DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 5. Benefits This chapter presents an analysis of the benefits associated with the proposed regulations under TSCA Section 402(c)(3). The proposed work practices, training and certification requirements will reduce lead exposure by increasing the containment and cleanup of dust and debris generated by renovation, repair, and painting (RRP) activities in child occupied facilities (COF). These reductions in exposure will in turn reduce the risks of adverse health and ecological effects in the vicinity of these activities. The chapter is organized around the analytical steps involved in estimating the benefits. These steps are outlined in Figure 5-1. Section 5.1 of this chapter presents an overview of these steps, including: (1) estimating the amount of lead contamination due to RRP events, (2) estimating the resultant changes in blood lead levels, (3) estimating the resultant adverse health effects, including reduction in IQ, and (4) estimating the dollar value of the reductions in adverse effects. Steps one through three are conducted for both the “without rule” situation (i.e. the baseline) and for the with rule situations. The difference between the loss of children’s IQ points avoided under the rule and the baseline are the quantified benefits of the rule, and are assigned dollar values in step four. Section 5.2 provides details on step one, the estimation of lead levels generated by RRP events and the effectiveness of cleaning. Section 5.3 presents the analytical details of steps two, three, and four as they relate to children. Section 5.4 briefly highlights effects and uncertainties associated with adult exposures. Section 5.5 describes the statistical model as well as the parameters used in the Monte Carlo analysis and their assumed values or distributions. Section 5.6 presents the numerical estimates of the value of the benefits for children. Five appendices supplement this chapter. They are Appendix 5A: Lead-Related Health Effects and Ecological Effects, Appendix 5B: Identifying and Characterizing Lead Loadings for Interior RRP Tasks in target housing COFs, Appendix 5C: Identifying and Characterizing Lead Loadings for Interior RRP Tasks in COFs in public or commercial buildings, Appendix 5D: Distributions of Inputs and Results, and Appendix 5E: Monte Carlo Analysis Inputs. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 1 5/17/2007 1 2 3 DRAFT—DELIBERATIVE Figure 5-1: Overview of Analytical Steps Amount of Lead Generated by RRP Activity (EPA Phase I Study – Appendix 3C) Estimate Amount of Lead Generated by Each Type of RRP Event (Section 2) Cleaning Efficiencies for Contractor and Routine Cleaning (Section 2) Estimate Level of Lead Exposure due to RRP Using Children’s BloodLead Models and Relationship between Blood-Lead and IQ, Estimate Changes in IQ (Section 3) Number of AtRisk Children Estimate Number of IQ Points Gained due to Regulation (Section 3) Value of IQ Point Gained (Life-time Earnings – Section 3) Estimate Value of IQ Points Gained (Section 6) Estimate Value of Quantified and Monetized Benefits (Section 6) §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 2 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 5.1 Overview of Approach Lead exposure causes many adverse health effects; a great deal of information on the health effects of lead is available from decades of medical observation and scientific research. Because inhaled or ingested lead is initially distributed throughout the body and is toxic to many organ systems, it damages a wide range of systems and its toxicity manifests itself in the form of many different types of adverse health effectsimpacts on many different organ systems as described below. A reduction in lead exposure resulting from the rule would lead to a reduction in these adverse health effects and the costs of treating them. Young children are particularly sensitive to lead, which impairs a child’s neuropsychological development (most commonly measured as reduced IQ). Increased blood-lead levels have also been associated with aberrant behavior in school-age children and a decrease in their growth rate and stature (Needleman 1996, Needleman 2002, Schwartz 1986, Shukla 1989, Shukla 1992). These cognitive and behavioral effects are strongly related to their future productivity and expected earnings (Salkever 1995; U.S. EPA 2000). Adverse health effects in adults may include hypertension, coronary heart disease (CHD), stroke, blood disorders, kidney damage, thyroid hormone abnormalities, immune system damage, many types of neurological abnormalities, increased incidence of stillbirths and miscarriage, low sperm rates, abnormal sperm, and infertility (ATSDR 1999). Both epidemiologic and toxicologic studies have shown that environmentally relevant levels of Pb affect many different organ systems (EPA 2006). It appears that some of these effects, particularly changes in the levels of certain blood enzymes and in aspects of children's neurobehavioral development, may occur at blood-lead levels so low as to be essentially without a threshold (IRIS EPA 2004). Appendix 5A presents an assessment of the animal toxicology and human epidemiology data available for a range of health effects associated with lead. While there are an assortment of adverse health effects associated with exposure to lead (see ATSDR 1999 and Appendix 5A), Both epidemiologic and toxicologic studies have shown that environmentally relevant levels of Pb affect many different organ systems (EPA 2006), but this analysis is able to include only a subset of children’s health effects, due to limitations in understanding and quantifying the dose response relationships for some of the health effects. Even where the dose-response relationships are known, many cases are not included in the estimates because exposure levels cannot be estimated for the relevant groups of potentially affected individuals. Consequently, this benefits assessment focuses on effects on cognitive function in young children (under the age of six). It should be noted that even for this effect (IQ loss), some potential cases of these health effects are not included in the benefits estimates. This is because exposures of some potentially affected individuals (for example, neighbors of households performing renovations) have not been estimated in this assessment. To estimate the benefits of the proposed rulemaking, the quantified adverse health effects associated with exposures to lead from RRP tasks in the baseline (i.e., without RRP regulation) are first calculated; then, health effects associated with exposures are calculated assuming the RRP regulations are in place. Since the rule requires actions intended to reduce contamination, fewer adverse health effects are expected with the rule. This reduction in adverse health effects is the rule’s major benefit. The most commonly used measure of the amount of lead in the body is blood-lead level (PbB), although lead also bioaccumulates in bone, hair, teeth, and other tissues. Published studies relate one or more of these measures such as blood or bone lead levels to adverse health effects. Blood lead is generally a biomarker of recent lead exposure. However, it is also affected by chronic exposure (i.e., lead released from bone from previous exposure may result in elevated blood lead levels). Published studies relate one §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 3 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 or more of these measures such as blood or bone lead levels to adverse health effects. Some studies have examined the question of whether the neurological effects of exposures in early childhood are ameliorated when blood-lead levels decline. The data are mixed on this issue. In a study that treated lead-exposed children with a chelating agent, Ruff (1993) found that children whose blood-lead levels had the greatest decline showed the most improvement in IQ. In contrast, Rogan (2001) found that treatment with a chelating agent lowered blood-lead levels in children but did not appear to improve neurological function. Liu (2002) also found that chelation therapy, while lowering blood-lead levels, did not improve neurological function in children at 5 years of age. While the study did detect a relationship between declining blood-lead and improved neurological function, this association was observed only in the untreated group, leading the authors to speculate that some other factor besides declining lead levels from chelation therapy (such as greater parental involvement), led to the neurological gains. Dietrich (2004) had similar findings in the same cohort of children at 7 years of age. One study cited in ATSDR (1999) showed impaired motor and cognitive function at a current mean level of 2.9 µg/dL, about 20 years after exposure when blood-lead levels were 40-50 µg/dL (Stokes 1998). These studies suggest that medical interventions aimed at lowering blood-lead levels may not lead to dramatic improvements in neurological function. This, further supportsing the concern that early exposures to lead (Pb) may lead to irreversible damage, and supportsing the benefits of regulatory interventions to prevent or reduce lead exposure. The estimation of the adverse health effects associated with renovation, repair, and painting projects involves four steps: 1. Estimate the amount of lead contamination due to the renovation project under various assumptions about cleaning; 2. Estimate the blood-lead levels resulting from this contamination; 3. Estimate the adverse health effects (i.e., loss in IQ points) and; 4. Assign medical costs, reduced income, or another proxy for willingness-to-pay to avoid the adverse health effects. Each of these steps is briefly discussed below; methods for implementing these steps are described in detail in Sections 5.2 and 5.3 Step 1: Estimate the amount of lead contamination due to the renovation project Most of the events covered by this rule are expected to generate lead dust levels that exceed the EPA floor hazard standard of 40 µg/ft2 promulgated in 2001.1 As described in detail in Appendix 5B and Appendix 5C, all RRP events undertaken where lead-based paint is present have the potential to exceed the 40 µg/ft2 hazard standard if precautions are not taken. In some cases contractors are already taking precautions even without the rule, and the number of RRP events where this is the case is estimated in Section 4.5 of this document. (Also see Section 5.2.4 of this chapter.) In addition, not all COFs contain lead-based paint; the likelihood that lead-based paint occurs in a COF is estimated in Section 4.2.3 of this document, and summarized in Table 4-7 for commercial or public buildings. In the majority of RRP events, however, contractors do not take these precautions.However, based on a survey of nine contractors and 1 The final lead hazard standards were established under the authority of Section 403 of the Toxic Substances Control Act. These standards were based in part on a benefit-cost analysis comparing clean-up costs to the value of IQ points lost in children without the cleanup. The objective of the Section 403 benefit-cost analysis was to identify a standard at which net benefits are maximized, assuming that all households with children under the age of seven undertake appropriate lead clean-up activities if their home exceeds the hazard standard. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 4 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 industry representatives described in section 5.2.4, contractors do not take these precautions in the majority of RRP events. RRP events with precautions already in place are not included in the benefits calculations; only those RRP events performed without appropriate precautions are included in the calculations. Unless the appropriate safe work practices, containment, and cleanup are used, lead dust from the renovation activities can migrate throughout the COF and children could be exposed to lead levels that are high enough to cause adverse health effects, even when accounting for normal cleaning. Step 2: Estimate blood-lead levels from this contamination Several studies establish a strong correlation between lead dust levels and blood-lead concentrations in children. Dust caught in carpets that is later released due to vacuuming, as well as dust in window wells and on uncarpeted floors, can contribute to increased blood-lead levels among children. In a study of lead exposure among urban children, Lanphear (1998a) showed a dramatically rising probability of blood-lead levels exceeding 10 µg/dL,2 with a probability of over 50 percent for lead dust greater than 100 µg/ft2. Another analysis that measured lead dust on multiple surfaces (carpeted and non-carpeted floors as well as window sills and window wells) found that lead dust was significantly correlated with the probability of increased blood-lead levels for every surface tested (Lanphear 1996). When compared against several other factors, window trough lead dust highly correlated with blood-lead levels in the Lead-Safe Cambridge program (Potula 2001). The studies discussed above evaluated the correlation between lead dust and blood-lead levels without specifying the activities that generated the leaded dust. Further, the studies either do not specify the duration of the exposure to the lead dust, since the dust and blood-lead measurements are concurrent, or in one case, specifically examined chronic lead dust exposure and excluded cases where there had been spikes of exposure such as renovation (Lanphear 1996). However, two other recent studies focused specifically on the impact of home renovation on lead exposure among children. In New York City, authors conducted a case-control study comparing children living in homes that had undergone renovation and/or repair (cases) to those children living in homes without these activities (controls). One criterion for including children in the study was that they had no prior history of increased blood-lead measurements; thus it is very unlikely that the renovations were carried out as abatements to reduce lead exposure to children living in the house (which could have biased the study). The study found that children with increased blood-lead levels were more likely to live in a house that had undergone renovations (1.2 times more likely), had interior surfaces prepared for painting (3.5 times more likely), or had work-related dust dispersed throughout the house (6.3 times more likely), as compared to control children (Reissman 2002). In a Department of Health study also from New York (but which excluded New York City), half of the families with a child with increased blood-lead levels reported more than one type of paint removal activity in their house (CDC 1997). Thus, the studies concluded that increased blood-lead levels are correlated with renovation activities where lead-based paint is present. Decreases in lead dust have been shown to decrease blood-lead levels among children. Several studies have investigated the effectiveness of lead dust controls in conjunction with or in the absence of other lead-control activities (NCHH 2002, Hilts 1998). One intervention study, which trained families in home dust-control measures, found a reduction of up to 4.0 µg/dL after one year of the program (Hilts 1998). 2 This is the CDC level of concern, which is intended to trigger community-wide prevention activities. It is not a threshold of effect. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 5 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 Thus, control of dust is an important element of any renovation and clean-up event where lead-based paint is present. Step 3: Estimate the adverse health effects (i.e., loss in IQ points) due to increased blood-lead levels As described above, this assessment estimates the adverse health impact of increased blood-lead levels on cognitive function and, more specifically, IQ values in young children. Appendix 5A provides a review of the recent literature related to the cognitive effects of lead in children. Young children are particularly sensitive to lead, which impairs a child’s neuropsychological development (most commonly measured as reduced IQ). Increased blood-lead levels have also been associated with aberrant behavior in school-age children and a decrease in their growth rate and stature. These cognitive and behavioral effects are strongly related to their future productivity and expected earnings (Salkever 1995; U.S. EPA 2000). EPA believes there is essentially no threshold for adverse health effects of lead in children. Indeed dose-effect curves for lead effects on children’s IQs show a non-linear, inverse relationship with the greatest effects occurring at the lowest detectable blood-lead levels. In an effort to determine what a blood-lead level of concern should be, the Workgroup of the Advisory Committee on Childhood Lead Poisoning Prevention to the Centers for Disease Control and Prevention (CDC 2005a) found that the overall weight of available evidence supports an inverse association between blood-lead levels and the cognitive function of children in the low range of exposure (less than 10 µg/dL blood). The evidence for such an association is bolstered by the consistency across both cross sectional and longitudinal studies in varied settings. Further, the association is not weaker in studies where the populations’ mean blood-lead levels are relatively lower (CDC 2005a). Thus, this analysis assumes that there is no evidence of a threshold below which the adverse health effects of lead are not experienced. Similarly, U.S. EPA’s Integrated Risk Information System (IRIS 2004) concluded: “by comparison to most other environmental toxicants, the degree of uncertainty about the health effects of lead is quite low. It appears that some of these effects, particularly changes in the levels of certain blood enzymes and in aspects of children's neurobehavioral development, may occur at blood-lead levels so low as to be essentially without a threshold.” Step 4: Assign medical costs, reduced income or other proxy for willingness to pay to avoid the adverse health effects In the following analysis, standard values from the economic literature are used for the benefits valuation. In lieu of willingness to pay, medical costs and reduced income are used as an estimate of the cost of chronic conditions. 5.2 Estimating Lead Dust Contamination Levels from RRP Activities The benefit estimates provided in this report reflects the following sequence of events, which vary somewhat between Target Housing (TH) COF’s and COFs in public or commercial buildings (i.e., centers, kindergartens, and pre-kindergartens). In the baseline, an RRP event occurs, generating lead dust in both the area where the work occurs and the adjacent area (e.g., the adjacent room). In the baseline (without rule), the contractor cleans the work area before he/she leaves by sweeping or vacuuming. While this contractor cleaning substantially reduces the lead dust in the work area, it rarely brings the lead levels down to the EPA hazard levels of 40 µg/sq.ft (Dixon 1999, Ettinger 2002). In most cases, this cleaning also does not address the lead dust that may have migrated into the rest of the COF during the §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 6 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 RRP event. Following the end of the RRP event, the COF cleans the work area and the rest of the COF on a regular basis. Periodic cleaning continues to reduce the lead dust levels over time. Under the rule, the same sequence of events occurs, but it is assumed that the containment, cleaning, and cleaning verification practices required by the rule reduce lead dust loadings to 40 µg/ft2 or less after contractor cleaning. For the benefits analysis, exposures to both indoor dust and to soil contaminated from exterior painting are evaluated. 5.2.1 Initial Dust Loadings from Interior RRP Activities For COFs in TH, EPA estimated the lead dust loadings associated with an interior RRP event using the U.S. Census Bureau’s 1997 and 2003 American Housing Survey (AHS) and EPA’s 1997 “Lead Exposure Associated with Renovation and Remodeling Activities: Environmental Field Sampling Study” (EFSS). The AHS provides information on the number of households in which renovation and repair tasks of various types are carried out during the prior two years.3 EPA used this information to estimate the number of renovation and remodeling activities that could potentially generate lead dust4. The AHS does not detail the amount of lead dust a particular RRP event could generate. EPA used data provided in the EFSS to estimate the lead dust loadings associated with each RRP task identified in the 1997 AHS. For a more detailed description of the approach and methods used to estimate the lead dust loadings for interior RRP tasks, see Appendix 5B. For COFs in public or commercial buildings EPA used data from HUD's First National Health Survey of Child Care Centers (HUD 2003). The survey data was collected in 2001 and was published in 2003; it includes data on 98 daycare centers that are known to be built before 1978. This methodology and data were also used to extrapolate to kindergartens. The basic steps for estimating the number of events are: 1. Estimate the number of COFs (rooms and centers), 2. Estimate the frequency of performing an event, 3. Estimate the likelihood that an event will be affected by the rule (disturbing paint, disturbing LBP). 4. Combine the results of the above three steps to estimate: (1) annual number of centers and classrooms where interior painting with sanding or scraping takes place, (2) annual number of centers and classrooms where interior painting with sanding and scraping lead-based paint (LBP) is performed, (3) number and distribution of children and dust loadings in centers where interior painting with sanding and scraping LBP is performed. In addition, data from Whitestone Research was used to estimate the types and frequency of RRP work for COFs in public or commercial buildings (including both elementary schools and childcare centers). For a more detailed description of the approach used to estimate the lead dust loadings for interior RRP tasks centers see Appendix 5C. 5.2.2 Effect of Cleaning on Interior Dust Loadings In the baseline an RRP event occurs, generating lead dust in both the area where the work occurs and the adjacent area(s). In the baseline (without rule), initial cleaning occurs in the work area by the contractor or the in-house staff performing the RRP work. While this initial cleaning substantially reduces the lead dust in the work area, it rarely brings the lead levels down to the EPA hazard levels of 40 µg/sq.ft.5 In 3 Details are provided in Chapter 4 of this report. The distribution of dust loadings estimated for the 2006 proposed LRRP TH rule was used as the distribution of dust loadings generated during events in COFs in target housing. 5 EPA and HUD have developed protocols for clean-up following abatement and interim control activities. 4 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 7 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 most cases, this cleaning also does not address the lead dust that may have migrated into the rest of the COF during the RRP event. Following the end of the RRP event, the COF cleans (routine cleaning) the work area and the rest of the COF on a regular basis. Periodic cleaning continues to reduce the lead dust levels over time. Under the rule, the same sequence of events occurs, but it is assumed that the containment, cleaning, and cleaning verification practices required by the rule reduce lead dust loadings to 40 µg/ft2 or less after initial cleaning. The blood-lead models used in the analysis require an estimate of the annual lead exposure to occupants of COFs where RRP occurs. To obtain this value the following calculations were performed6: 1. Lead dust levels in the work area due to the RRP event are estimated as described in Section 5.2 and Appendix 5B and Appendix 5C. 2. For TH-COFs lead dust levels due to the RRP event in the room(s) adjacent to the work area are estimated to be 16 percent of the lead levels in the work area (i.e., before initial cleaning). See Appendix 5B for the derivation of this factor. For COFs in public or commercial buildings EPA assumes that all COF classrooms are painted at the same time, so that the work area is the size of all COF rooms that contain LBP and the adjacent room is either the same size as the work area or the remaining space in the center not containing LBP, whichever is smaller. 3. Lead dust levels in the rest of the TH-COF and COF in public or commercial buildings are assumed to be zero.7 4. Average work area lead levels when the RRP is completed are estimated by applying normal (initial) cleaning efficiencies to the lead levels estimated in step 1. See Section 5.2.3 and 5.2.4, Appendix 5B, and Appendix 5C. 5. For TH-COFs, average lead dust levels when the RRP is complete are estimated as the weighted average of the lead dust levels due to the RRP event in the work-room after initial cleaning, the adjacent room, and the rest of the TH-COF. Each of these three values is weighted by the relative square feet of the area involved. The average work area size is estimated for each type of RRP event in Chapter 4 and then expressed as a percentage of the total size of the housing unit.8 The size of any household’s work area is the sum of percentages shown in Table 5-1 for the events that are performed. The maximum work area size for a COF is calculated as the average size of the largest event (a Non-Room-Specific event), which is 30 percent of the unit.9 The adjacent area is assumed to equal the work area in size. Applying these protocols will reduce lead levels below the hazard standard. However, the cleaning currently performed following renovation projects is typically less extensive. 6 The following example is specific to interior painting events. Interior non-painting (i.e., wall disturbing) events have the same overall approach but there are differences in the assumptions; see section 4.2.3 for details. 7 It is likely that in some circumstances the lead dust from RRP events can contaminate other areas of the COF beyond the work room and the adjacent room. Because no data were located specifically addressing lead dust levels in the rest of the house from RRP events, the level was assumed to be zero. However, this assumption is likely to understate the extent of lead dust contamination from RRP events, and to underestimate the benefits of the rule. 8 For large events, the work area size is estimated as the size of the room where the RRP work is performed (or 50 percent and 25 percent of the housing unit for non-room specific and interior painting events, respectively); for small events the work area is estimated as an area the size of one wall by five feet (to reflect the spreading of plastic out 5 feet from the work). To maintain consistency with the assumptions regarding the mix of large and small events estimated for the cost analysis, the typical work area size is estimated as the midpoint between these two estimates. 9 Note that 30 percent of the unit is the average work area of the largest event, a non-room-specific event. While some non-room-specific events are larger, and others are smaller; the average size is used in the benefits calculations. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 8 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 6. For COFs in public or commercial buildings, average lead dust levels when the contractor leaves are estimated in a similar manner but the work area is the size of all COF rooms that contain LBP and the adjacent room is either the same size as the work area or the remaining space in the center, whichever is smaller. In addition, instead of assuming one average dust loading for interior painting (as in TH-COFs), the entire distribution of loadings presented in Table 3 of Appendix 5C is used. 7. Periodic COF routine cleaning is applied to the lead dust levels and the decline is tracked over time so that the annual average concentration can be calculated. 8. The benefits are estimated based on exposure that starts when the RRP and initial cleaning are complete. Thus the exposure estimates used in the benefits estimation do not include any exposure to COF occupants or neighbors (or the RRP workers) that might occur during the RRP event itself. Table 5-1: Work Area Size by Event Type (Percent of Housing Unit) Work Area Event Type Size Kitchen 6% Bathroom 3% Addition 5% Non-Room-Specific 30% Interior Painting 16% Average Household Work Area 24% Source: Calculated from the American Housing Survey, see Chapter 4 for details. 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 5.2.3 Estimating Cleaning Efficiencies Initial cleaning (i.e., by the contractor or in-house staff performing RRP work) immediately after the RRP work is completed decreases the amount of residual lead dust on COF surfaces. Following the initial cleaning, the COF is expected to undertake routine cleaning, further decreasing lead levels generated by the RRP event. This section describes the development of cleaning effectiveness estimates for various specific cleaning methods. Section 5.5 presents the modeling assumptions about the types of cleaning and the frequency of cleaning undertaken by contractors and COFs in the baseline and under the rule. The amount of lead present at any particular time is a function of the initial lead levels, the effectiveness of the cleaning and the number of times that cleaning has occurred in the intervening period. This relationship can be represented as: Pb Concentration at time t = [Pb concentration at time 0] * (1-Efficiency)Frequency where: Efficiency = the percent of lead dust removal associated with each cleaning activity; and §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 9 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 Frequency = the number of times the cleaning technique is employed between time 0 and time t. As described in more detail below, cleaning efficiency rates vary with the type of surfaced cleaned, the type of cleaning done and the amount of dust present. While cleaning efficiencies typically decline as the amount of dust declines (Yiin 2002, Rich 20024), this analysis assumes a constant rate for any particular situation, and thus overestimates the effectiveness of multiple cleanings. This in turn results in an underestimate of the benefits of the regulation. The model further assumes that routine cleaning by the COF will continue to reduce floor lead levels until the floor lead dust loading reaches a minimum of 1.1 µg/ft2. At this point, lead levels will no longer decline. The 1.1 µg/ft2 level is the U.S. household geometric mean for floor lead dust (Jacobs 2002).10 The length of time to reduce floor lead dust loadings from a particular RRP event is a function of the amount of lead generated by the RRP event, the type of cleaning method used, the frequency of cleaning, and whether the work practices specified in the proposed rule are used. EPA assumes that after initial cleaning of the workspace, the COF will perform subsequent routine cleanings using a typical vacuum or broom. To estimate the efficiency of clean-up activities performed by the COF (i.e. non-HEPA portable or central vacuum and/or broom sweeping), EPA conducted a literature review of studies that performed clean-up activities after RRP events. Of the studies reviewed by EPA, the majority described clean-up activities consisting of vacuuming, mopping, sweeping and/or using a wet clean-up method (i.e., phosphate detergent or water) on either a carpeted or non-carpeted surface. Of the studies selected, EPA only considered studies that described the effectiveness of clean-up activities associated with RRP events. The studies identified by EPA used a variety of clean-up methods and reported a wide range of cleaning efficiencies. To determine the most representative cleaning efficiency value, EPA analyzed each study and selected the studies with the most robust study design to establish a COF clean-up efficiency for both carpeted and non-carpeted surfaces. None of the studies, however, looked at cleanup of dust from the full range of activities that could occur during RRP. Rich 2002, Yiin 2004, Dixon 1999, and Ettinger 2002 each use a clean-up method consisting of a HEPA vacuum alone or a HEPA or non-HEPA vacuum in conjunction with either a tri-sodium phosphate (TSP) or non-TSP detergent. The HEPA studies were excluded from the analysis because the majority of cleanup activities do not employ a HEPA vacuum. Studies reporting efficiencies from the use of non-HEPA vacuuming combined with wet cleaning were eliminated because wet cleaning was not included as an initial cleaning method in the analysis. 11 This analysis assumes that COFs clean non-carpeted floors using broom sweeping or a non-HEPA vacuum. Some COFs may mop non-carpeted floors instead of sweeping or vacuuming. EPA believes that most such COFs are likely to use a single bucket mopping method instead of the two bucket method that would be required by the proposed rule. (The single bucket method uses the same water throughout, while the two bucket method uses separate wash and rinse water.) While EPA does not have information on the cleaning efficiency of the single bucket method nor data supporting the assumption about the 10 This value is used to represent the lowest lead level that cleaning can feasibly achieve over time. It is used in the analysis to indicate the background lead level before an RRP event was commenced, hence the lowest level one could expect to achieve after an event if all lead from the RRP event was eventually removed by cleaning. 11 While contractors can use wet cleaning methods under the rule, such methods are not appropriate for carpeted and upholstered surface. To simplify the benefits estimation, dry methods were assumed to always be used. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 10 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 utilization of single bucket mopping, the Agency does not believe that single bucket mopping is more effective than the 94.8% to 98.5% reduction in lead dust loadings used in the analysis for non-carpeted surfaces. However, if both cleaning methods are used, the cleaning effectiveness may be enhanced. A sensitivity analysis using alternate assumptions about cleaning efficiencies after RRP events in schools and daycare centers is shown in Chapter 7. Yiin 2002, Clemson Environmental Technologies Laboratory (CETL) 2001, Figley and Makohon (Canadian Mortgage and Housing Corporation (CMCH)) 1992, and EFSS 1997 describe clean-up efficiencies on both carpeted and/or non-carpeted surfaces. The Yiin, CETL, and CMCH studies used similar non-HEPA vacuums with an agitator head while the CMCH also used a non-HEPA vacuum with a plain tool head. It would be expected that each type of vacuum would remove a similar amount of lead dust from a carpeted surface, but Yiin 2002 reported that non-HEPA vacuums remove 14.0-36.6 percent of lead dust deposited after a RRP event, the CMCH 1992 study reported a non-HEPA clean-up efficiency of 23.7-65.3 percent, and the CETL 2001 study reported a non-HEPA clean-up efficiency of 66-84 percent. The large discrepancy between the clean-up efficiencies reported in each of these studies could be attributed to three factors: (1) the lead dust level applied to the carpet prior to clean-up, (2) whether the study accounted for the size of the dust particles, and (3) whether and how the study accounted for the redistribution/deposition of lead dust after clean-up. Yiin 2002 had both a lead dust level similar to those found in RRP events and accounted for redistribution/deposition of lead dust after cleanup. The study measured the efficiency of removing lead dust from carpet using a non-HEPA vacuum in two different experiments. The first experiment observed the clean-up efficiency of non-HEPA vacuums on carpets with low levels of lead dust (geometric mean: 54.4 µg/ft2), while the second experiment measured the clean-up efficiency of non-HEPA vacuums on “soiled” or high levels of lead dust (geometric mean: 113 µg/ft2). After vacuuming the carpeted surface Yiin et al. waited at least 1 hour to allow for the redistribution/deposition of lead dust prior to measuring the amount of lead dust remaining in the carpet. This practice was intended to account for the remaining lead dust levels because under normal vacuuming conditions airborne lead dust will be generated as a result of the agitator action of the vacuum head. Using the clean-up efficiencies (as measured by vacuum) from both the low dust level and high dust level experiments, EPA calculated a cleaning efficiency range for Yiin 2002 of 14.0-36.6 percent for non-HEPA vacuums on carpets. This is applied to both contractor and householder non-HEPA cleaning of carpets. The CETL 2001 study measured the clean-up efficiency on carpeted and non-carpeted surfaces of a variety of non-HEPA vacuums by applying a defined amount of lead dust to either a carpeted or noncarpeted surface. Unlike the Yiin 2002 study in which the lead dust level applied to the carpeted surface is more indicative of what would be found in a “real world” situation, CETL applied extremely high levels of lead dust, 1,000,000 µg/ft2, to both carpeted and non-carpeted surfaces. As a result, the clean-up efficiencies detailed within the study may be an overestimation because it is easier to reduce the percentage of lead dust found on a surface if there is more lead dust initially. Further, the CETL study does not take into account the redistribution/deposition of lead dust after clean-up. The CMCH 1992 study measured the clean-up efficiency of a variety of non-HEPA vacuums on carpeted and non-carpeted surfaces. In addition to using an extremely high level of dust to determine these cleanup efficiencies, 3,700,000 µg/ft2, the CMCH study did not use lead dust generated from RRP events. Instead the study used “a representative dust sample… produced by gathering floor debris including wallboard/plaster scraps, dust and paint chips from an existing building renovation.” To produce a dust §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 11 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 sample containing lead, lead stearate was added to the sample, but it represents only a “very small mass component of the test dust… [and artificially skewed] the lead content analysis… towards the fine particle sizes.” EPA excluded the non-HEPA vacuum clean-up efficiencies detailed within the CMCH study for carpeted and non-carpeted surfaces because the study does not use lead dust generated from a RRP event and, therefore, it is unclear if the efficiencies detailed are consistent with what would be found if the experiment used lead dust generated from a RRP event. EPA did include the broom clean-up efficiencies in this analysis although the study did not use lead dust generated from a RRP event because the broom clean-up efficiencies are consistent with those reported in the EFSS. The EFSS 1997 study measured the clean-up efficiency of brooms on non-carpeted surfaces, i.e. linoleum. Unlike the other studies considered by EPA, the EFSS study performed the clean-up operations on lead dust generated from drilling and sanding activities on wood door surfaces covered with leadbased paint. After the generation of lead dust, the lead loadings in the workroom were measured at 0 ft and 6 ft from the activity.12 The study found that at distances from 0-1 ft from the activity the broom had clean-up efficiencies similar to non-HEPA vacuums for lead dust generated from work activities, 98.8 percent for drilling and 99.5 percent for sanding. The study did wait one hour after each clean-up activity to allow for redistribution/deposition of lead dust prior to determining the clean-up efficiency. The broom clean-up efficiencies obtained in the EFSS are similar to those found in CMCH 1992 for the lead dust levels generated at 0-1 ft, which supports the clean-up efficiencies in Table 5-2Table 5-2. The main discrepancy with broom clean-up efficiencies reported in EFSS compared to those reported in CMCH exists at 5-6 ft from the work activity. EFSS found at 5-6 ft from the work activity broom clean-up efficiencies decreased to 25.1 percent - 38.3 percent. In using the higher clean-up efficiencies, the analysis produces lower benefit estimates than would be the case if the lower efficiency numbers had been used The values used in the benefits estimate are shown in Table 5-2Table 5-2. Table 5-2: Efficacy of Clean-up Activities for Initial and COF Cleaning Surface Reduction in Lead Dust Loadings Citation Portable and Central Vacuum. No HEPA, no washing Carpet 14% - 36.6% Yiin 2002 Portable and Central Vacuum or Sweeping. No HEPA, no washing Non-carpeted surface 94.8% - 98.5% Clemson Environmental Technologies Laboratory (CETL) 2001; Figley and Makohon (CMCH) 1992. Clean-up Option 28 29 30 31 32 33 Because cleaning efficiencies vary depending on whether floors are carpeted or not, it was also necessary to estimate the percentage of floors covered by carpet in TH-COFs and COFs in public or commercial buildings. The percent of single family and multi-family homes with any carpet is presented in HUD (2000) and was used to estimate the percentage of carpeting in TH-COFs. For homes with carpet, these data also provide the percentage of area of flooring (in square feet) that is covered in carpet. Using these 12 At 0 ft from the activity the geometric mean lead loading was 26,700 µg/ft2 for drilling and 653,000 µg/ft2 for sanding, while the geometric mean lead loadings at 6 ft from the activity were 65 µg/ft2 for drilling and 1,380 µg/ft2 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 12 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 two values, the analysis calculated the total percentage of the area of flooring covered in carpet among all TH-COFs covered by the proposed regulation. This value ranges from 33 to 38 percent. Using these values, a weighted average percent of area of flooring that is carpeted across all types of homes covered by the regulation was created. This weighted average is 36 percent. EPA used data from HUD's First National Health Survey of Child Care Centers (2003) to estimate an average 33.8% carpeting in COFs in public or commercial buildings. In order to model the effect of cleaning on the levels of residual dust over time, it was necessary to estimate a frequency of cleaning in TH-COFs and COFs in public or commercial buildings. Simcox (1995) conducted a study of pesticide exposures from household dust. As part of this study, the authors surveyed families about the frequency of cleaning activities. The study reported 40 percent of survey respondents vacuumed more than once per week, 45 percent vacuumed once per week and 16 percent vacuumed less often than once per week. For this analysis, 40 percent of TH-COFs were assumed to vacuum twice per week; 45 percent were assumed to vacuum once per week and 15 percent were assumed to vacuum once every two weeks.13 By comparison, 100 percent of COFs in public or commercial buildings were assumed to clean daily, based on data reported in HUD's First National Health Survey of Child Care Centers (2003). 5.2.4 Adjustment for Baseline Lead Safe Work Practices (i.e., Scenarios 1 and 2) Even without the proposed regulation, contractors (or in-house staff performing RRP) already perform some containment and clean-up. In order to determine how often various work practices are used in the baseline, nine industry experts were questioned. (The data collection effort is described in Section 4.5.4). The questions and responses on work site preparation (containment) and clean-up are summarized in Table 5-3Table 5-3Table 5-3. As shown in Table 5-3Table 5-3Table 5-3, the baseline frequency varies widely across different work practices. For example, respondents indicated that floors, outdoor areas, and doorways were covered with impermeable material (e.g., polyethylene plastic) in approximately 40 percent of events, work areas were vacuumed daily with a HEPA-filtered vacuum on a daily basis in less than 20 percent of events, and that exterior debris was wet misted and the impermeable material was rolled and sealed with duct tape in less than 5 percent of events. While the survey did not ask about cleaning verification, there are probably few renovation contractors performing the cleaning verification required by the rule. While the survey did not ask about cleaning verification, the specific cleaning verification method in the proposal did not exist before it was developed for this rule, so it is unlikely that anyone is following it in the baseline. Care must be taken in interpreting these data. First, there were only 9 respondents in the survey. Second, the questions only asked how frequently each of the various activities or practices was used (e.g. rarely, often, usually), not how often they are used when needed. Thus it is not clear whether the relatively low value for some activities reflects situations where the activity is not needed, or situations where they should be but were not undertaken. Third, some of the activities might be considered substitutes (e.g. turning off HVAC system versus sealing HVAC vents). Fourth, the survey asked about individual for sanding. 13 Note that the analysis modeled each cleaning as having the same level of efficiency. In fact, several studies have shown that cleaning efficiency is dependent on the lead concentration. Initial cleanings can be more efficient than later cleanings after much of the lead dust has been removed. Further, at least one study (Ewers 1994) showed that vacuuming could actually increase the surface lead dust level on carpets by bringing lead trapped in carpets to the surface. To the extent that this analysis overestimates baseline cleaning efficiencies, the benefits are underestimated. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 13 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 practices, but a combination of practices is needed to provide adequate protection. Finally, the survey did not address whether the activities were being performed properly to protect occupants. For example, contractors using containment may not do so effectively unless they have been trained in how to properly lay, affix, and dispose of plastic sheeting.14 Improper containment can release lead dust and contaminate the housing unit. Table 5-3Table 5-3Table 5-3 presents the actual survey questions while Table 4-23 relates to the work practices required by the rule. Because the survey was completed before the current proposed rule was developed, some of the survey questions do not match the rule’s requirements. In these cases the survey results in Table 5-3Table 5-3Table 5-3 were used to develop an assumption about the percentage of events where some form of a work practice required by the rule is already in use. For example, the survey asked how often a HEPA vacuum was used on a daily basis (shown in Table 5-3Table 5-3Table 5-3 to be estimated at 18.8 percent of events), but the rule only requires the use of a HEPA vacuum when the renovation is completed. It was assumed that some events that are not cleaned with a HEPA vacuum on a daily basis are cleaned with a HEPA vacuum when the renovation is completed. It was further assumed that a HEPA vacuum was used at the end of the renovation with the same frequency that roll down polyethylene sheeting was used, or in approximately 40 percent of renovation jobs (as shown in Table 423). 14 A 2004 evaluation of the EPA/HUD Lead Safety for Remodeling, Repair, and Painting curriculum found that trainees were more likely than not to use plastic sheeting properly on subsequent jobs. (HUD 2004) §402(c) COF Economic Analysis Chapter 5 14 Draft - Do Not Cite or Quote 5/17/2007 DRAFT—DELIBERATIVE Table 5-3: Summary of Current Practice Survey Results Activity Percentage of time the activity is already taking place Preparation Restrict access to the work area by placing impermeable material (e.g. polyethylene plastic) over doorways 39.3% Turn off the HVAC systems and close vents 16.9% Seal off HVAC vents with impermeable material 29.6% Cover floor of work area with impermeable material Create a runner of impermeable material between work area and outside door 41.2% 29.1% Cover occupants belongings with impermeable material Cover surrounding ground/soil with impermeable material during outdoor renovation projects 61.9% Place warning signs outside unit in a multi-unit building Restrict access to outdoor work area by constructing barriers and/or placing warning signs 13.2% 37.6% 23.0% Clean-Up Sweep work area daily 74.6% Mop work area daily 21.1% Vacuum work area with a shop-vac daily Vacuum work area with a two-stage high efficiency filtered vacuum daily 58.1% 27.2% Vacuum work area with a HEPA-filtered vacuum daily 18.8% Wrap debris in impermeable material and duct tape shut prior to removal Place removed carpet in impermeable material and duct tape shut prior to removal Wet mist debris generated by outdoor projects; roll and seal with duct tape before disposal 23.0% 4.8% 4.8% Results are based on a 1999 survey of nine contractors and industry representatives. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Because of uncertainties about the type of baseline containment and cleaning practices used, how often they are performed, whether they are carried out properly, and how these practices reduce lead levels, benefits are calculated under two different scenarios. The two scenarios both start with the assumption (based on the results in Table 5-3) that the more common work practices required by the rule (such as cleaning with a HEPA vacuum at the completion of the renovation) are used in 30 to 40 percent of baseline events. The first scenario assumes that the equivalent of 20 percent of all at-risk occupants are protected in the baseline commensurate with the results of the rule. This could be because 40 percent of all at-risk occupants receive 50 percent of the full protection benefits provided by the rule, or 31 percent of all atrisk occupants receive 65 percent of the full protection benefits provided by the rule, 25 percent of all atrisk occupants receive 80 percent of the full protection benefits provided by the rule, 20 percent of all atrisk occupants receive the full protection benefits provided by the rule, or some other computational equivalent. This scenario is meant to cover all baseline activities. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 15 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 The second scenario also assumes that 20 percent of all at-risk occupants are protected in the baseline commensurate with the results of the rule. In this scenario, the 20 percent represents occupants receiving the full protection benefits of the rule, and there are additional occupants who may receive some partial protection that may not be commensurate with the rule. Given that work practices such as cleaning with a HEPA vacuum at the completion of the job are assumed to be used in 30 to 40 percent of the events, this scenario assumes that additional cleaning activities are performed in another 10 to 20 percent of events. This additional cleaning goes beyond the efficiency of normal initial or COF cleaning, but does not necessarily reach the level of protection achieved by the rule. There are three such activities assumed in Scenario 2: • Additional cleaning by contractor (or in-house staff performing RRP). In 15 percent of the baseline events, contractors (or in-house staff performing RRP) are assumed to achieve dust reduction levels beyond those credited to initial cleaning. This could be due to the use of a HEPA vacuum, or through a second cleaning with normal cleaning methods immediately after the RRP event. Computationally, the cleaning efficiency is calculated as two successive cleanings. The first cleaning yields normal initial cleaning efficiencies (a 14 to 37 percent removal rate on carpets and a 94.8 to 98.5 percent removal rate on non-carpeted surfaces). The second cleaning for carpets is credited with the same cleaning efficiency (14 to 37 percent), while non-carpeted surfaces are credited with a 70 percent reduction. Thus, the 15 percent of events in this category achieve a 26 to 60 percent cleaning efficiency on carpets and a 98.4 to 99.6 percent cleaning efficiency on non-carpeted surfaces. These events are in addition to the baseline activities that protect 20 percent of the population described above. If there are any events where the lead loadings are sufficiently low so that the cleaning efficiency for this additional cleaning reduces lead loadings to below the 40 µg/ft2 level in the baseline, then the same level is used in the postrule analysis. That is, lead levels are generally assumed to be 40 µg/ft2 post-rule, but are not assumed to increase from the baseline as a result of the rule. Contractors (or in-house staff performing RRP) are assumed to cease this additional cleaning after the rule. • Contractor (or in-house staff performing RRP) cleaning of adjacent room. In 10 percent of baseline events, contractors (or in-house staff performing RRP) are assumed to clean the adjacent room as well as the work room. These events are independent of the additional cleaning described above. To the extent that an event is in both the 15 percent of events that achieve dust reduction above normal levels and the 10 percent of events where the adjacent room is cleaned, the cleaning efficiency is assumed to be the same in the adjacent room as in the work room. If there are any events where the lead loadings are low enough that the cleaning efficiency for this contractor cleaning of the adjacent room is calculated to reduce lead levels to below 40 µg/ft2 in the baseline, the same level is used in the post-rule analysis. That is, lead levels are generally assumed to be 40 µg/ft2 post-rule, but are not assumed to increase from the baseline as a result of the rule. Contractors (or in-house staff performing RRP) are assumed to cease this cleaning of the adjacent room after the rule. • Additional cleaning by COF. In 15 percent of the events, COFs are assumed to achieve dust reduction levels beyond those credited to routine COF cleaning in their first cleaning event. This could be due to the use of a HEPA vacuum, or through a second cleaning with normal cleaning methods immediately after the RRP event. Computationally, the cleaning efficiency is calculated as two successive cleanings. The first cleaning yields typical efficiencies (a 14 to 37 percent §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 16 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 removal rate on carpets and a 94.8 to 98.5 percent removal rate on non-carpeted surfaces). The second cleaning for carpets is credited with the same cleaning efficiency (14 to 37 percent), while non-carpeted surfaces are credited with a 70 percent reduction. Thus, the 15 percent of events in this category achieve a 26 to 60 percent cleaning efficiency on carpets and a 98.4 to 99.6 percent cleaning efficiency on non-carpeted surfaces. These events are independent of the contractor cleaning assumptions. The additional COF cleaning in this scenario is assumed to continue after the rule. The two scenarios are summarized in Table 5-4Table 5-4. Table 5-4: Scenarios for Benefits Analysis Category Scenario 1 Assumption Treatment of Equivalent of 20% of atbaseline protection risk occupants protected commensurate with the rule Initial cleaning by Contractor or in-house contractors (or instaff performing RRP house staff conducts initial cleaning performing RRP) in in all events. the baseline Initial cleaning by contractor (or inhouse staff performing RRP) of adjacent room in the baseline First COF cleaning Contractor or in-house staff performing RRP conducts initial cleaning in work room but does not clean the adjacent room. Routine COF cleaning for all events Scenario 2 Assumption 20% of at-risk occupants protected In addition to initial cleaning in all events, in 15% of baseline events contractors (or staff performing RRP) are assumed to achieve dust reduction levels beyond those credited to initial cleaning. Events in this category achieve a 26% to 60% cleaning efficiency on carpets and a 98.4% to 99.6% cleaning efficiency on non-carpeted surfaces. These events are in addition to the baseline events that protect 20% of the population as described above. In 10% of events, contractors (or staff performing RRP) are assumed to clean the adjacent room as well as the work room. These events are independent of the additional cleaning described above. The cleaning efficiency is assumed to be the same in the adjacent room as in the work room. In addition to routine COF cleaning, in 15% of the baseline events, COFs are assumed to achieve dust reduction levels beyond those credited to the first routine cleaning event. The 15% of events in this category achieve a 26% to 60% cleaning efficiency on carpets and a 98.4% to 99.6% cleaning efficiency on non-carpeted surfaces. These events are independent of the contractor cleaning assumptions. 11 12 13 14 15 16 17 18 19 20 21 5.2.5 Soil Loadings from Exterior RRP events According to the AHS survey, slightly over one-half of households who performed RRP conducted an exterior event; of these approximately 88 percent are exterior painting. However, estimates of lead loadings from exterior activities were not included as part of EPA’s 1997 Lead Exposure Associated with Renovation and Remodeling Activities: Environmental Field Sampling Study (EFSS). A more recent study done by the University of Illinois (2002) investigated five different methods used to remove lead paint from exterior surfaces of homes in preparation of painting these surfaces. This study contains data that can be used to estimate lead loadings from exterior painting. For the analysis of exterior events, only lead exposures resulting from exterior painting were evaluated. Other exterior events were not modeled. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 17 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 As a result, benefits of avoiding exposure to lead from these events may be underestimated. However, given that these un-modeled events constitute a relatively small proportion of the number of exterior events, and given that the area of lead paint disturbed in these events (e.g., replacing a deck) are relatively small compared to the amount disturbed by exterior painting, it is unlikely the benefits are significantly underestimated by this omission. To estimate exposures to lead resulting from exterior paint, the analysis first estimated the increase in lead concentrations in soil that could result from exterior RRP activities involving lead-based paint. In the University of Illinois study, lead loadings from the five paint removal methods were measured on six, 12inch by 12-inch collection plates evenly placed at designated intervals within a 6.5 foot by 11 foot area on the ground directly under, and centered on, the work area. Lead deposited to surface soils is mostly immobilized and is retained within the top 2 inches if left undisturbed. EPA (1986) documents that lead deposited from air is retained within 2-5 cm of topsoil. Therefore, in this analysis, the lead deposits were assumed to be distributed within the top 2 inches of soil. Transport of lead contaminated soil can occur through soil erosion, however, it is assumed that this erosion is negligible in residential areas due to appropriate land grading and good vegetative cover commonly found in these areas. Estimation of the concentration of lead in soils for a mass loading is a function of the dry density of the soil. Dry densities can vary from 1.1 g/cm3 for clays to 1.6 g/cm3 for sands. This analysis assumes an average density of 1.36 g/cm3 corresponding to a loam soil (EPA 1986). By dividing the lead loading on the collection plates by the mass of soil, the analysis obtained the additional concentration of lead in soil that could be contributed by exterior paint from each of the five methods. In the analysis, the five values were used to represent a distribution (each value with equal probability) of possible soil lead concentrations from exterior paint removal. This additional lead was added to background levels of lead assumed to be present in soil around TH-COF or COF in public or commercial buildings painted with lead based paint. The background soil lead level (490 ug/g) used for pre-1980 TH COFs was derived from the National Survey of Lead and Allergens in Housing (HUD 2000). For COFs in public or commercial buildings the background average lead concentration in soil was 74 ug/g and was derived from HUD's First National Health Survey of Child Care Centers (2003). The study data found that almost all (94.0 percent to 99.8 percent) of the lead fell on the front center plate, located directly beneath, and centered on, the work area. This 12-inch by 12-inch plate was placed so that its nearest edge was 6 inches from the base of the house. Therefore, the analysis assumed that the observed lead loading in this plate would fall over an area 18 inches around the perimeter of the house and calculated the resulting soil concentrations in this area. When adults or children are exposed to soil during outdoor play, yard work, gardening or other outdoor activities, they can be exposed to soil anywhere in the yard surrounding the COF. Therefore, the analysis averaged the concentration in the contaminated area with the background soil concentrations assumed to be present in the remainder of the yard. The perimeter and total yard areas of TH-COFs (i.e., single and multi-family homes) were calculated in the EPA Section 403 Economic Analysis (U.S. EPA 2001). For COFs in public or commercial buildings the total yard areas of single-family homes was used. The average perimeters of these buildings were estimated using the average square footage per-floor according to CBECS data (DOE 2003). When adults or children are exposed to soil during outdoor play, yard work, gardening or other outdoor activities, they can be exposed to soil anywhere in the yard surrounding the COF. Therefore, the analysis averaged the concentration in the contaminated area with the background soil concentrations assumed to §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 18 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 be present in the remainder of the yard. The area that is affected by the renovation is assumed to be the 18 inches along the perimeter of the building or house. The rest of the yard is assumed be at background levels of soil lead concentration. Single-family homes have an estimated perimeter of 157 feet, which can be multiplied by 18 inches to get a contaminated area of 235 square feet (U.S. EPA 2001). This is 8% of 2,988 square feet, the average yard size of a single-family home (U.S. EPA 2001).15 Since data on yard sizes were not available for daycare centers or schools, it was assumed that the share of the yard that is contaminated in public or commercial building COFs would be the same as estimated for single-family homes. Thus, the analysis implicitly assumes that the ratio of building size to yard size is the same for single-family homes and COFs in public or commercial buildings. The inputs used to calculate the soil concentrations from exterior paint activities are summarized in Table 5-5Table 5-5. Table 5-5: Soil Concentrations from Exterior Paint Activities Parameter Value Alkaline Chemical 10,242 mg/ft2 Paste Heat Gun 58,218 mg/ft2 Lead loadings in Paint Shaver 10,071 mg/ft2 front center plate Safe Stripper 46,556 mg/ft2 Wet Scrape 43,328 mg/ft2 Density of soil 1.36 g/cm3 Size of yard – single family home (applied to TH and COFs in public or commercial 2,988 ft2 buildings) Size of yard – multi-family home (applied to 6,417 ft2 TH COFs) Area within 18 inches of perimeter – single 235.37 ft2 family home Area within 18 inches of perimeter – multi402.47 ft2 family home 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Reference University of Illinois 2002 U.S. EPA 1986 U.S. EPA 2001 U.S. EPA 2001 See Chapter 4.5.5 See Chapter 4.5.5 The analysis assumes that the sizeone-quarter of exterior paint jobs varies from one wall of the building to the entire perimeter of the building, i.e., all four walls. (See Appendix 5D for details.)involve all sides of the building, one-quarter involve only one side of the building, and the other one-half involve areas in between one side and all sides. In calculating soil lead concentrations due to RRP events that involve all four sides of the building, the contaminated area is calculated as the entire area within 18 inches of the structure. These are the perimeter values presented in the table above. For exterior painting events that involve only one side of the structure, the soil concentration is calculated assuming that only one-quarter of the perimeter is contaminated, and the rest of the yard (including the other three sides of the structure) have background lead levels. The average of these two contamination levels are used to characterize contamination levels for the exterior painting events that are “in between.” For exterior renovations without the rule, it is assumed no cleaning or soil replacement occurs. Furthermore, no degradation of lead is assumed to occur over time. The containment that would be required under the rule consists of placing Under the rule, contractors (or in-house staff conducting RRP) 15 Two and one wall exterior events would have contaminated areas that are 4% and 2% of the yard. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 19 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 place plastic sheeting on the ground 10 feet out around the perimeter of the house to catch the dust and debris. This plastic must be properly removed and disposed of at the completion of the RRP activity., and they remove it prior to possible exposure. 5.3 Benefits Assessment -- Children Section 5.1 summarized the four steps for estimating the adverse health effects associated with renovation, repair, and painting projects. Section 5.3 presents the analytical details of steps two, three, and four as they relate to children. In step two the changes in blood lead levels due to exposure from RRP activities (with and without the rule) are estimated. In step three the resultant adverse health effects, including a reduction in IQ (with and without the rule) are estimated. In step four the dollar values of the reductions in adverse effects as a result of the rule are estimated. 5.3.1 Estimation of the change in blood-lead levels resulting from lead contamination Two models were used to estimate the change in blood-lead levels in children per change in lead dust loadings or soil lead concentrations: the EPA’s Integrated Exposure Uptake Biokinetic (IEUBK) model (U.S. EPA 2005) and a regression equation (also referred to as the Empirical Model) based on the Rochester Lead-in-Dust study (URSM and NCLSH, 1995; EPA 2000a). Exposure inputs (i.e., dust and soil loading or concentration) to both models were adjusted to reflect the time children spent in COF each day and over the course of a year. In the first approach, the analysis used EPA’s Integrated Exposure Uptake Biokinetic (IEUBK) model (U.S. EPA 2005). The IEUBK model estimates age-specific lead intake from environmental media (air, water, soil, dust, diet), models absorption and excretion and then simulates uptake and biokinetics in a child’s body to estimate the level of lead in blood. The IEUBK model predicts a central tendency estimate of blood lead concentration for children who might experience the inputted average exposures. However, within a group of similarly exposed children, blood lead concentrations would be expected to vary among children as a result of inter-individual variability in media uptakes (e.g., daily average intakes of soil-derived dust, drinking water, or food), absorption, and biokinetics. The model simulates the combined impact of these sources of variability as a lognormal distribution of blood lead concentration for which the geometric mean is given by the central tendency blood lead concentration outputted from the biokinetics model, and the geometric standard deviation is an input parameter (U.S. EPA 2006). The model can be used to predict age-specific blood-lead levels for children ages 6 months to 84 months. For dust exposures from interior RRP activities, exposures occur from the time that RRP activity is completed (includingafter initial clean-up) until the time that the dust falls to background levels due to routine COF cleaning (background is around 1.1 µg/ft2, the typical background level of floor dust cited in Jacobs et al., 2002). In the modeling exercise performed for this assessment, where initial and routine cleaning are assumed to reduce dust levels over time, the dust levels generally fall to background levels in less than one year, especially for COFs with daily cleaning. However, because IEUBK predicts blood-lead for children in one-year age brackets, the analysis calculated the average lead dust during the entire year when RRP activities take place and used this value to represent the average dust exposure during the year that a child is exposed. The average lead dust was further adjusted for the time children spend in COF, which is described in more detail below. At all other times during early childhood (from ages 6 months §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 20 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 to 6 years), children are assumed to be exposed to background levels of household lead dust, equal to 1.1 µg/ft2. For soil exposures from exterior painting, the analysis assumes that lead loadings remain in the soil and are not cleaned up or abated after exterior painting work is completed. Therefore, children are assumed to be exposed to contaminated soil from the year when exposure begins through the age of 5 years. The IEUBK requires an estimate of dust concentration rather than loadings. To estimate the relationship between floor dust loadings and hand lead dust concentrations, this analysis used a regression equation to describe hand dust-lead measures (PbH, µg) as a function of floor dust-lead loading (PbF, µg/ft2), estimated using geometric mean floor dust-lead loadings and mean hand dust loadings from Clark (1985, as cited in Battelle 2005).(Topping 2005). An alternative approach to the use of the IEUBK is to estimate the blood-lead level directly as a function of floor lead dust loadings. In fact, the Rochester Lead-in-Dust study (URSM and NCLSH, 1995) concluded that lead dust loading is a better predictor of children's blood-lead levels than is lead dust concentration. This analysis used a regression equation that directly links dust loadings and blood-lead levels, previously presented in U.S. EPA (2000a), which was based on data from the Rochester Lead-inDust Study. The regression model used (Model C) was16: log(PbB) = 1.337 + 0.140 * log(PbF) + 0.004 * (PbP) + error where: PbB = PbF = PbP = Error = the blood-lead concentration, µg/dL; the floor lead dust loading, µg/ft2; and the percent of painted surfaces containing deteriorated lead-based paint, assumed to be 0% in this analysis. 0.580 This model applies only to lead dust and is not used to describe blood-lead concentrations from soil lead exposures. Children spend only a portion of the day (and the year) in COFs. Age specific number of awake hours per week children spend in non-parental care and duration of school year was used to time-weight exposure to lead-based hazards generated by renovation and repair activities. Table 5-6Table 5-6Table 5-6 shows adjustment factors for 0 to 5 year olds. Mean hours per week in COF were obtained from Mulligan et al (National Center for Education Statistics, 2005). Awake hours per day was calculated as the difference between 24 hours and the average number of hours children sleep.17 Mean hours awake per week was the product of awake hours per day and seven days/week. For children in kindergarten and prekindergarten and kindergartenschool settings (kindergarten, pre-kindergarten, and daycare centers in schools) an attendance of 36 weeks per year was used,18 while in all other settings year-round attendance (52 weeks) was assumed. Values for kindergarten schools were further adjusted to reflect the 60 percent of kindergarteners in full-day (6 hour) and 40 percent in half-day (3 hour) settings (Wirt et al., 2004). The exposure adjustment factor was applied to the loadings estimates prior to modeling the blood lead levels 16 This is logarithm to the base e or natural log. Accessed December 12, 2006 http://www.webmd.com/content/article/122/114620.htm Edited by Cynthia Haines, MD, WebMD, February 2006. SOURCE: WebMD Feature: "How Much Sleep Do Children Need?" © 2006 WebMD Inc. All rights reserved. 18 Centers or preschools housed in a K-12 school setting are assumed to follow the typical school year of approximately 36 weeks. 17 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 21 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 in order to account for the fraction of time that children spend in a COF. Children spend only a portion of the day (and the year) in COFs. Age-specific number of awake hours per week children spend in COFs and duration of school year were used to time-weight exposure to leadbased hazards generated by renovation and repair activities (Table 5-6Table 5-6Table 5-6). Since dust ingestion is only likely to occur during waking hours, the average hours awake is used as the denominator in the adjustment equation. Mean hours per week in COF were obtained from Mulligan et al19. Awake hours per day were calculated as the difference between 24 hours and the average number of hours children sleep.20 Mean hours awake per week were the product of awake hours per day and seven days/week. For children in school settings (kindergarten, pre-kindergarten, and daycare centers in schools) an attendance of 36 weeks per year was used,21 while in all other settings year-round attendance (52 weeks) was assumed. Values for schools were further adjusted to reflect the 60 percent of kindergarteners in full-day (6 hour) and 40 percent in half-day (3 hour) settings22. The exposure adjustment factor was applied to the loadings estimates prior to modeling the blood lead levels in order to account for the fraction of time that children spend in a COF. 19 Mulligan, G.M., Brimhall, D., and West, J. (2005). Child Care and Early Education Arrangements of Infants,Toddlers, and Preschoolers: 2001 (NCES 2006-039). U.S. Department of Education, National Center for Education Statistics. Washington, DC: U.S. Government Printing Office. 20 There are a number of estimates on how much sleep children need. These estimates vary slightly but are all similar. The following table includes estimates from WebMD, the American Academy of Pediatrics (AAP), and the Nemours Foundation. This information in Table 5-6 is taken from WebMD. Reference 1-4 weeks old 1-4 months old Between Birth-6 Months 4 – 12 months Between 6-12 Months, Between Ages 1-3 Between Ages 3-5 1 Estimate is for 3-6 year olds. WebMD 15-16 hrs 14-15 hrs AAP 16-20 hrs 14-15 hrs 14-15 hrs 12-14 hrs 10-13 hrs 10-12 hrs 1 10-12 hrs 2 2 Estimate is for 3-10 year olds. Nemours Foundation 16-20 hrs 14 hrs 10-13 hrs 10-12 hrs WebMD, Kids and Sleep: How Much Sleep Do Children Need? http://www.webmd.com/sleepdisorders/guide/sleep-children AAP. Cohen, George J., M.D., F.A.A.P. (Ed.). (1999). American Academy of Pediatrics Guide to Your Child’s Sleep. New York: Villard Nemours Foundation. The Nemours Foundation brochure All About Sleep is provided at MedLine Plus, http://www.nlm.nih.gov/medlineplus/childrenshealth.html . Nemours Foundation, All About Sleep, http://kidshealth.org/parent/general/sleep/sleep.html 21 Centers or preschools housed in a K-12 school setting are assumed to follow the typical school year of approximately 36 weeks. 22 Wirt, J., Choy, S., Rooney, P., Provasnik, S., Sen, A., and Tobin, R. (2004). The Condition of Education 2004(NCES 2004-077). U.S. Department of Education, National Center for Education Statistics. Washington, DC: U.S. Government Printing Office. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 22 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 Table 5-6: Age Specific Adjustment Factors for Time Weighted Inputs into Blood Lead Models Mean hours Mean hours Attendance Exposure Hours awake Total weeks Age (years) per week in awake per (weeks per Adjustment per day per year COF week year) Factor COFs Outside of Schools (Daycare Centers not in Schools, and COFs in Target Housing) 0 30.3 9.5 66.5 52 52 0.456 1 31.7 11 77 52 52 0.412 2 32.3 11 77 52 52 0.419 3 29.7 11 77 52 52 0.386 4 29.6 13 91 52 52 0.325 5 29 13 91 52 52 0.319 COFs in Schools (Daycare Centers in Schools, Pre-Kindergartens, and Kindergartens) 0 30.3 9.5 66.5 36 52 0.315 1 31.7 11 77 36 52 0.285 2 32.3 11 77 36 52 0.290 3 29.7 11 77 36 52 0.267 4 29.6 13 91 36 52 0.225 5 29 13 91 36 52 0.183 4 5 6 7 8 9 10 11 An example calculation for 1 year olds in COFs outside of schools is shown below. • • • • 31.7 hours/week * 52 weeks/year = 0.412 77 hours/week * 52 weeks/year 12 13 14 15 16 17 18 19 20 21 22 23 24 Mean number of hours per week 1 year olds spend in COF (31.7 hrs). Mean number of hours per week 1 year olds are awake (77 hrs). Number of weeks a 1 year old attends non-parental care arrangements (52 weeks). Total number of weeks per year (52 weeks). In this example, RRP floor loadings are adjusted downward by 0.588 (or 1-0.412) to estimate the lead load to which the child is exposed. Research indicates that for children in kindergarten through 2nd grade about 20% have a non-parental care arrangement before school and 48% have a non-parental care arrangement after school.23 The likelihood of after school arrangements does not appear to differ substantially among age groupings (i.e., 54% of 6-8 graders have non-parental arrangements after school). This is important because the average number of hours spent in these settings is presented in the aggregate for kindergarten through 8th grade, i.e., data are not reported for kindergartners alone. Overall, children who were in before-school arrangements spent an average of 4.7 hours per week in this setting (i.e., about 1 hour/day). Children who were in after-school arrangements spent an average of 9.0 hours per week in this setting (i.e., about 2 hours/day). 23 U.S. Department of Education, National Center for Education Statistics. Before- and After-School Care,Programs, and Activities of Children in Kindergarten Through Eighth Grade: 2001, NCES 2004-008, by Brian Kleiner, Mary Jo Nolin, and Chris Chapman. Project Officer: Chris Chapman. Washington, DC: 2004. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 23 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 The COF analysis does not take into account RRP activities that may occur in before and after-school care arrangements for kindergarteners. This occurs primarily because of the expected complexity of adding this new configuration to the analysis and in part the limited data on kindergarteners. This could lead to underestimates of benefits. For example, kindergarteners who attend half-day programs may spend an equal or greater amount of time in before and after-school arrangements. However, there has been a long-term trend from half day to full day kindergartens. Thus, any underestimate of benefits that occurs as a result of not accounting for before- and after-school programs in children attending half day programs could diminish with time. 5.3.2 Estimation of the adverse health effects (e.g., loss in IQ points) due to increased bloodlead Estimation of IQ loss must consider both the dose (peak or average) or blood-lead level and the timing of the exposure. Because the levels of lead dust generated from a single RRP event are expected to remain at increased levels for only a short period of time (generally less than a year), the exposure occurs only during a fraction of a child’s life between the ages of 0 and 6 years. Furthermore, due to the particular vulnerability of the nervous systems of very young children (around the age of 2 years old), the timing of the short-term exposure may also influence the magnitude of adverse health effects associated with the exposure. This section briefly reviews and highlights the literature that serves as the source in this analysis for blood lead (dose)-IQ decrements. Note that the literature on this topic often discusses both exposure dose and timing together. The next section (5.3.3) describes several exposure conditions used in the analysis that vary the duration and timing of the exposure. Numerous cross-sectional and longitudinal studies have estimated the decrement in IQ points associated with an increase in blood-lead levels. In cross-sectional studies, blood-lead levels are measured only once; usually in very early childhood between 6 and 24 months, when children’s mouthing behavior leads to relatively high rates of ingestion of dust and soil, and when children are believed to be particularly vulnerable to neurological effects from exposure to lead. In longitudinal studies, blood-lead levels are typically measured several times during the course of young childhood (around 6 months to around 6 years) and may be presented in several ways: as the peak blood-lead concentration (typically occurring around 24 months of age), as the average “lifetime” blood-lead concentration (where lifetime is equal to the time of first measurement until the time of the administration of the IQ testing); and blood-lead levels concurrent to the time of testing. These metrics are, of course, closely correlated. The studies reviewed for this analysis are listed on Table 5-7Table 5-7. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 24 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5-7: Association between Blood-lead Levels and IQ Initial Blood-lead Levels of Study Study Metric Participants Dietrich 1993, Peak and concurrent blood-lead 20 ug/dL or 1995 levels greater Peak and child lifetime average 20 ug/dL or Tong 1996 blood-lead levels greater Peak, average, concurrent 20 ug/dL or Chen 2005 blood-lead levels greater Child lifetime average, Child lifetime concurrent, average during average of 7.4 Canfield 2003 infancy (6-24 months) and peak µg/dL; Peak value concentrations of blood-lead of 11.1 µg/dL levels Child lifetime average, Lanphear 2005 concurrent, peak, and very early Meta-analysis childhood blood-lead levels Peak Exposure: Cross sectional studies: Bloodlead level measured once Schwartz 1994 Meta-analysis Longitudinal studies: integrated blood-lead up to age three and the 24-month blood-lead level Canfield 2003 Average Exposure: Child lifetime average, concurrent, average during infancy (6-24 months) and peak concentrations of blood-lead levels 7.4 µg/dL child lifetime average among all participants Study Results Concurrent blood-lead levels were more strongly associated with IQ than peak-blood levels Child lifetime average blood-lead levels were more strongly associated with IQ than peak-blood levels Concurrent blood-lead levels always had the strongest association with IQ, and this association grew stronger with age All four measures were significantly related to IQ declines, but the child lifetime average and concurrent blood-lead levels were more highly significant than average infancy and peak blood-lead concentrations Concurrent or child lifetime average blood-lead levels more strongly associated with IQ declines than peak or very early childhood blood-lead concentrations; IQ decline was significantly greater among children whose peak blood-lead was below 7.5 µg/dL Overall decline in IQ points: 0.257 IQ points per 1 µg Pb/dL increase in blood-lead; higher IQ point losses among children with blood-lead levels below 15 µg/dL Overall: IQ points decreased by 0.46 per 1 µg/dL change in child lifetime average blood-lead Sub-analysis: Children with peak blood-lead level <10 µg/dL had greater IQ points loss per 1 µg/dL change in blood-lead than those with blood-lead levels >10 µg/dL Linear model: estimated a 1.37 reduction in IQ points for an increase of 1 µg/dL in child lifetime average blood-lead concentrations Non-linear model: estimated a 7.4 reduction in IQ points for an increase of 10 µg/dL in child lifetime average blood-lead. Children with peak blood-lead levels above 10 µg/dL have an estimated reduction of 2.5 IQ points as blood-lead levels increase from 10 µg/dL to 30 µg/dL 2 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 25 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 Peak blood-lead levels typically occur when children are young (around 2 years of age), but neurological function is difficult to measure until children are older (at around age 4 years or more). The question remains whether peak levels of exposures early in life are the cause of neurological effects reflected later in life or if the effects are the result of cumulative effects of exposures during early childhood. Because of the difficulty in answering this question, different theories have emerged. Some researchers have argued that peak lead exposures occurring around the ages of 1 to 2 years will affect neurological development more critically than exposures at other ages. Before the age of 2 years, the neural network in the brain is still undergoing substantial development, and many basic cognitive functions form during this period (ATDSR 1999). In a longitudinal study of lead-exposed children in Boston, researchers found a significant relationship between IQ and blood-lead levels at 2 years of age, but not with bloodlead levels at 57 months or 10 years of age (Bellinger 1992). Several more recent studies have found a stronger association with child lifetime average or with concurrent, rather than peak exposure, and have found an attenuation of association between peak bloodlead levels and IQ over time. Note that in this circumstance, “lifetime average” is defined as the mean blood lead from the first blood lead test (usually around 6 months) to concurrent blood lead tests. For clarity, this assessment will henceforth refer to this measure as “child lifetime average.” The “concurrent blood lead” test is the blood lead measurement made closest in time to the IQ tests, which are typically administered between 5 and 7 years of age. Dietrich (1995) and Dietrich (1993) found that concurrent blood-lead levels were more strongly associated with IQ than peak blood-lead levels, while Tong (1996) found that child lifetime average blood-lead levels were more strongly associated with IQ than peak blood-lead levels. Chen (2005) examined the question by following children from about 2 years until 7 years of age, and comparing peak, average and concurrent blood-lead levels to IQ at different ages. Chen found that concurrent blood-lead levels always had the strongest association with IQ, and this association grew stronger with age. This result suggests that peak exposures at the age of 2 did not fully account for the neurological effects seen in older children, and that lower level, concurrent exposures are also influential. Schwartz et al. (1994) conducted a meta-analysis based on the results of seven studies (three longitudinal and four crosssectional). Blood-lead was only measured once in the cross-sectional studies, and these point-in-time measures were used in the meta-analysis as the exposure metric. For two of the longitudinal studies, the integrated blood-lead up to age 3 was used, while for one of the longitudinal studies, the 24-month bloodlead was used. An increase in blood lead from 10 to 20 µg/dL was associated with a decrease in 2.6 IQ points. All of the studies were generally conducted on children with higher blood-lead concentrations (around 20 µg/dL or more). Other recent studies have examined the question of whether these results applied to children with lower blood-lead levels. Canfield (2003) studied the relationship of blood-lead measurements throughout early childhood to IQ measured at 3 and 5 years of age, using a variety of regression models. Among children included in the Canfield (2003) data analysis, the average blood-lead level was relatively low (child lifetime average of 7.4 µg/dL and a peak value of 11.1 µg/dL). The authors evaluated child lifetime average, concurrent, average during infancy (6-24 months) and peak concentrations of blood-lead levels. Although all four measures were significantly related to IQ declines, even after adjustment for covariates, the authors found that the child lifetime average and concurrent blood-lead levels were more highly significant (as measured by the P-value) than average in infancy and §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 26 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 peak blood-lead concentrations. In a recent meta-analysis, Lanphear et al. (2005) also found the concurrent or child lifetime average blood-lead levels to be more strongly associated with IQ decrements than peak or very early childhood blood-lead concentrations. This analysis uses two different metrics for IQ loss associated with increased blood-lead levels. In the first approach, peak exposures are assumed to be the most relevant exposures for predicting IQ point loss. In this case, the blood-lead levels associated with floor lead dust exposures from RRP activities are averaged over one year, and are assumed to represent a peak of exposure between the ages of 0 and 6 years old. The change in IQ from a 1-µg/dL increase in peak blood-lead concentration is predicted using the results from Schwartz (1994). The overall estimated decrement in IQ was estimated to be 0.257 IQ points per 1µg Pb/dL increase in blood-lead. In the second approach, the analysis assumed that the IQ decrement is associated with the average exposure over the lifetime of the child between the ages of 6 months and 6 years old (that is, over the course of 5 and one-half years). This analysis used the Canfield (2003) overall estimate of -0.46 IQ point per 1 µg/dL. Because the estimated blood-lead levels among children in COFs may range from below 10 µg/dL to above 10 µg/dL the COF analysis used the Canfield estimate (-0.46 IQ per 1 µg/dL), which was derived using data across all blood-lead ranges observed in that study, rather than the greater IQ loss estimated for children with blood-lead levels below 10 µg/dL. 5.3.3 Alternative Blood-lead Assumptions Based on the epidemiological evidence discussed above and elsewhere in the extensive literature on the topic, children 0 to 6 years old are considered to be the population at highest risk from lead contaminated dust and soil. The appropriate blood-lead metric is an important consideration for this proposed rulemaking, because the exposure associated with RRP activities is not expected to be a chronic exposure. If, in fact, this short-term lead exposure occurs during a critical window of development, where lead can cause permanent cognitive and neurobehavioral deficits, then the full IQ decrement can be attributed to this period of exposure, and the appropriate metric of exposure is the blood-lead concentration measured during this developmental period. If in contrast, the IQ decrement is associated with the cumulative exposure of the child, then the more appropriate blood-lead exposure metric is the child lifetime average blood-lead concentration, where the blood-lead levels resulting from exposures to RRP generated dust are averaged with blood-lead levels during the rest of early childhood. While the causal relationship between lead exposure and IQ reduction in children is well established, there remains some uncertainty about the exact form of the dose-response functions and whether exposure should be measured in terms of peak exposure or average exposure over the first six years of life. This analysis varies two possible assumptions to address this uncertainty. In the first condition, the analysis evaluates risks to children 1 to 2 years (i.e., 12-23 months) old only (Table 5-8Table 5-8, set 1a-1e). At this age, behaviors such as mouthing and crawling lead to relatively high levels of dust and soil exposure, and, as discussed earlier, nervous system structures are considered particularly vulnerable to adverse effects from lead exposures at this age. In this case the analysis calculates peak blood-lead and child lifetime average blood-lead concentrations, assuming that the child is exposed to RRP-contaminated dust between the ages of 1 and 2 years old [12-23 months], and to RRP-contaminated soil beginning at age 1 year. When calculating benefits, the size of the population of children ages 1 to 2 years [12-23 months] estimated to be attending a COF with RRP events was used. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 27 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Children of other ages will be exposed to lead from RRP activities, and may have health effects associated with these exposures; however, it is not known if exposures of the duration associated with RRP events occurring only at these other ages will have the same influence on IQ as exposures that occur at around the age of 2 years (23 months). Therefore, in this first condition, risks to other age groups are not assessed quantitatively. Thus, IQ benefits may be underestimated. In the second condition, the analysis assumes that short-term lead exposures that occur at any age under 6 years have the same magnitude of effect on neurological decrement as exposures that occur when children are around the age of 2 years (Table 5-8Table 5-8, set 2a and 2b). In this case, risks to all children under the age of 6 were assessed. Because exposures to lead in dust or soil for young children could occur at any age under 6 years old, the analysis calculated the effect on blood-lead for exposures occurring at different times during young childhood. To do so, the analysis first calculated the effect of a single year of exposure to lead-contaminated dust for each age group separately (i.e., 1, 2, 3, 4, and 5 year olds). That is, the analysis first calculated the peak and child lifetime average blood-lead for a child, assuming exposure to RRP dust occurred when the child was 6 months to 1 year old; then assuming the exposure occurred when the child was 1 to 2 years (i.e., 12-23 months) old; then assuming exposure occurred when the child was 2 to 3 years (i.e., 24 to 35 months) old, and so on, for each one-year age group until the age of 5 to 6 years. The peak and child lifetime average blood-lead values for each age group were then weighted by the proportion of children in each age group in the U.S. population, according to the 2000 Census to derive “population-weighted average” peak and child lifetime average blood-lead values associated with exposures to dust. Similar calculations were made for exposures to lead-contaminated soil for each one-year age range, assuming exposures start at the beginning of the age range and continue until the age of 6 years. All children under the age of 6 years within the populations for each regulatory option were then used to calculate benefits. Using a population-weighted average for benefits assessment assumes that using a population average will yield equivalent results to calculating lost IQ points based on the blood-lead metrics derived for each age group separately. Table 5-8: Alternative Blood-Lead Assumptionsa,b Blood-lead Age Group Blood-Lead Model Assumptions Set 1a 1-2 years old IEUBK Set 1b 1-2 years old IEUBK Set 1c 1-2 years old Empirical Model Set 1e 1-2 years old Empirical Model Set 2a 0 thru 5 years old IEUBK Set 2b 0 thru 5 years old IEUBK 31 32 33 34 35 36 37 38 39 Lead Exposure Metric Peak (age 1-2 year old) Average (age 0 thru 5 years old) Peak (age 1-2 year old) Average (age 0 thru 5 years old) Peak (at any age 0 thru 5 years) Average (age 0 thru 5 years old) a Age Group refers to the age at which exposure has an effect, and Lead Exposure Metric refers to whether exposure level is measured in terms of peak or six-year average lead exposure. b The Empirical Model was not used for Assumption Sets 2a or 2b because these sets assume that the relevant age group is children from 0 through 5 years of age. The Empirical Model, however, was derived from observations of very young children (around 2 years old) and automatically incorporates the behaviors of these young children when estimating how floor dust relates to blood lead. Because older children have different behaviors than very young children, the Empirical Model does not represent the relationship between blood lead and floor dust lead in older children. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 28 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 5.3.4 Assignment of cost of adverse health effects The estimated value of an IQ point is $12,953 (2005 dollars), which is derived from coefficients provided by Salkever (1995). The IQ value is modeled as the present value of a loss in expected lifetime earnings due to a one point IQ drop.24 The present value is calculated assuming that while most people start working at age 18, average income in the early adult years is reduced because some are still in school. In addition, the present value assumes a retirement age of 67 years old, due to the revisions of the Social Security law that are incrementally increasing the retirement age so that it will be at age 67 by the time today's children are retiring. Further, the analysis assumed that children would be affected by lead at 3 years of age, the median of the range when children are most susceptible to lead hazards. As a result, the value of an IQ point is only discounted back to age 3. Limiting the valuation estimation to reduced income underestimates the value of children’s neurological benefits. Additional measures of the impact on IQ are: additional education costs for special and remedial education, and medical costs to treat very high levels of lead. This analysis does not generate the information needed to estimate the number of such cases, so these measures are not included in the valuation of children’s benefits. 5.4 Benefits Assessment – Adults There is evidence that adult exposure to lead is linked to various adverse health effects such as hypertension, coronary heart disease, stroke, and premature mortality. Epidemiologic studies have consistently demonstrated associations between Pb exposure and enhanced risk of deleterious cardiovascular outcomes, including increased blood pressure and increased hypertension (EPA 2006). However, there is sufficient uncertainty about the level of exposure that adults working in COF will experience that this analysis did not attempt to estimate the number of cases that would be avoided due to the regulations under consideration. Thus the benefits valuation estimates do not include adult benefits. However, the analysis does present the number of adults who will potentially experience reduced exposures to lead as a result of implementation of the rule. 5.5 5.5.1 Benefits Assessment – Statistical Models and Assumptions. The Monte Carlo Model A Monte Carlo modeling approach was used to estimate blood-lead (and thus health effect) levels in both the baseline and with the regulations in place. A Monte Carlo approach was used because it provides a means of incorporating the uncertainty around each input parameter, as long as the uncertainty can be described in terms of a statistical distribution. (For details, see section 5.5.2 and Appendix 5E.) For each run (iteration) of the Monte Carlo model, a single value for each uncertain parameter is drawn from the distributions describing the uncertain parameters. In this analysis, for example, an iteration consists of drawing from a (pseudo) random number generator the (1) cleaning efficiencies and (2) the IQ reduction parameter. Given these 2 randomly selected variables, the iteration is evaluated for every possible floor type, cleaning frequency, and dust characteristic: 2,682 possible combinations, which are weighted according to their likelihood. 24 Present value of earnings calculated at a 3 percent discount rate. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 29 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 This analysis treats each input as a statistically independent parameter. The expected benefits are calculated given this set of values for the uncertain input parameters. This is done multiple times, keeping track of the expected benefits calculated at each model run. The result of the Monte Carlo approach is a distribution of values for expected benefits that is described by its mean, standard deviation and percentile values. Uncertainty distributions are incorporated into the benefits model for key inputs. For example, the initial lead loadings used in estimating exposures are drawn from distributions developed based on the frequency of different RRP activities and associated lead dust and soil concentration estimates. Further, for indoor exposure, the frequency of COF cleaning and the cleaning efficiency of vacuums on carpet and vacuuming/sweeping on non-carpeted floors are uncertain. Using values from the literature the analysis characterizes the uncertainty of these important inputs. Several inputs that are known to be uncertain are treated as point estimates in the benefits model because the data that are necessary to statistically describe the uncertainty are not available. Therefore, although the benefits model incorporates as much input uncertainty as feasible, the outputs described below underestimate the level of uncertainty in the estimation of the benefits associated with the rule options. 5.5.2 Summary of Inputs to the Monte Carlo Model The input parameters are listed in Appendix 5E along with their value (for those with a point estimate) or their distribution. Certain inputs to the Monte Carlo model account for variability by incorporating a range of possible values. Where all values in this series are assumed to have equal probability of occurring, for example different parameters for initial cleaning efficiency, these are termed uniform distributions in Appendix 5E. Other inputs given by a range of values consist of custom distributions, with specific probabilities for each value in the range. For example, for the variable of soil lead concentration outdoors without the rule, different concentrations are likely depending upon the unit type and the system of paint removal used. These values are not equally likely to occur because of the different frequencies of unit types. The specific distributions used for the custom distribution parameters are given in Appendix 5D. The size of the population that has a reduced exposure to lead because of the rule is described in detail in Chapter 2. Five major steps are taken to estimate this population: (1) estimate the number of individuals attending a COF where there is a regulated RRP event; (2) adjust the time spent in the COF for awake hours per week and school year duration; (3) estimate the proportion of regulated RRP events where leadbased paint (LBP) is present; (4) adjust for the baseline use of containment and clean-up practices to account for individuals who would avoid exposure in the absence of regulation; and (5) adjust for compliance rates. The basic analysis presented in this chapter assumes a 75 percent compliance rate, based on analyses of compliance rates with OSHA construction regulations. (See section 4.2 for a discussion of the compliance rate assumptions.) Alternative compliance rates are presented in the Sensitivity Analyses shown in Chapter 7. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 30 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 5.6 Results This analysis estimates the benefits of the proposed regulation in terms of IQ deficits in children. Quantitative estimates of benefits are provided by regulatory option (A, B, C, D, or E) and year (first or second) of benefit. Options A, B, C, and E are described as flexible, which means that certified renovators can rely upon their training to determine how much containment is necessary and practical in any particular situation. These options differ in terms of the universe of COFs subject to the rule’s requirements during the first year of the rule, i.e., Option A applies to pre-1978 construction, Options B, D, and E to pre-1960, and Option C to pre-1950. The containment, cleaning, and verification practices required under Option D are the same as those in the flexible options; however, Option D prescribes the size of the work area that must be contained, cleaned, and verified. Option E is the same as Option B but includes estimated benefits for COFs in target housing where renovation work is conducted by do-ityourselfers (TH-DIY). Starting in the second year the rule applies to pre-1978 construction under each of the five options. The proposed rule is Option B.Option B is the option being proposed in the supplemental proposed rule. Scenarios (1 and 2) are provided as well. Scenario 1 assumes that the equivalent of 20 percent of all atrisk occupants are protected in the baseline commensurate with the results of the rule. This scenario is meant to cover all baseline activities. Scenario 2 assumes that additional cleaning activities are performed in another 10 to 20 percent of events. This additional cleaning goes beyond the efficiency of initial or routine cleaning, but does not necessarily reach the level of protection achieved by the rule. 5.6.1 Summary of Results Table 5- 9Table 5- 9 displays the number of cases of avoided exposures by adults and children protected by due to the rule along with the number of RRP events.25 The number of avoided exposuresindividuals protected in Table 5- 9Table 5- 9 reflects the following adjustments: 1) the proportion of regulated RRP events during a 1-year period where lead-based paint (LBP) is present; 2) the baseline use of containment and clean-up practices to account for reductions in exposure in the absence of regulation; and 3) an assumption of a 75% compliance rate with the rule. For adults, the number of cases of avoided exposuresindividuals benefiting from the rule areis estimated but the benefits of thesecases avoided exposures have not been quantified or monetized. Table 5- 9Table 5- 9 shows that during the first year of the rule approximately 409 to 920 thousand cases of exposures to children are avoidedprotected, while during the second year 916 thousand cases of exposure to children are avoidedprotected in Options A through D and 987 thousand cases of exposure to children are avoidedprotected under Option E. Similarly, there are 112 to 225 thousand cases of avoided exposures in adults are protected during the first year andwhile 224 to 273 thousand are protected during in the second year. Children 6-17 years old could also experience reduced exposures as a result of the rule. These individuals could be older children living in target housing COFs, attending schools where RRP occurs in common areas (i.e., the cafeteria), or attending locations (i.e., churches) that may host daycare centers and where RRP also occurs in a common area. Table 5- 9Table 5- 9 indicates that there are 916,000 annual exposures in children 25 There may be more cases of avoided exposures than individuals protected by the rule, since an individual can be exposed to lead from more than one event (for example, from both an interior renovation event and an exterior event). §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 31 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 estimated to be avoided as a result of the supplemental rule, starting in the second year. Table 5- 10Table 5- 10Table 5- 10 provides the estimates by interior and exterior events, and by type of COF. The benefits displayed in Table 5-11Table 5-11 through Table 5-14Table 5-14Table 5-14 are annual benefits.26 They result from the avoided reductions in IQ among children who are living in units where RRP events occur during a single year’s time. Because these benefits occur from avoiding exposure to lead from a single RRP event, and each year a new group of regulated RRP events occurs, each with its own population of at-risk individuals potentially exposed thus generating a new group of at-risk individuals potentially exposed, every year there will be benefits of approximately this magnitude.27 Table 5-11Table 5-11 presents benefits in thousands of IQ points gained as a result of the rule while Table 5-12Table 5-12 presents the benefits in dollars and Table 5-13Table 5-13 presents median, 5th, and 95th percentile values of dollar benefits. In terms of the value of the benefits of the proposed options; they closely track the total number of IQ points. Looking at first year children’s benefits for Scenario 1 (Table 5-12Table 5-12 or Table 5-13Table 5-13), Option A provides the greater benefits (year 1 range $69 to $272 million) relative to Options B and C. Under Option B, first year benefits (range $55 to $210 million) are about 20 percent smaller than first year benefits under Option A. Under Option C, first year benefits (range $38 to $144 million) are about 45 percent smaller than first year benefits under Option A. These differences are attributable to the greater number of persons protected under Option A in the first year. These differences disappear in the second year when all three options cover the same housing units and thus the same number of people. Option E benefits are as much as 70 percent larger relative to Option B due to the additional children protected. Table 5-11Table 5-11, Table 5-12Table 5-12, and Table 5-13Table 5-13 also show that estimating children’s IQ benefits using the Empirical Model for children age 1 to 2 years yields consistently higher results than using the IEUBK model for children age 1 to 2 years, but lower results than using the IEUBK model for children age 0 to 5 yearspeak and 6 year-average exposures in 1-year olds. Limiting the estimated benefits to 1 to 2 year olds The highest benefits in terms of number of children’s IQ points are estimated in IEUBK using peak blood lead as the exposure metric, and attributing the same magnitude of neurological decrement associated with this peak to all children under the age of six years, regardless of the age at which exposure occurs (Table 5-10 and Table 5-11). The lowest benefit in terms of number of children’s IQ points gained are derived using IEUBK model to estimate a six-year blood lead average for 1 to 2 year olds; however, this estimate almost certainly underestimates benefits because it assumes that there are no benefits in avoiding exposures to other age groups. Scenario 2, which assumes that additional cleaning activities are performed in another 10 to 20 percent of events, has lower benefits relative to Scenario 1. Benefits under Scenario 2 range from two to five percent lower than Scenario 1. For example, first year benefits under Option B range from $55 to $210 million in Scenario 1 and from $54 to $200 million in Scenario 2. 26 Benefits for Option D are not displayed in the following tables because the analysis treats them the same as those for Option B; Options B and D cover the same universe of households. 27 As discussed in more detail in Chapter 4, the number of RRP events covered by the rule will decline slowly as the regulated housing stock declines due to demolitions. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 32 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 Benefits are also examined in terms of the proportion attributable to reductions in interior and exterior event exposures. Table 5-14Table 5-14Table 5-14 and Table 5-15Table 5-15Table 5-15 present the number of IQ points gained and monetary benefits by regulatory option, year, and interior/exterior event. In the IEUBK peak exposure models the majority of child benefits (54% to 75%) are due to reductions in interior event exposure. However, in the IEUBK average exposure models the majority of benefits (55% to 73%) are due to reductions in outdoor event exposures. In the empirical model runs, for both peak and average exposures, the majority of child benefits (67 to 85%) are due to reductions in interior event exposures.28 Table 5-16Table 5-16 shows annualized mean 50 year benefits of avoided exposures at 3% and 7%, with Scenario 1 in Table 5-16Table 5-16a and Scenario 2 in Table 5-16Table 5-16b. Table 5-17Table 5-17 through Table 5-18Table 5-18 summarize median, 5th, and 95th percentile total 50-year benefits and 50year annualized benefits by regulatory option, year, and discount rate. Table 5-19Table 5-19Table 5-19 summarizes the range of 50-year annualized benefit estimates across the six alternative sets of estimates using different blood lead models, exposure estimates, and ages of children. Using a 3 percent discount rate, mean children’s IQ benefits under Scenario 1 range from about $65 to $386 million across the six different modeling estimates; under Scenario 2 they range from about $64 to $361 million. Using a 7 percent discount rate, all the mean annualized values are slightly higher than they are when a 3 percent discount rate is used. Using Option B as an example, Table 5-20Table 5-20Table 5-20 shows the contribution to benefits by each COF type. The largest proportion of benefits are derived from contractor RRP work in TH COFs (range 62 to 75 percent based on mean values). This pattern was similar between Scenarios and discount rates. Table 5-21Table 5-21Table 5-21 shows the dollar value of estimated benefits for interior and exterior events, and by type of COF, under the six different alternative sets of blood-lead modeling results. As shown below, regardless of which scenario is considered, the estimated benefits are substantial. In addition, in both scenarios a number of benefit categories have been excluded from the estimated benefits. Among the categories of benefits excluded from either scenario are: • Other children’s health and developmental effects for which there were not adequate data to develop a dose-response curve, and thus a benefits estimate. These outcomes include academic achievement, disturbances in mood, behavior, and social conduct, and hearing loss deficits in attention, reduced ability to inhibit inappropriate responding; impulsivity, distractibility, reactivity to the environment, social behavior, and auditory function. Recent research suggest that IQ loss is most strongly associated with concurrent blood lead levels and that this relationship is stronger in older children. • Benefits that accrue to adults, including avoided cases of hypertension, coronary heart disease (CHD), stroke, and death. • Adverse effects on plants and animals, and • Instead of Wwillingness-to-pay values, – values based on lost income are used instead. 28 For the empirical model runs, benefits for exterior events are estimated using the IEUBK model. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 33 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5- 9 Annual Number of RRP Events; Number of Cases of Exposures Avoided by Children under 6 and Adults due to the Proposed Regulations. Number of Cases of Exposure AvoidedIndividuals Protected (thousands)* Number of Events (thousands) Option First Year First Year with LSWP Second Year Second Year with LSWP Option A Option B Option C Option D Option E 580 343 217 343 397 399 243 155 243 287 578 578 578 578 668 140 140 140 140 177 Children Adults First Year Second Year First Year Second Year 920 633 409 633 693 916 916 916 916 987 225 166 112 166 207 224 224 224 224 273 * Number is increment above those occupying units where LSWP are currently practiced in the baseline and assume 75% compliance rate. 2 3 4 5 6 Table 5- 10: Annual Number of Exposures Avoided in Children Under the Age of 6 Due to the Rule, by Interior/Exterior Event and COF Type. Schools with PreTarget Daycare Schools with Kindergarten/ Housing Total Center Kindergarten Kindergarten COF Interior Events 295,804 136,338 246,772 61,433 740,348 Exterior Events 49,429 22,782 41,236 62,298 175,744 Total 345,233 159,121 288,008 123,731 916,092 7 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 34 5/17/2007 DRAFT—DELIBERATIVE 1 2 Table 5-11a: Total Number of IQ Points Gained (in thousands) from Interior and Exterior Events – Scenario 1 Exposure Option A Year 1 Year 2 Options B & D Year 1 Year 2 Option C Year 1 Year 2 Option E Year 1 Year 2 Mean Mean Mean Mean Mean Mean Mean 5 4 8 9 6 5 10 12 3 3 6 7 6 5 10 12 8 6 14 16 10 7 17 19 11 11 20 20 25 20 30 25 Mean Child Exposure - IQ Points Gained (thousands) - Interior and Exterior Children, age 1-2 years Using IEUBK model Peak 6 6 For Interior Estimates 6-Year Average 5 5 Using Empirical model Peak 11 10 For Interior Estimates 6-Year Average 12 12 Children, age 0-5 years Using IEUBK model Peak 21 20 16 20 For Interior Estimates 6-Year Average 21 20 16 20 Based on post-rule compliance rate of 75%. Option E = Option B with DIY. All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. 3 4 5 6 7 8 9 10 11 12 13 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 35 5/17/2007 1 2 3 4 5 6 7 8 9 10 DRAFT—DELIBERATIVE Table 5-11b: Total Number of IQ Points Gained (in thousands) from Interior and Exterior Events – Scenario 2 Option A Options B & D Option C Option E Year 1 Year 2 Year 1 Year 2 Year 1 Year 2 Year 1 Year 2 Exposure Mean Mean Mean Mean Mean Mean Mean Mean 5 4 8 9 6 5 10 11 3 3 6 6 6 5 10 11 8 5 14 15 9 7 16 18 10 10 19 19 23 20 28 24 Child Exposure - IQ Points Gained (thousands) - Interior and Exterior Children, age 1-2 years Using IEUBK model Peak 6 6 For Interior Estimates 6-Year Average 5 5 Using Empirical model Peak 10 10 For Interior Estimates 6-Year Average 12 11 Children, age 0-5 years Using IEUBK model Peak 20 19 15 19 For Interior Estimates 6-Year Average 20 19 15 19 Based on post-rule compliance rate of 75%. All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 36 5/17/2007 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 DRAFT—DELIBERATIVE Table 5-12a: Total Quantified Benefits of Avoided Exposures (Children Only) from Interior and Exterior Events – Scenario 1 ($ millions) Exposure Option A Year 1 Year 2 Options B & D Year 1 Year 2 Option C Year 1 Year 2 Option E Year 1 Year 2 Mean Mean Mean Mean Mean Mean $45 $38 $75 $85 $78 $67 $134 $151 $107 $74 $185 $204 $128 $89 $224 $246 $144 $138 $263 $259 $322 $265 $394 $328 Mean Mean Child Exposure - Interior and Exterior Children, age 1-2 years Using IEUBK Model Peak $81 $78 $64 $78 for Interior Estimates 6-Year Average $69 $67 $55 $67 Using Empirical Model Peak $139 $134 $108 $134 for Interior Estimates 6-Year Average $157 $151 $123 $151 Children, age 0-5 years Using IEUBK Model Peak $272 $263 $210 $263 for Interior Estimates 6-Year Average $267 $259 $205 $259 Based on post-rule compliance rate of 75%. Option E = Option B with DIY included. All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 37 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 Table 5-12b: Total Quantified Benefits of Avoided Exposures (Children Only) from Interior and Exterior Events – Scenario 2 ($ millions) Option A Options B & D Option C Option E Year 1 Year 2 Year 1 Year 2 Year 1 Year 2 Year 1 Year 2 Exposure Mean Mean Mean Child Exposure - Interior and Exterior Children, age 1-2 years Using IEUBK Model Peak $76 $74 $61 for Interior Estimates 6-Year Average $68 $65 $54 Using Empirical Model Peak $133 $128 $104 for Interior Estimates 6-Year Average $151 $146 $119 Children, age 0-5 years Using IEUBK Model Peak $258 $249 $199 for Interior Estimates 6-Year Average $260 $252 $200 Based on post-rule compliance rate of 75%. Option E = Option B with DIY included. All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. Mean Mean Mean Mean Mean $74 $65 $128 $146 $42 $37 $72 $82 $74 $65 $128 $146 $100 $71 $176 $196 $120 $86 $213 $236 $249 $252 $136 $135 $249 $252 $302 $254 $368 $315 7 8 9 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 38 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5-13a: Total Quantified of Avoided Exposures (Children Only)Benefits for Child Exposure from Interior and Exterior Events – Scenario 1 ($ millions) Option A Year 1 Year 2 Percentiles Exposure Options B & D Median 5 th Year 1 Percentiles 95 th Median 5 th 95 Year 2 Percentiles th Option C Median 5 th 95 Year 1 Percentiles th Median 5 th th 95 Year 2 Percentiles Median 5 th th 95 Percentiles Median 5th 95th Child Exposure - Interior and Exterior Children, age 1-2 years $2 $3 $5 $4 $267 $204 $422 $487 $18 $27 $43 $38 $3 $4 $6 $5 $10 $15 $25 $22 $2 $2 $4 $3 $188 $141 $288 $337 $18 $27 $43 $38 $3 $4 $6 $5 $321 $247 $542 $608 $12 $1,039 $69 $12 $1,004 $54 $9 $17 $931 $109 $17 $900 $13 $86 Based on post-rule compliance rate of 75%. All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. $819 $725 $69 $109 $12 $1,004 $36 $17 $900 $57 $6 $9 $570 $495 $69 $109 $12 $17 $1,004 $900 Using IEUBK Model Using Empirical Model Peak 6-Year Average Peak 6-Year Average $19 $28 $45 $39 $3 $4 $7 $5 $332 $255 $560 $628 $18 $27 $43 $38 $3 $4 $6 $5 $321 $247 $542 $608 $15 $22 $36 $31 $321 $247 $542 $608 Children, age 0-5 years Using IEUBK Model Peak 6-Year Average $71 $113 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 39 5/17/2007 DRAFT—DELIBERATIVE 1 2 Table 5-13Table 5-13b: Total Quantified Benefits of Avoided Exposures (Children Only)for Child Exposure from Interior and Exterior Events – Scenario 2 ($ millions) Option A Year 1 Options B & D Year 2 Percentiles Year 1 Percentiles Median 5 95 Median 5 $3 $4 $4 $4 $314 $248 $542 $607 $18 $27 $38 $35 $3 $4 $4 $4 $253 $198 $409 $471 $18 $27 $38 $35 $3 $4 $4 $4 $304 $240 $524 $587 $10 $15 $22 $20 $2 $2 $2 $2 $71 $12 $975 $69 $12 $943 $54 $9 $769 Peak $109 $17 $873 $86 $13 $702 6-Year Average $113 $17 $903 Based on post-rule compliance rate of 75%. Scenario 2 assumes additional contractor and COF cleaning in the baseline, as described in Section 5.2.4. $69 $109 $12 $17 $943 $873 $36 $57 $6 $9 th th 95 th th th th th Year 2 Percentiles 5 th Median 5 Year 1 Percentiles Median Median 95 Year 2 Percentiles 95 Exposure 5 Option C th Percentiles Median 5th 95th $178 $137 $280 $326 $18 $27 $38 $35 $3 $4 $4 $4 $304 $240 $524 $587 $535 $479 $69 $109 $12 $17 $943 $873 95 th Child Exposure - Interior and Exterior Children, age 1-2 years Using IEUBK Model Peak 6-Year Average Using Empirical Model Peak 6-Year Average $19 $28 $40 $36 $304 $240 $524 $587 $15 $22 $32 $29 $2 $3 $3 $3 Children, age 0-5 years Using IEUBK Model All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. 3 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 40 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 Table 5-14a: Total Number of IQ Points (in thousands) Gained per Year – Scenario 1 Option A Options B & D Year 1 Year 2 Year 1 Year 2 Exposure Mean Mean Int.* Ext.** Child Exposure - IQ Points Gained (thousands) Children, age 1-2 years Using IEUBK model Peak 4 3 For Interior Estimates 6-Year Average 2 4 Using Empirical model Peak 8 3 For Interior Estimates 6-Year Average 8 4 Int. Mean Ext. Int. Option C Year 1 Mean Ext. Int. Mean Ext. Int. 2 3 2 3 4 1 8 8 2 4 2 4 11 6 9 14 9 5 7 11 11 6 9 14 Int. Mean Ext. Int. Mean Ext. Int. Ext. 4 1 8 8 2 4 2 4 6 3 12 13 2 3 2 3 7 3 15 15 3 4 3 4 6 3 * = Interior, ** = Exterior. Based on post rule compliance rate of 75%. Option E = Option B with TH DIY. All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. 5 7 11 6 9 14 18 9 7 11 21 11 9 14 9 15 3 1 6 6 Mean Ext. Year 2 1 2 1 2 12 6 2 4 2 4 Year 1 2 1 4 5 Children, age 0-5 years Using IEUBK model Peak For Interior Estimates 6-Year Average 4 1 8 8 Option E Year 2 5 6 7 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 41 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5-14b: Total Number of IQ Points Gained (in thousands) per Year – Scenario 2 Option A Options B & D Year 1 Year 2 Year 1 Year 2 Exposure Mean Mean Int.* Ext.** Child Exposure - IQ Points Gained (thousands) Children, age 1-2 years Using IEUBK model Peak 3 3 For Interior Estimates 6-Year Average 1 4 Using Empirical model Peak 8 3 For Interior Estimates 6-Year Average 8 4 Int. Mean Ext. Int. Option C Mean Ext. Int. Ext. Year 1 Year 2 Year 1 Year 2 Mean Mean Mean Mean Int. 2 3 2 3 3 1 7 8 2 4 2 4 10 5 9 14 8 4 7 11 10 5 9 14 Ext. Int. Ext. Int. Ext. 3 1 7 8 2 4 2 4 6 2 12 12 2 3 2 3 7 3 14 14 3 4 3 4 6 3 * = Interior, ** = Exterior. Based on post rule compliance rate of 75%. Option E = Option B with TH DIY. All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. 5 7 10 5 9 14 16 8 7 11 19 10 9 14 9 15 3 1 6 6 Int. 1 2 1 2 11 6 2 4 2 4 Ext. 2 1 4 4 Children, age 0-5 years Using IEUBK model Peak For Interior Estimates 6-Year Average 3 1 7 8 Option E 2 3 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 42 5/17/2007 DRAFT—DELIBERATIVE 1 2 Table 5-15a: Total Quantified Benefits of Avoided Exposures (Children Only) – Scenario 1 ($ millions) Option A Options B & D Year 1 Year 2 Year 1 Year 2 Exposure Mean Mean Mean Mean Option C Option E Year 1 Year 2 Year 1 Year 2 Mean Mean Mean Mean Int.* Ext.** Int. Ext. Int. Ext. Int. Ext. Int. Ext. Int. Ext. $48 $20 $106 $107 $33 $50 $33 $50 $46 $19 $103 $103 $32 $48 $32 $48 $38 $16 $83 $84 $26 $39 $26 $39 $46 $19 $103 $103 $32 $48 $32 $48 $27 $11 $58 $59 $18 $27 $18 $27 $46 $19 $103 $103 $32 $48 $32 $48 $80 $33 $159 $163 $27 $41 $27 $41 $95 $39 $191 $196 $33 $50 $33 $50 Using IEUBK model Peak $154 $118 $148 $114 $119 $90 $148 $114 For Interior Estimates 6-Year Average $79 $189 $76 $183 $61 $144 $76 $183 * = Interior, ** = Exterior. Based on post rule compliance rate of 75%. Option E = Option B with TH DIY. All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. $83 $42 $60 $96 $148 $76 $114 $183 $229 $116 $93 $149 $276 $140 $117 $188 Child Exposure Children, age 1-2 years Using IEUBK model Peak For Interior Estimates 6-Year Average Using Empirical model Peak For Interior Estimates 6-Year Average Children, age 0-5 years Int. Ext. Int. Ext. 3 4 5 6 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 43 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 Table 5-15b: Total Quantified Benefits of Avoided Exposures (Children Only) – Scenario 2 ($ millions) Option A Options B & D Year 1 Year 2 Year 1 Year 2 Exposure Child Exposure Children, age 1-2 years Using IEUBK model Peak For Interior Estimates 6-Year Average Using Empirical model Peak For Interior Estimates 6-Year Average Mean Mean Mean Mean Option C Option E Year 1 Year 2 Year 1 Year 2 Mean Mean Mean Mean Int.* Ext.** Int. Ext. Int. Ext. Int. Ext. Int. Ext. Int. Ext. $44 $18 $100 $102 $33 $50 $33 $50 $42 $17 $97 $98 $32 $48 $32 $48 $35 $14 $78 $80 $26 $39 $26 $39 $42 $17 $97 $98 $32 $48 $32 $48 $25 $10 $55 $56 $18 $27 $18 $27 $42 $17 $97 $98 $32 $48 $32 $48 $73 $30 $150 $155 $27 $41 $27 $41 $87 $36 $180 $186 $33 $50 $33 $50 $76 $39 $60 $96 $135 $69 $114 $183 $208 $106 $93 $149 $251 $127 $117 $188 Children, age 0-5 years Using IEUBK model Peak $139 $118 $135 $114 $108 $90 $135 $114 For Interior Estimates 6-Year Average $72 $189 $69 $183 $55 $144 $69 $183 * = Interior, ** = Exterior. Based on post rule compliance rate of 75%. Option E = Option B with TH DIY. All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. Int. Ext. Int. Ext. 7 8 9 10 11 12 13 14 15 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 44 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 Table 5-16a: Total Annualized Mean 50-Year Benefits of Avoided Exposures (Children Only) from Interior and Exterior Events – Scenario 1 ($ millions) Option A 3% 7% Exposure Mean Mean Options B & D 3% 7% Mean Mean Children, age 1-2 years Using IEUBK Model Peak $77 $82 $76 $81 for Interior Estimates 6-Year Average $66 $71 $66 $70 Using Empirical Model Peak $133 $141 $132 $139 for Interior Estimates 6-Year Average $150 $159 $148 $157 Children, age 0-5 years Using IEUBK Model Peak $260 $276 $257 $272 for Interior Estimates 6-Year Average $256 $272 $253 $268 Based on post-rule compliance rate of 75%. Option E = Option B with DIY included. All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 Option C 3% 7% Option E 3% 7% Mean Mean Mean Mean $76 $65 $130 $147 $79 $68 $137 $154 $126 $88 $220 $241 $133 $93 $232 $255 $255 $251 $267 $263 $386 $321 $408 $339 45 5/17/2007 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 DRAFT—DELIBERATIVE Table 5-16b: Total Annualized Mean 50-Year Benefits of Avoided Exposures (Children Only) from Interior and Exterior Events – Scenario 2 ($ millions) Option A 3% 7% Exposure Mean Mean Options B & D 3% 7% Mean Mean Children, age 1-2 years Using IEUBK Model Peak $73 $78 $72 $77 for Interior Estimates 6-Year Average $65 $69 $64 $68 Using Empirical Model Peak $127 $135 $126 $133 for Interior Estimates 6-Year Average $145 $154 $143 $152 Children, age 0-5 years Using IEUBK Model Peak $246 $262 $244 $258 for Interior Estimates 6-Year Average $249 $265 $247 $261 Based on post-rule compliance rate of 75%. Option E = Option B with DIY included. All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 Option C 3% 7% Option E 3% 7% Mean Mean Mean Mean $72 $64 $125 $142 $75 $67 $131 $149 $117 $84 $209 $232 $124 $89 $221 $245 $241 $244 $253 $256 $361 $309 $382 $326 46 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 Table 5-17a: Total 50- Year and 50-Year Annualized Quantified Benefits for Avoided Child Exposure (Interior & Exterior Events) – Scenario 1 (3 Percent Discount Rate, $ millions) Option A Total 50-Year Benefit Option B Annualized Benefit Percentiles Total 50-Year Benefit Percentiles Annualized Benefit Total 50-Year Benefit Percentiles Percentiles Percentiles Median 5 $8,098 $6,231 $13,640 $15,312 $18 $27 $42 $37 $3 $4 $6 $5 $1,753 $303 $25,547 $68 $12 $993 $1,736 $300 $25,327 Peak 6-Year Average $2,776 $424 $22,905 $108 $16 $890 $2,749 $419 $22,698 Based on post-rule compliance rate of 75%. Scenario 2 assumes additional contractor and COF cleaning in the baseline, as described in Section 5.2.4. $67 $107 Exposure Median th 5 95 th Median 5 th th 95 Median 5 Option C th 95 th Percentiles Median 5 95th $18 $26 $42 $37 $3 $4 $6 $5 $312 $240 $525 $589 $12 $984 $1,718 $297 $25,078 $67 $16 $882 $2,720 $415 $22,468 $106 $12 $16 $975 $873 th th 95 Median Annualized Benefit 5 th 95 th th Child Exposure - Interior and Exterior Children, age 1-2 years Using IEUBK Model Using Empirical Model $466 $76 Peak 6-Year Average $689 $98 $1,096 $162 Peak 6-Year Average $966 $121 $8,163 $6,283 $13,778 $15,454 $18 $27 $43 $38 $3 $4 $6 $5 $317 $462 $75 $244 $683 $97 $535 $1,088 $161 $601 $958 $120 $315 $458 $75 $242 $676 $96 $530 $1,077 $160 $595 $948 $119 $8,019 $6,169 $13,506 $15,162 Children, age 0-5 years Using IEUBK Model All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. 7 8 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 47 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5-17Table 5-17b: Total 50- Year and 50-Year Annualized Quantified Benefits for Avoided Child Exposure (Interior and Exterior Events) – Scenario 2 (3 Percent Discount Rate, $ millions) Option A Total 50-Year Benefit Percentiles Exposure Median th 5 th 95 Option B Annualized Benefit Total 50-Year Benefit Percentiles Median 5 $18 $27 $38 $35 $3 $4 $4 $4 th 95 th Percentiles Median th 5 th 95 Option C Annualized Benefit Percentiles Median 5th 95th Total 50-Year Benefit Percentiles Median 5 th 95 th Annualized Benefit Percentiles Median 5th 95th Child Exposure - Interior and Exterior Children, age 1-2 years $18 $26 $38 $34 $3 $4 $4 $4 $298 $235 $513 $575 $3 $4 $4 $4 $295 $233 $508 $569 $1,745 $301 $23,989 $68 $12 $932 $1,728 $298 $23,783 $67 Peak 6-Year Average $2,772 $423 $22,202 $108 $16 $863 $2,745 $419 $22,002 $107 Based on post-rule compliance rate of 75%. Scenario 2 assumes additional contractor and COF cleaning in the baseline, as described in Section 5.2.4. $12 $16 $924 $1,710 $295 $23,549 $66 $11 $855 $2,716 $414 $21,779 $106 $16 $915 $846 Using Indoor IEUBK Using Indoor Empirical Peak 6-Year Average Peak 6-Year Average $463 $75 $7,729 $687 $98 $6,103 $975 $99 $13,330 $895 $104 $14,928 $300 $237 $518 $580 $459 $75 $681 $97 $967 $98 $888 $103 $7,667 $6,052 $13,197 $14,792 $454 $674 $958 $879 $74 $7,593 $96 $5,992 $97 $13,068 $102 $14,647 $18 $26 $37 $34 Children, age 0-5 years Using Indoor IEUBK All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. 2 3 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 48 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5-18a: Total 50- Year and 50-Year Annualized Quantified Benefits for Avoided Child Exposure (Interior and Exterior Events) – Scenario 1 (7 Percent Discount Rate, $ millions) Option A Total 50-Year Benefit Percentiles Exposure Median Annualized Benefit 5 95 Media n $43 $56 $93 $69 $4,660 $3,587 $7,866 $8,822 $19 $28 $45 $40 th th Option B Total 50-Year Benefit Percentiles th th 5 95 $3 $4 $7 $5 $338 $260 $570 $639 Annualized Benefit Total 50-Year Benefit Percentiles Percentiles Percentiles Median 5 th th 95 Option C Median 5 Median 5 $333 $256 $560 $629 $257 $380 $606 $534 $42 $54 $90 $67 Percentiles Median 5 $19 $28 $44 $39 $3 $4 $7 $5 $327 $252 $550 $618 $12 $1,041 $966 $167 $14,116 $70 $17 $933 $1,529 $233 $12,640 $111 $12 $17 $1,023 $916 th 95 Annualized Benefit th th 95 th th 95th Child Exposure - Interior and Exterior Children, age 1-2 years Using Indoor IEUBK Peak 6-Year Average Using Indoor Empirical Peak 6-Year Average $266 $393 $626 $551 $262 $387 $617 $543 $43 $55 $91 $68 $4,595 $3,535 $7,727 $8,681 $19 $28 $45 $39 $3 $4 $7 $5 $4,516 $3,473 $7,594 $8,531 Children, age 0-5 years $1,001 $173 $14,584 $73 $13 $1,057 $984 $170 $14,364 $71 Peak 6-Year Average $1,585 $242 $13,076 $115 $18 $947 $1,558 $238 $12,870 $113 Based on post-rule compliance rate of 75%. Scenario 2 assumes additional contractor and COF cleaning in the baseline, as described in Section 5.2.4. Using Indoor IEUBK All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. 2 3 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 49 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5-18Table 5-18b: Total 50- Year and 50-Year Annualized Quantified Benefits for Avoided Child Exposure (Interior and Exterior Events) – Scenario 2 (7 Percent Discount Rate, $ millions) Option A Total 50-Year Benefit Percentiles Option B Annualized Benefit Total 50-Year Benefit Percentiles Median 5th 95th Median Percentiles Option C Annualized Benefit Total 50-Year Benefit Percentiles Median 5th 95th Median Annualized Benefit Percentiles Percentiles 5th 95th Median 5th 95th $42 $54 $55 $57 $4,276 $3,373 $7,348 $8,241 $19 $27 $39 $36 $3 $4 $4 $4 $310 $244 $532 $597 $996 $172 $13,695 $72 $12 $992 $979 $169 $13,489 $71 $12 $977 $961 $166 $13,255 Peak 6-Year Average $1,582 $241 $12,675 $115 $17 $918 $1,555 $237 $12,475 $113 $17 $904 $1,526 $233 $12,251 Based on post-rule compliance rate of 75%. Scenario 2 assumes additional contractor and COF cleaning in the baseline, as described in Section 5.2.4. $70 $111 $12 $17 $960 $888 Exposure Median 5th 95th $43 $56 $57 $59 $4,412 $3,484 $7,610 $8,522 5th 95th $42 $55 $56 $59 $4,351 $3,434 $7,477 $8,386 Child Exposure - Interior and Exterior Children, age 1-2 years Using Indoor IEUBK Using Indoor Empirical Peak 6-Year Average Peak 6-Year Average $264 $392 $557 $511 $19 $28 $40 $37 $3 $4 $4 $4 $320 $252 $551 $617 $260 $386 $549 $503 $19 $28 $40 $36 $3 $4 $4 $4 $315 $249 $542 $608 $256 $379 $539 $494 Children, age 0-5 years Using Indoor IEUBK All estimates of benefits due to avoided exposure from exterior events based on IEUBK model. 2 3 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 50 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5-19: Summary of the Range of Annualized Benefits, by Scenario and Option, using 3 percent and 7 percent discount rates over 50 years. Children’s IQ Benefits – Annualized (millions 2005$) Scenario 1 Scenario 2 Lowest Model Est. Highest Model Est. Lowest Model Est. Highest Model Est. Annualized using 3 Percent Discount Rate Option A $66 $4 - $244 $260 $12 - $993 $65 $4 - $237 $249 $16 - $863 Options B & D $66 $4 - $242 $257 $12 - $984 $64 $4 - $235 $247 $16 - $855 Option C $65 $4 - $240 $255 $12 - $975 $64 $4 - $233 $244 $16 - $846 Option E $88 $386 $84 $4 - $354 $12 - $1684 $4 - $340 Annualized using 7 Percent Discount Rate $361 $17 - $1184 Option A $71 $4 - $260 $276 $13 - $1057 $69 $4 - $252 $265 $17 - $918 Options B & D $70 $4 - $256 $272 $12 - $1041 $68 $4 - $249 $261 $17 - $904 Option C $68 $4 - $252 $267 $12 - $1023 $67 $4 - $244 $256 $17 - $888 Option E $93 $408 $89 $382 $4 - $375 $13 - $1782 $4 - $360 $18 - $1252 The top line of numbers in each cell represents the range of mean values, reflecting the lowest and highest estimates across the six alternative models for blood lead, exposure estimates and population of children. The bottom line of numbers in each cell, in italics, represents the 5th and 95th percentile values associated with that mean value. 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 51 5/17/2007 DRAFT—DELIBERATIVE 1 2 Table 5-20: Summary of the Range of Annualized Benefits for Option B by Scenario and Type of COF, using 3 percent and 7 percent discount rates over 50 years. Children’s IQ Benefits – Annualized (millions 2005$) Scenario 1 Lowest Model Highest Model Est. Est. Annualized using 3 Percent Discount Rate Scenario 2 Lowest Highest Model Model Est. Est. Public or Commercial Building COFs $17 $1 - $48 $97 $7 - $241 $16 $1 - $48 $115 $10 - $322 Target Housing COFs $49 $2 - $194 $160 $4 - $744 $48 $2 - $187 $131 $6 - $533 $64 $4 - $235 $247 $16 - $855 $66 $257 $4 - $242 $12 - $984 Annualized using 7 Percent Discount Rate Total Public or Commercial Building COFs $18 $1 - $51 $102 $8 - $254 $17 $1 - $50 $122 $11 - $340 Target Housing COFs $52 $2 - $205 $170 $5 - $787 $51 $2 - $198 $139 $6 - $564 $70 $272 $68 $261 $4 - $256 $12 - $1041 $4 - $249 $17 - $904 The top line of numbers in each cell represents the range of mean values, reflecting the lowest and highest estimates across the six alternative models for blood lead, exposure estimates and population of children. The bottom line of numbers in each cell, in italics, represents the 5th and 95th percentile values associated with that mean value. Total 3 4 5 6 7 Table 5-21:Estimated benefits for interior and exterior events by type of COF and blood-lead modeling assumptions Annualized Benefits - $ million (3 percent), Option B: IEUBK Model, 0-5 year-olds, Peak Daycare Center Schools with Kindergarten Schools with PreKindergarten/ Kindergarten Target Housing COF Total Interior Events $22 $5 $11 $107 $145 Exterior Events $41 $6 $12 $53 $112 Total $63 $11 $23 $160 $257 Annualized Benefits- $ million (3 percent), Option B: Empirical Model, 1-year olds, Peak (Exterior from IEUBK) §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 52 5/17/2007 DRAFT—DELIBERATIVE Daycare Center Schools with Kindergarten Schools with PreKindergarten/ Kindergarten Target Housing COF Total Interior Events $19 $0 $7 $75 $101 Exterior Events $8 $0 $1 $21 $31 $27 $0 $8 $97 $132 Total Annualized Benefits- $ million (3 percent), Option B: Empirical Model, 1-year olds, Average (Exterior from IEUBK) Daycare Center Schools with Kindergarten Schools with PreKindergarten/ Kindergarten Target Housing COF $79 Total Interior Events $17 $0 $6 $101 Exterior Events $12 $0 $2 $33 $47 Total $29 $0 $8 $111 $148 Annualized Benefits- $ million (3 percent), Option B: IEUBK Model, 0-5 year-olds, Average Daycare Center Schools with Kindergarten Schools with PreKindergarten/ Kindergarten Target Housing COF Total Interior Events $12 $3 $6 $54 $74 Exterior Events $67 $9 $20 $82 $179 Total $79 $12 $26 $136 $253 Annualized Benefits - $ million (3 percent), Option B: IEUBK Model, 1-year olds, Peak Daycare Center Schools with Kindergarten Schools with PreKindergarten/ Kindergarten Target Housing COF Total Interior Events $4 $0 $1 $40 $45 Exterior Events $8 $0 $1 $21 $31 $12 $0 $3 $62 $76 Total Annualized Benefits- $ million (3 percent), Option B: IEUBK Model, 1-year olds, Average Schools with Schools with PreTarget Housing Daycare Center Kindergarten Kindergarten/ Kindergarten COF Total Interior Events $2 $0 $0 $17 $19 Exterior Events $12 $0 $2 $33 $47 Total $14 $0 $3 $49 $66 1 2 3 4 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 53 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Appendix 5A Lead-Related Health Effects and Ecological Effects Lead exposure can cause many adverse health and ecological effects. The quantitative benefits estimates in Chapter 5 are based only on the value of reduced lifetime earnings due to IQ loss from exposures to children under the age of six. This appendix supplements the benefits chapter by providing a broader, qualitative discussion of lead-related effects (including adult effects and ecological effects that are not included in the quantitative benefits estimates), based on EPA’s Air Quality Criteria for Lead. The information provided in this Appendix is an excerpt from the Executive Summary of the document Air Quality Criteria for Lead (United States Environmental Protection Agency, October 2006, EPA/600/R-5/144aF, this document is available at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=158823). Specifically, the information provided in this Appendix is directly from the following sections of the Executive Summary: E.4 Health Effects Associated with Lead Exposure E.5 Human Population Groups at Special Risk and Potential Public Health Impacts E.6 Environmental Effects of Lead Background The purpose of the 2006 Lead Air Quality Criteria document (AQCD) is to critically assess the latest scientific information on lead. The final version of the revised Lead AQCD mainly assesses pertinent literature published or accepted for publication through December 2005. The First External Review Draft (dated December 2005) of the revised Lead AQCD underwent public comment and was reviewed by the Clean Air Scientific Advisory Committee (CASAC) at a public meeting held in Durham, NC on February 28-March 1, 2006. The public comments and CASAC recommendations received were taken into account in making appropriate revisions and incorporating them into a Second External Review Draft (dated May, 2006) which was released for further public comment and CASAC review at a public meeting held June 28-29, 2006. In addition, still further revised drafts of the Integrative Synthesis chapter and the Executive Summary were then issued and discussed during an August 15, 2006 CASAC teleconference call. Public comments and CASAC advice received on these latter materials, as well as Second External Review Draft materials, were taken into account in making and incorporating further revisions into this final version of the Lead AQCD. Health Effects Associated With Lead Exposure Both epidemiologic and toxicologic studies have shown that environmentally relevant levels of lead affect many different organ systems. Research completed since the 1986 AQCD/Addendum and 1990 Supplement indicates that lead effects occur at blood-lead levels even lower than those previously reported for many endpoints. Remarkable progress has been made since the mid-1980s in understanding the Pb effects on health. Recent studies have focused on details of the associations, including the shapes of concentration-response relationships, especially at levels well within the range of general population exposures, and on those biological and/or socioenvironmental factors that either increase or decrease an individual’s risk. Key findings and conclusions regarding important outcomes of newly available toxicologic and epidemiologic studies of Pb health effects are highlighted below. Neurotoxic Effects of Pb Exposure • Neurobehavioral effects of Pb-exposure early in development (during fetal, neonatal, and later §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 54 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 postnatal periods) in young infants and children (<7 years old) have been observed with remarkable consistency across numerous studies involving varying study designs, different developmental assessment protocols, and diverse populations. Negative Pb impacts on neurocognitive ability and other neurobehavioral outcomes are robust in most recent studies even after adjustment for numerous potentially confounding factors (including quality of care giving, parental intelligence, and socioeconomic status). These effects generally appear to persist into adolescence and young adulthood. • The overall weight of the available evidence provides clear substantiation of neurocognitive decrements being associated in young children with blood-Pb concentrations in the range of 5-10 µg/dL, and possibly somewhat lower. Some newly available analyses appear to show Pb effects on the intellectual attainment of preschool and school age children at population mean concurrent blood-Pb levels ranging down to as low as 2 to 8 µg/dL. A decline of 6.2 points in full scale IQ for an increase in concurrent blood Pb levels from 1 to 10 µg/dL has been estimated, based on a pooled analysis of results derived from seven well-conducted prospective epidemiologic studies. • In the limited literature examining the effects of environmental Pb exposure on adults, mixed evidence exists regarding associations between Pb and neurocognitive performance. No associations were observed between cognitive performance and blood Pb levels; however, significant associations were observed in relation to bone Pb concentrations, suggesting that longterm cumulative Pb exposure may contribute to neurocognitive deficits in adults. • Animal toxicology data indicate that developmental Pb exposures creating steady-state blood-Pb concentrations of ~10 µg/dL result in behavioral impairments that persist into adulthood in rats and monkeys. No evident threshold has yet been found; and Pb-induced deficits, for the most part, have been found to be very persistent, even with various chelation treatments. However, experimental studies indicate that environmental enrichment during development can partially mitigate the effects of Pb on cognitive function. In rats, neurobehavioral deficits that persisted well into adulthood were observed with prenatal, preweaning, and postweaning Pb exposure. In monkeys, such neurobehavioral deficits were observed both with in utero-only exposure and with early postnatal-only exposure when peak blood-Pb levels did not exceed 15 µg/dL and steadystate levels were ~11 µg/dL. • Learning impairment has been observed in animal studies at blood levels as low as 10 µg/dL, with higher level learning showing greater impairment than simple learning tasks. The mechanisms associated with these deficits include: response perseveration; insensitivity to changes in reinforcement density or contingencies; deficits in attention; reduced ability to inhibit inappropriate responding; impulsivity; and distractibility. • Lead affects reactivity to the environment and social behavior in both rodents and nonhuman primates at blood Pb levels of 15 to 40 µg/dL. Rodent studies also show that Pb exposure potentiates the effects of stress in females. • Auditory function has also been shown to be impaired at blood Pb levels of 33 µg/dL, while visual functions are affected at 19 µg/dL. • Neurotoxicological studies in animals clearly demonstrated that Pb mimics calcium and affects neurotransmission and synaptic plasticity. • Epidemiologic studies have identified genetic polymorphisms of two genes that may alter §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 55 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 susceptibility to the neurodevelopmental consequences of Pb exposure in children. Variant alleles of the ALAD gene are associated with differences in absorption, retention, and toxicokinetics of Pb. Polymorphisms of the vitamin D receptor gene have been shown to affect the rate of resorption and excretion of Pb over time. These studies are only suggestive, and parallel animal studies have not been completed. Cardiovascular Effects of Lead • Epidemiologic studies have consistently demonstrated associations between Pb exposure and enhanced risk of deleterious cardiovascular outcomes, including increased blood pressure and incidence of hypertension. A meta-analysis of numerous studies estimates that a doubling of blood-Pb level (e.g., from 5 to 10 µg/dL) is associated with ~1.0 mm Hg increase in systolic blood pressure and ~0.6 mm Hg increase in diastolic pressure. Studies have also found that cumulative past Pb exposure (e.g., bone Pb) may be as important, if not more, than present Pb exposure in assessing cardiovascular effects. The evidence for an association of Pb with cardiovascular morbidity and mortality is limited but supportive. • Experimental toxicology studies have confirmed Pb effects on cardiovascular functions. Most have shown that exposures creating blood-Pb levels of ~20 to 30 µg/dL for long periods result in arterial hypertension that persists long after cessation of Pb exposure in genetically normal animals. One study reported blood pressure increases at blood-Pb levels as low as 2 µg/dL in rats. A number of in vivo and in vitro studies provide compelling evidence for the role of oxidative stress in the pathogenesis of Pb-induced hypertension. However, experimental investigations of cardiovascular effects of Pb in animals are unclear as to why low, but not high, levels of Pb exposure cause hypertension in experimental animals. Renal Effects of Lead • In the general population, both circulating and cumulative Pb was found to be associated with longitudinal decline in renal function. Effects on creatine clearance have been reported in human adult hypertensives to be associated with general population mean blood-Pb levels of only 4.2 µg/dL. The public health significance of such effects is not clear, however, in view of more serious signs of kidney dysfunction being seen in occupationally exposed workers only at much higher blood-Pb levels (>30-40 µg/dL). • Experimental studies using laboratory animals demonstrated that the initial accumulation of absorbed Pb occurs primarily in the kidneys. This takes place mainly through glomerular filtration and subsequent reabsorption, and, to a small extent, through direct absorption from the blood. Both low dose Pb-treated animals and high dose Pb-treated animals showed a “hyperfiltration” phenomenon during the first 3 months of Pb exposure. Investigations into biochemical alterations in Pb-induced renal toxicity suggested a role for oxidative stress and involvement of NO, with a significant increase in nitrotyrosine and substantial fall in urinary excretion of NOx. • Iron deficiency increases intestinal absorption of Pb and the Pb content of soft tissues and bone. Aluminum decreases kidney Pb content and serum creatinine in Pb-intoxicated animals. Age also has an effect on Pb retention. There is higher Pb retention at a very young age and lower bone and kidney Pb at old age, attributed in part to increased bone resorption and decreased bone accretion and, also, kidney Pb. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 56 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Effects of Lead on the Immune System • Findings from recent epidemiologic studies suggest that Pb exposure may be associated with effects on cellular and humoral immunity. These include changes in serum immunoglobulin levels. Studies of biomarkers of humoral immunity in children have consistently found significant associations between increasing blood-Pb concentrations and serum IgE levels at blood-Pb levels <10 µg/dL. • Toxicologic studies have shown that Pb targets immune cells, causing suppression of delayed type hypersensitivity response, elevation of IgE, and modulation of macrophages into a hyperinflammatory phenotype. These types of changes can cause increased risk of atopy, asthma, and some forms of autoimmunity and reduced resistance to some infectious diseases. Lead exposure of embryos resulting in blood-Pb levels <10 µg/dL can produce persistent later-life immunotoxicity. Effects of Lead on Heme Synthesis • Lead exposure has been associated with disruption of heme synthesis in both children and adults. A 10% probability of anemia (hematocrit <35%) is estimated to be associated with a blood-Pb level of ~20 µg/dL at age 1 year. Increases in blood Pb concentration of about 20-30 µg/dL are sufficient to halve erythrocyte ALAD activity and sufficiently inhibit ferrochelatase to double erythrocyte protoporphyrin levels. • Toxicological studies demonstrated that Pb intoxication interferes with red blood cell (RBC) survival and alters RBC mobility. Hematological parameters, such as mean corpuscular volume, mean corpuscular hemoglobin, and mean corpuscular hemoglobin concentration, are also significantly decreased upon exposure to Pb. These effects are due to internalization of Pb by RBC. The transport of Pb across the RBC membrane is energy-independent and carrier-mediated; and the uptake of Pb appears to be mediated by an anion exchanger through a vanadate-sensitive pathway. • Erythrocyte ALAD activity ratio (ratio of activated/non activated enzyme activity) has been shown to be a sensitive, dose-responsive measure of Pb exposure, regardless of the mode of administration of Pb. Competitive enzyme kinetic analyses in RBCs from both humans and Cynomolgus monkeys indicated similar inhibition profiles by Pb. Effects of Lead on Bones and Teeth • Experimental studies in animals demonstrate that Pb substitutes for calcium and is readily taken up and stored in the bone and teeth of animals, potentially allowing bone cell function to be compromised both directly and indirectly by exposure. Relatively short term exposure of mature animals to Pb does not result in significant growth suppression. However, chronic Pb exposure during times of inadequate nutrition has been shown to adversely influence bone growth, including decreased bone density, decreased trabecular bone, and growth plates. • Exposure of developing animals to Pb during gestation and the immediate postnatal period has clearly been shown to significantly depress early bone growth in a dose-dependent fashion, though this effect is not manifest below a certain threshold. • Systemically, Pb has been shown to disrupt mineralization of bone during growth, to alter §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 57 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 calcium binding proteins, and to increase calcium and phosphorus concentration in the blood stream, in addition to potentially altering bone cell differentiation and function by altering plasma levels of growth hormone and calciotropic hormones such as vitamin D3 [1,25- (OH2)D3. • Periods of extensive bone remodeling, such as occur during weight loss, advanced age, altered metabolic state, and pregnancy and lactation are all associated with mobilization of Pb stores from bone of animals. • Numerous epidemiologic studies and, separately, animal studies (both post-eruptive Pb exposure and pre- and perinatal Pb exposure studies) suggest that Pb is a caries-promoting element. However, whether Pb incorporation into the enamel surface compromises the integrity and resistance of the surface to dissolution, and ultimately increases risk of dental decay, is unclear. • Increased risk of dental caries has been associated with Pb exposure in children and adults. Lead effects on caries were observed in populations whose mean blood-Pb levels were less than 10 µg/dL. Reproductive and Developmental Effects of Lead • Epidemiologic evidence suggests small associations between Pb exposure and male reproductive outcomes, including perturbed semen quality and increased time to pregnancy. There are no adequate epidemiologic data to evaluate associations between Pb exposure and female fertility. Most studies have yielded no associations, or weak associations, of Pb exposure with thyroid hormone status and male reproductive endocrine status in highly exposed occupational populations. • New toxicologic studies support earlier conclusions, presented in the 1986 Lead AQCD, that (a) Pb can produce both temporary and persisting effects on male and female reproductive function and development and (b) Pb disrupts endocrine function at multiple points along the hypothalamic-pituitary-gonadal axis. Although there is evidence for a common mode of action, consistent effects on circulating testosterone levels are not always observed in Pb-exposed animals. Inconsistencies in reports of circulating testosterone levels complicate derivation of a dose-response relationship for this endpoint. • Lead-induced testicular damage (ultrastructural changes in testes of monkeys at blood-Pb >35 to 40 µg/dL) and altered female sex hormone release, imprinting during early development, and altered female fertility all suggest Pb-induced reproductive effects. However, Pb exposure does not generally produce total sterility. Pre- and postnatal exposure to Pb has been demonstrated to result in fetal mortality and produce a variety of sublethal effects in the offspring. Many of the Pb-induced sublethal developmental effects occur at maternal blood-Pb levels that do not result in clinical (overt) toxicity in the mothers. Teratogenic effects resulting from Pb exposure reported in a few studies appear to be confounded by maternal toxicity. Lead Effects on Other Organ Systems • Lead impacts the hypothalamic-pituitary-adrenal axis, elevating corticosterone levels and altering stress responsivity. This may be a potential mechanism contributing to Pb-induced hypertension, with further possible roles in the etiology of diabetes, obesity and other disorders. • Studies of hepatic enzyme levels in serum suggest that liver injury may be present in Pb workers; §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 58 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 however, associations specifically with Pb exposures are not evident. Children exposed to relatively high levels of Pb (blood Pb >30 µg/dL) exhibit depressed levels of circulating 1,25dihydroxy vitamin D (1,25-OH-D). However, associations between serum vitamin D status and blood Pb were not evident in a study of calcium-replete children who had average lifetime bloodPb concentrations <25 µg/dL. • Field studies that evaluated hepatic enzyme levels in serum suggest that liver injury may be present in Pb workers; however, associations specifically with Pb exposures have not been well established. • Simultaneous induction of the activities of phase II drug metabolizing enzymes and decreased phase I enzymes with a single exposure to Pb nitrate in rat liver suggest that Pb is capable of causing biochemical phenotype similar to hepatic nodules. • Newer studies examined the induction of GST-P at both transcriptional and translational levels using in vitro systems and indicated a role for Pb-nitrate and Pb-acetate in the induction process. • Lead-induced alterations in cholesterol metabolism appear to be mediated by the induction of several enzymes related to cholesterol metabolism and the decrease of 7 α-hydroxylase, a cholesterol catabolizing enzyme. This regulation of cholesterol homeostasis is modulated by changes in cytokine expression and related signaling. • Newer experimental evidence suggests that Pb-induced alterations in liver heme metabolism involve perturbations in ALAD activity, porphyrin metabolism, alterations in Transferrin gene expression, and associated changes in iron metabolism. • Gastrointestinal (GI) absorption of Pb is influenced by a variety of factors, including chemical and physical forms of the element in ingested media, age at intake, and various nutritional factors. The degeneration of intestinal mucosal epithelium leading to potential malabsorption and alterations in the jejunal ultrastructure (possibly associated with distortion of glycocalyx layer) have been reported in the intestine of Pb-exposed rats. • Nutritional studies that varied Pb, Ca, and vitamin D levels in the diet have demonstrated competition of Pb with Ca absorption. Supplementation with vitamin D has been reported to enhance intestinal absorption of Ca and Pb. Physiological amounts of vitamin D, when administered to vitamin D-deficient rats, resulted in elevated Pb and Ca levels. In the case of severe Ca deficiency, Pb ingestion results in a marked decrease in serum 1,25 hydroxy vitamin D. Genotoxic and Carcinogenic Effects of Lead • Epidemiologic studies of highly exposed occupational populations suggest a relationship between Pb and cancers of the lung and the stomach; however the evidence is limited by the presence of various potential confounders, including metal coexposures (e.g., to arsenic, cadmium), smoking, and dietary habits. The 2003 NTP and 2004 IARC reviews concluded that Pb and Pb compounds were probable carcinogens, based on limited evidence in humans and sufficient evidence in animals. Similarly, Pb and Pb compounds would likely be classified as likely to be carcinogenic to humans according to the new 2005 EPA Cancer Assessment Guidelines for Carcinogen Risk Assessment, based on animal data even though the human data are inadequate. • Studies of genotoxicity consistently find associations of Pb exposure with DNA damage and §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 59 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 micronuclei formation; however, the associations with the more established indicator of cancer risk, chromosomal aberrations, are inconsistent. • Pb is an animal carcinogen and extends our understanding of mechanisms involved to include a role for metallothionein. Specifically, the recent data show that metallothionein may participate in Pb inclusion bodies and, thus, serves to prevent or reduce Pb-induced tumorigenesis. • In vitro cell culture studies that evaluated the potential for Pb to transform rodent cells are inconsistent, and careful study of a time course of exposure is necessary to determine whether Pb actually induces transformation in cultured rodent cells. There is increased evidence suggesting that Pb may be co-carcinogenic or promotes the carcinogenicity of other compounds. Cell culture studies do support a possible epigenetic mechanism or co-mutagenic effects. Lead-Binding Proteins • Proteins depending upon sulfur-containing side chains for maintaining conformity or activity are vulnerable to inactivation by Pb, due to its strong sulfur-binding affinity. • The enzyme, ALAD, a 280 kDa protein, is inducible and is the major Pb-binding protein within the erythrocyte. • The Pb-binding protein in rat kidney has been identified as a cleavage product of α-2microglobulin. The low molecular weight Pb-binding proteins in human kidney have been identified as thymosin β 4 (molecular weight 5 kDa) and acyl-CoA binding protein (molecular weight 9 kDa). In human brain, Pb-binding proteins include thymosin β4 and an unidentified protein of 23 kDa. • Animal toxicology studies with metallothionein-null mice demonstrated a possible role for metallothionein as a renal Pb-binding protein. Human Population Groups At Special Risk And Potential Public Health Impacts • Children, in general and especially low SES (often including larger proportions of AfricanAmerican and Hispanic) children, have been well-documented as being at increased risk for Pb exposure and Pb-induced adverse health effects. This is due to several factors, including enhanced exposure to Pb via ingestion of soil-Pb and/or dust-Pb due to normal hand-to-mouth activity and/or pica. • Even children with low Pb exposure levels (having blood Pb of 5-10 µg/dL or, possibly, somewhat lower) are at notable risk, due to apparent non-linear dose-response relationships between blood Pb and neurodevelopmental outcomes. It is hypothesized that initial neurodevelopmental lesions occurring at blood-Pb levels <10 µg/dL may disrupt different developmental processes in the nervous system than more severe high level exposures. • Adults with idiosyncratic exposures to Pb through occupations, hobbies, make-up use, glazed pottery, native medicines, and other sources are at risk for Pb toxicity. Certain ethnic and racial groups are known to have cultural practices that involve ingestion of Pb-containing substances, e.g., ingestion of foods or beverages stored in Pb-glazed pottery or imported canned food from countries that allow Pb-soldered cans. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 60 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 • Cumulative past Pb exposure, measured by bone Pb, may be a better predictor of cardiovascular effects than current blood-Pb levels. African-Americans are known to have substantially higher baseline blood pressure than other ethnic groups, so Pb’s impact on an already higher baseline could indicate a greater susceptibility to Pb for this group. • Effects on adults of low-level Pb exposures also include some renal effects (i.e., altered creatinine clearance) at blood-Pb levels <5 ug/dL. Lead exposure combined with other risk factors, such as diabetes, hypertension, or chronic renal insufficiency may result in clinically relevant effects in individuals with two or more other risk factors. • At least two genetic polymorphisms, of the ALAD and the vitamin D receptor gene, have been suggested to play a role in susceptibility to Pb. In one study, African-American children were found to have a higher incidence of being homozygous for alleles of the vitamin D receptor gene thought to contribute to greater Pb blood levels. This work is preliminary and further studies will be necessary to determine implications of genetic differences that may make certain populations more susceptible to Pb exposure. • What was considered “low” for Pb exposure levels in the 1980s is an order of magnitude higher than the current mean level in the U.S. population, and current average blood-Pb levels in U.S. populations remain perhaps as much as two orders of magnitude above preindustrial “natural” levels in humans. There is no level of Pb exposure that has yet been identified, with confidence, as being clearly not being associated with possible risk of deleterious health effects. Some recent studies of Pb neurotoxicity in infants have observed effects at population average blood-Pb levels of only 1 or 2 µg/dL; and some cardiovascular, renal, and immune outcomes have been reported at blood-Pb levels below 5 µg/dL. • Public health interventions have resulted in declines, over the last 25 years, of more than 90% in the mean blood-Pb level within all age and gender subgroups of the U.S. population, substantially decreasing the numbers of individuals at likely risk for Pb-induced toxicities. Nevertheless, estimates of the magnitude of potential public health impacts of Pb exposure can be substantial for the U.S. population. For example, in estimating the effect of Pb exposure on intelligence, it was projected that the fraction of individuals with an IQ >120 would decrease from ~9% with no Pb exposure to less than 3% at a blood-Pb level of 10 µg/dL. Also, the fraction of individuals with an IQ >130 points was estimated as being likely to decrease from 2.25% to 0.5% with a blood-Pb level change from 0 to 10 µg/dL. In addition, an estimate of hypertension-related risk for serious cardiovascular events (coronary disease, stroke, peripheral artery disease, cardiac failure) indicates that a decrease in blood Pb from 10 to 5 µg/dL could result in an annual decrease of 27 events per 100,000 women and 39 events per 100,000 men. Environmental Effects Of Lead Terrestrial Ecosystems Methodologies Used in Terrestrial Ecosystem Research • Electron probe microanalysis (EPMA) techniques provide the greatest information on metal speciation. Other techniques, such as EXAFS (extended X-ray absorption fine structure) and EXANES (extended X-ray absorption near edge spectroscopy), show great promise and will be important in solving key mechanistic questions. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 61 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 • In situ methodologies have been developed to lower soil-Pb relative bioavailability. These amendments typically fall within the categories of phosphate, biosolid, and Al/Fe/Mn-oxide amendments. Some of the drawbacks to soil amendment include phosphate toxicity to plants and increased arsenic mobility at high soil phosphate concentrations. The use of iron(III) phosphate seems to mitigate arsenic mobility, however increased concentrations of phosphate and iron limit their application when drinking water quality is a concern. Distribution of Atmospherically Delivered Lead in Terrestrial Ecosystems • Total Pb deposition during the 20th century has been estimated at 1 to 3 g Pb m-2, depending on elevation and proximity to urban areas. Total contemporary loadings to terrestrial ecosystems are ~1 to 2 mg m-2 year-1. This is a relatively small annual flux of Pb compared to the reservoir of ~0.5 to 4 g m-2 of gasoline additive-derived Pb already deposited in surface soils over much of the United States. • Dry deposition can account for 10% to >90% of total Pb deposition. Because Clean Air Act Legislation has preferentially reduced Pb associated with fine particles, relative contributions of dry deposition have changed in the last few decades. • Although inputs of Pb to ecosystems are currently low, Pb export from watersheds via groundwater and streams is substantially lower than inputs. Therefore, even at current input levels, watersheds are accumulating anthropogenic Pb. • Species of Pb delivered to terrestrial ecosystems can be inferred by emission source. For example, Pb species emitted from automobile exhaust are dominated by particulate Pb halides and double salts with ammonium halides (e.g., PbBrCl, PbBrCl2NH4Cl), while Pb emitted from smelters is dominated by Pb-sulfur species. Halides from automobile exhaust break down rapidly in the atmosphere, via redox reactions in the presence of atmospheric acids. Lead phases in the atmosphere, and presumably the compounds delivered to the surface of the earth (i.e., to vegetation and soils), are suspected to be in the form of PbSO4, PbS, and PbO. • The importance of humic and fulvic acids and hydrous Mn- and Fe-oxides for scavenging Pb in soils was discussed in some detail in the 1986 Lead AQCD. The importance of these Pb binding substrates is reinforced by studies reported in the more contemporary literature. • The amount of Pb that has leached into mineral soil appears to be on the order of 20 to 50% of the total anthropogenic Pb deposition. • The vertical distribution and mobility of atmospheric Pb in soils was poorly documented prior to 1986. Techniques using radiogenic Pb isotopes have been developed to differentiate between gasoline-derived Pb and natural, geogenic (native) Pb. These techniques provide more accurate determinations of the depth-distribution and potential migration velocities for atmospherically delivered Pb in soils. • Selective chemical extractions have been used extensively over the past 20 years to quantify amounts of a particular metal phase in soil or sediment rather than total metal concentration. However, some problems persist with the selective extraction technique: (a) extractions are rarely specific to a single phase; and (b) in addition to the nonselectivity of reagents, significant metal redistribution has been found to occur during sequential chemical extractions. Thus, although chemical extractions provide some useful information on metal phases in soil or sediment, the §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 62 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 results should be treated as “operationally defined,” e.g., “H2O2-liberated Pb” rather than “organic Pb.” • Soil solution dissolved organic matter content and pH typically have very strong positive and negative correlations, respectively, with the concentration of dissolved Pb species. Effects of Lead on Natural Terrestrial Ecosystems • Atmospheric Pb pollution has resulted in the accumulation of Pb in terrestrial ecosystems throughout the world. In the United States, anthropogenically-derived Pb represents a significant fraction of the total Pb burden in soils, even in sites remote from smelters and other industrial plants. However, few significant effects of Pb pollution have been observed at sites that are not near point sources of Pb. • Evidence from precipitation collection and sediment analyses indicates that atmospheric deposition of Pb has declined dramatically (>95%) at sites unaffected by point sources of Pb, and there is little evidence that Pb accumulated in soils at these sites represents a threat to ground water or surface water supplies. • The effects of Pb and other chemical emissions on terrestrial ecosystems near smelters and other industrial sites decrease downwind from the Pb source. Several studies using the soil burden as an indicator have shown that much of the contamination occurs within a radius of 20 to 50 km around the emission source. Elevated metal concentrations around smelters have been found to persist despite significant reductions in emissions. The concentrations of Pb in soils, vegetation, and fauna at these sites can be two to three orders of magnitude higher than in reference areas. Assessing the risks specifically associated with Pb is difficult, because these sites also experience elevated concentrations of other metals and because of effects related to SO2 emissions. The confounding effect of other pollutants makes the assessment of Pb-specific exposure-response relationships impossible at the whole ecosystem level. • In the most extreme cases, near smelter sites, the death of vegetation causes a near-complete collapse of the detrital food web, creating a terrestrial ecosystem in which energy and nutrient flows are minimal. • More commonly, stress in soil microorganisms and detritivores can cause reductions in the rate of decomposition of detrital organic matter. Although there is little evidence of significant bioaccumulation of Pb in natural terrestrial ecosystems, reductions in microbial and detritivorous populations can affect the success of their predators. Thus, at present, industrial point sources represent the greatest Pb-related threat to the maintenance of sustainable, healthy, diverse, and high-functioning terrestrial ecosystems in the United States. Terrestrial Species Response/Mode of Action • Plants take up Pb via their foliage and through their root systems. Surface deposition of Pb onto plants may represent a significant contribution to the total Pb in and on the plant, as has been observed for plants near smelters and along roadsides. • There are two possible mechanisms (symplastic or apoplastic) by which Pb may enter the root of a plant. The symplastic route is through the cell membranes of root hairs; this is the mechanism of uptake for water and nutrients. The apoplastic route is an extracellular route between epidermal §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 63 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 cells into the intercellular spaces of the root cortex. The symplastic route is considered the primary mechanism of Pb uptake in plants. • Recent work supports previous conclusions that the form of metal tested, and its speciation in soil, influence uptake and toxicity to plants and invertebrates. The oxide form of Pb is less toxic than the chloride or acetate forms, which are less toxic that the nitrate form of Pb. However, these results must be interpreted with caution, as the counter ion (e.g., the nitrate ion) may also be contributing to the observed toxicity. • Lead may be detoxified in plants by deposition in root cell walls, and this may be influenced by calcium concentrations. Other hypotheses put forward recently include the presence of sulfur ligands and the sequestration of Pb in old leaves as detoxification mechanisms. Lead detoxification has not been studied extensively in invertebrates. Glutathione detoxification enzymes were measured in two species of spider. Lead may be stored in waste nodules in earthworms or as pyromorphite in the nematode. • Lead effects on heme synthesis (as measured primarily by ALAD activity and protoporphyrin concentration) were documented in the 1986 Lead AQCD and continue to be studied. However, researchers caution that changes in ALAD and other enzyme parameters are not always related to adverse effects, but may simply indicate exposure. Other effects on plasma enzymes, which may damage other organs, have been reported. Lead also may cause lipid peroxidation, which may be alleviated by vitamin E, although Pb poisoning may still result. Changes in fatty acid production have been reported, which may influence immune response and bone formation. • Insectivorous mammals may be more exposed to Pb than herbivores, and higher trophic-level consumers may be less exposed than lower trophic-level organisms. Nutritionally-deficient diets (including low calcium) cause increased uptake of Pb and greater toxicity in birds. • Interactions of Pb with other metals are inconsistent, depending on the endpoint measured, the tissue analyzed, the animal species, and the metal combination. Exposure/Response of Terrestrial Species • Recent critical advancements reported in the current Lead AQCD in understanding toxicity levels relies heavily on the work completed by a multi-stakeholder group, consisting of federal, state, consulting, industry, and academic participants, led by the EPA to develop Ecological Soil Screening Levels (Eco-SSLs). • Eco-SSLs are concentrations of contaminants in soils that would result in little or no measurable effect on ecological receptors. The Eco-SSLs are intentionally conservative in order to provide confidence that contaminants which could present an unacceptable risk are not screened out early in the evaluation process. That is, at or below these levels, adverse effects are considered unlikely. Due to conservative modeling assumptions (e.g., metal exists in most toxic form or highly bioavailable form, high food ingestion rate, high soil ingestion rate) which are common to screening processes, several Eco-SSLs are derived below the average background soil concentration for a particular contaminant. • The Eco-SSLs for terrestrial plants, birds, mammals, and soil invertebrates are 120, 11, 56, and 1700 mg Pb/kg soil, respectively. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 64 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Aquatic Ecosystems Methodologies Used in Aquatic Ecosystem Research • Many of the terrestrial methods can also be applied to suspended solids and sediments collected from aquatic ecosystems. Just as in the terrestrial environment, the speciation of Pb and other trace metals in natural freshwaters and seawater plays a crucial role in determining their reactivity, mobility, bioavailability, and toxicity. Many of the same speciation techniques employed for the speciation of Pb in terrestrial ecosystems are applicable in aquatic ecosystems. • There is now a better understanding of the potential effects of sampling, sample handling, and sample preparation on aqueous-phase metal speciation. Thus, a need has arisen for dynamic analytical techniques that are able to capture a metal's speciation, in-situ and in real time. • With few exceptions, ambient water quality criteria (AWQC) are derived based on data from aquatic toxicity studies conducted in the laboratory. In general, both acute (short term) and chronic (long term) AWQCs are developed. Depending on the species, the toxicity studies considered for developing acute criteria range in length from 48 to 96 hours. • Acceptable chronic toxicity studies should encompass the full life cycle of the test organism, although for fish, early life stage or partial life cycle toxicity studies are considered acceptable. Acceptable endpoints include reproduction, growth and development, and survival, with the effect levels expressed as the chronic value. • The biotic ligand model (BLM), which considers the binding of free metal ion to the site of toxic action and competition between metal species and other ions, has been developed to predict the toxicity of several metals under a variety of water quality conditions. However, there are limitations to this tool in deriving AWQC because, currently, limited work has been conducted in developing chronic BLMs (for any metals, let alone Pb) and the acute BLMs to date do not account for dietary metal exposures. Distribution of Lead in Aquatic Ecosystems • Atmospheric Pb is delivered to aquatic ecosystems primarily through deposition (wet and/or dry) or through erosional transport of soil particles. • A significant portion of Pb in the aquatic environment exists in the undissolved form (i.e., bound to suspended particulate matter). The ratio of Pb in suspended solids to Pb in filtrate varies from 4:1 in rural streams to 27:1 in urban streams. • The oxidation potential of Pb is high in slightly acidic solutions, and Pb2+ binds with high affinity to sulfur-, oxygen-, and nitrogen-containing ligands. Therefore, speciation of Pb in the aquatic environment is controlled by many factors (e.g., pH, redox, dissolved organic carbon, sulfides). The primary form of Pb in aquatic environments is divalent (Pb2+), while Pb4+ exists only under extreme oxidizing conditions. Labile forms of Pb (e.g., Pb2+, PbOH+, PbCO3) are a significant portion of the Pb inputs to aquatic systems from atmospheric washout. Lead is typically present in acidic aquatic environments as PbSO4, PbCl4, ionic Pb, cationic forms of Pb-hydroxide, and ordinary Pb-hydroxide (Pb(OH)2). In alkaline waters, common species of Pb include anionic forms of Pb-carbonate (Pb(CO3)) and Pb(OH)2. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 65 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 • Lead concentrations in lakes and oceans were generally found to be much lower than those measured in the lotic waters assessed by NAWQA. In open waters of the North Atlantic the decline of Pb concentrations has been associated with the phasing out of leaded gasoline in North America and Western Europe. However, in estuarine systems, it appears that similar declines following the phase-out of leaded gasoline are not necessarily as rapid. • Based on a synthesis of NAWQA data from the United States, Pb concentrations in surface waters, sediments, and fish tissues (whole body) respectively range from: 0.04 to 30 µg/L (mean = 0.66, median = 0.50, 95th %tile = 1.1); 0.5 to 12,000 mg/kg (mean = 120, median = 28, 95th %tile = 200); and 0.08 to 23 mg/kg (mean = 1.03, median = 0.59, 95th %tile = 3.24). Effects of Lead on Natural Aquatic Ecosystems • Lead exposure may adversely affect organisms at different levels of organization, i.e., individual organisms, populations, communities, or ecosystems. Generally, however, there is insufficient information available for single materials in controlled studies to permit evaluation of specific impacts on higher levels of organization (beyond the individual organism). Potential effects at the population level or higher are, of necessity, extrapolated from individual level studies. Available population, community, or ecosystem level studies are typically conducted at sites that have been contaminated or adversely affected by multiple stressors (several chemicals alone or combined with physical or biological stressors). Therefore, the best documented links between Pb and effects on the environment are with effects on individual organisms. • Natural systems frequently contain multiple metals, making it difficult to attribute observed adverse effects to single metals. For example, macroinvertebrate communities have been widely studied with respect to metals contamination and community composition and species richness. In these studies, multiple metals were evaluated and correlations between observed community level effects were ascertained. The results often indicate a correlation between the presence of one or more metals (or total metals) and the negative effects observed. While, correlation may imply a relationship between two variables, it does not imply causation of effects. • In simulated microcosms or natural systems, environmental exposure to Pb in water and sediment has been shown to affect energy flow and nutrient cycling and benthic community structure. • In field studies, Pb contamination has been shown to significantly alter the aquatic environment through bioaccumulation and alterations of community structure and function. • Exposure to Pb in laboratory studies and simulated ecosystems may alter species competitive behaviors, predator-prey interactions, and contaminant avoidance behaviors. Alteration of these interactions may have negative effects on species abundance and community structure. • In natural aquatic ecosystems, Pb is often found coexisting with other metals and other stressors. Thus, understanding the effects of Pb in natural systems is challenging given that observed effects may be due to cumulative toxicity from multiple stressors. Aquatic Species Response/Mode of Action • Recent research has suggested that due to the low solubility of Pb in water, dietary Pb (i.e., Pb adsorbed to sediment, particulate matter, and food) may contribute substantially to exposure and toxicity in aquatic biota. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 66 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 • Generally speaking, aquatic organisms exhibit three Pb accumulation strategies: (1) accumulation of significant Pb concentrations with a low rate of loss, (2) excretion of Pb roughly in balance with availability of metal in the environment, and (3) weak net accumulation due to very low metal uptake rate and no significant excretion. • Protists and plants produce intracellular polypeptides that form complexes with Pb. Macrophytes and wetland plants that thrive in Pb-contaminated regions have developed translocation strategies for tolerance and detoxification. • Like aquatic plants and protists, aquatic animals detoxify Pb by preventing it from being metabolically available, though their mechanisms for doing so vary. Invertebrates use lysosomalvacuolar systems to sequester and process Pb within glandular cells. They also accumulate Pb as deposits on and within skeletal tissue, and some can efficiently excrete Pb. Fish scales and mucous chelate Pb in the water column, and potentially reduce visceral exposure. • Numerous studies have reported the effects of Pb exposure on blood chemistry in aquatic biota. Plasma cholesterol, blood serum protein, albumin, and globulin concentrations were identified as bioindicators of Pb stress in fish. • Nutrients affect Pb toxicity in aquatic organisms. Some nutrients seem capable of reducing toxicity. Exposure to Pb has not been shown to reduce nutrient uptake ability, though it has been demonstrated that Pb exposure may lead to increased production and loss of organic material (e.g., mucus and other complex organic ligands). • Avoidance responses are actions performed to evade a perceived threat. Some aquatic organisms have been shown to be quite adept at avoiding Pb in aquatic systems, while others seem incapable of detecting its presence. • The two most commonly reported Pb-element interactions are between Pb and calcium and between Pb and zinc. Both calcium and zinc are essential elements in organisms and the interaction of Pb with these ions can lead to adverse effects both by increased Pb uptake and by a decrease in Ca and Zn required for normal metabolic functions. Exposure/Response of Aquatic Species • The 1986 Lead AQCD reviewed data in the context of sublethal effects of Pb exposure. The document focused on describing the types and ranges of Pb exposures in ecosystems likely to adversely impact domestic animals. As such, the 1986 AQCD did not provide a comprehensive analysis of the effects of Pb to most aquatic primary producers, consumers, and decomposers. • Waterborne Pb is highly toxic to aquatic organisms, with toxicity varying with the species and life stage tested, duration of exposure, form of Pb tested, and water quality characteristics. • Among the species tested, aquatic invertebrates, such as amphipods and water fleas, were the most sensitive to the effects of Pb, with adverse effects being reported at concentrations as low as 0.45 µg/L (range: 0.45 to 8000 µg/L). • Freshwater fish demonstrated adverse effects at concentrations ranging from 10 to >5400 µg/L, depending generally upon water quality parameters. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 67 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 • Amphibians tend to be relatively Pb tolerant; however, they may exhibit decreased enzyme activity (e.g., ALAD reduction) and changes in behavior (e.g., hypoxia response behavior). Critical Loads for Lead in Terrestrial and Aquatic Ecosystems • Critical loads are defined as threshold deposition rates of air pollutants that current knowledge indicates will not cause long-term adverse effects to ecosystem structure and function. A critical load is related to an ecosystem's sensitivity to anthropogenic inputs of a specific chemical. • The critical loads approach for sensitive ecosystems from acidification has been in use throughout Europe for about 20 years. Its application to Pb and other heavy metals in Europe is more recent. European critical load values for Pb have been developed but are highly specific to the bedrock geology, soil types, vegetation, and historical deposition trends in each European country. To date, the critical loads framework has not been used for regulatory purposes in the United States for any chemical. Considerable research is necessary before critical load estimates can be formulated for ecosystems extant in the United States. • Speciation strongly influences the toxicity of Pb in soil and water and partitioning between dissolved and solid phases determines the concentration of Pb in soil drainage water, but it has not been taken into account in most of the critical load calculations for Pb performed to date. • Runoff of Pb from soil may be the major source of Pb into aquatic systems. However, little attempt has been made to include this source into critical load calculations for aquatic systems due to the complexity of including this source in the critical load models. In summary, due to the deposition of Pb from past practices (e.g., leaded gasoline, ore smelting) and the long residence time of Pb in many aquatic and terrestrial ecosystems, a legacy of environmental Pb burden exists, over which is superimposed much lower contemporary Pb loadings. The potential for ecological effects of the combined legacy and contemporary Pb burden to occur is a function of the bioavailability or bioaccessibility of the Pb, which, in turn, is highly dependent upon numerous site factors (e.g., soil organic carbon content, pH, water hardness). Moreover, while the more localized ecosystem impacts observed around smelters are often striking, these perturbations cannot be attributed solely to Pb. Many other stressors (e.g., other heavy metals, oxides of sulfur and nitrogen) can also act singly or in concert with Pb to cause such notable environmental impacts. This is a summary of the health and ecological effects from renovation, repair, and painting activities involving lead-based paint.Lead exposure can cause many adverse health and ecological effects. The quantitative benefits estimates in Chapter 5 are based only on the value of reduced lifetime earnings due to IQ loss from exposures to children under the age of six. This appendix supplements the benefits chapter by providing a broader, qualitative discussion of lead-related effects (including adult effects and ecological effects that are not included in the quantitative benefits estimates), starting with a discussion of EPA’s Air Quality Criteria for Lead. 5A.1 Review of Effects in EPA’s Lead Air Quality Criteria DocumentHealth Effects §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 68 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 National Ambient Air Quality Standards (NAAQS) are promulgated by the United States Environmental Protection Agency (EPA) to meet requirements set forth in Sections 108 and 109 of the U.S. Clean Air Act. Those two Clean Air Act sections require the EPA Administrator to (1) list widespread air pollutants that reasonably may be expected to endanger public health or welfare; (2) issue air quality criteria for them that assess the latest available scientific information on nature and effects of ambient exposure to them; (3) set “primary” NAAQS to protect human health with adequate margin of safety and to set “secondary” NAAQS to protect against welfare effects (e.g., effects on vegetation, ecosystems, visibility, climate, manmade materials, etc); and (4) periodically review and revise, as appropriate, the criteria and NAAQS for a given listed pollutant or class of pollutants. Lead (Pb) was first listed in the mid-1970’s as a “criteria air pollutant” requiring NAAQS regulation. The scientific information pertinent to Pb NAAQS development available at the time was assessed in the EPA document Air Quality Criteria for Lead; published in 1977. Based on the scientific assessments contained in that 1977 lead air quality criteria document (1977 Lead AQCD), EPA established a 1.5 µg/m3 (maximum quarterly calendar average) Pb NAAQS in 1978. To meet Clean Air Act requirements noted above for periodic review of criteria and NAAQS, new scientific information published since the 1977 Lead AQCD was later assessed in a revised Lead AQCD and Addendum published in 1986 and in a Supplement to the 1986 AQCD/Addendum published by EPA in 1990. A 1990 Lead Staff Paper, prepared by EPA’s Office of Air Quality Planning and Standards (OPQPS), drew upon key findings and conclusions from the 1986 Lead AQCD/Addendum and 1990 Supplement (as well as other OAQPS sponsored lead exposure/risk analyses) in posing options for the EPA Administrator to consider with regard to possible revision of the Pb NAAQS. However, EPA chose not to revise the Pb NAAQS at that time. Rather, as part of implementing a broad 1991 U.S. EPA Strategy for Reducing Lead Exposure, the Agency focused primarily on regulatory and remedial clean-up efforts to reduce Pb exposure from a variety of non-air sources that posed more extensive public health risks, as well as other actions to reduce air emissions. The purpose of the revised Lead AQCD is to critically assess the latest scientific information that has become available since the literature assessed in the 1986 Lead AQCD/Addendum and 1990 Supplement, with the main focus being on pertinent new information useful in evaluating health and environmental effects of ambient air lead exposures. This includes discussion in this document of information regarding: the nature, sources, distribution, measurement, and concentrations of lead in the environment; multimedia lead exposure (via air, food, water, etc.) and biokinetic modeling of contributions of such exposures to concentrations of lead in brain, kidney, and other tissues (e.g., blood and bone concentrations, as key indices of lead exposure).; characterization of lead health effects and associated exposure response relationships; and delineation of environmental (ecological) effects of lead. The final version of the revised Lead AQCD mainly assesses pertinent literature published or accepted for publication through December 2005. The First External Review Draft (dated December 2005) of the revised Lead AQCD underwent public comment and was reviewed by the Clean Air Scientific Advisory Committee (CASAC) at a public meeting held in Durham, NC on February 28-March 1, 2006. The public comments and CASAC recommendations received were taken into account in making appropriate revisions and incorporating them into a Second External Review Draft (dated May, 2006) which was released for further public comment and CASAC review at a public meeting held June 28-29, 2006. In addition, still further revised §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 69 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 drafts of the Integrative Synthesis chapter and the Executive Summary were then issued and discussed during an August 15, 2006 CASAC teleconference call. Public comments and CASAC advice received on these latter materials, as well as Second External Review Draft materials, were taken into account in making and incorporating further revisions into the final version of this Lead AQCD, which was issued to meet an October 1, 2006 court-ordered deadline. Evaluations contained in the present document provide inputs to an associated Lead Staff Paper prepared by EPA’s Office of Air Quality Planning and Standards (OAQPS), which poses options for consideration by the EPA Administrator with regard to proposal and, ultimately, promulgation of decisions on potential retention or revision, as appropriate, of the current Pb NAAQS. The full text for the final Lead Criteria Document is available at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=158823. 5A.2Adult Health Effects The possible link between adverse health effects in adults and lead exposure has been widely studied. However, there is uncertainty about the level of exposure adults working in COF will experience. Thus the analysis did not attempt to estimate the number of cases that would be avoided due to the regulations under consideration. While adult exposure is highly uncertain it can be assumed that the temporal pattern might be similar to what children experience, which is a relatively short term (weeks to months) increase in blood lead levels. Epidemiological studies of adult health effects are typically conducted in occupational settings or in the general population. The general population studies reflect common environmental exposures and are more likely to be indicative of health effects that adults could experience as a result of renovation, repair, and painting activities. The following discussion focuses on cardiovascular and renal outcomes, which are the more likely adverse effects at relatively low levels of exposure, but other outcomes are discussed as well. 5A.2.1Cardiovascular Epidemiologic studies support the relationship between increased lead exposure and increased deleterious cardiovascular outcomes (U.S. EPA 2006). The evidence to date appears strongest for risk of increased blood pressure and increased incidence of hypertension. Studies in humans (U.S. EPA 2006) also have suggested that cumulative past lead exposure (as indexed by bone lead) may be as important, if not more, than present exposure in assessing cardiovascular effects. However, animal toxicologic studies have found that cell, tissue, and organ response to lead is immediate and may provide clues to the mechanisms by which lead contributes to cardiovascular disease in humans (U.S. EPA 2006). A recent meta-analysis, which included 19 general population and 12 occupational studies, indicated that for blood lead and blood pressure, every doubling of blood lead is associated with a ~1.0 mm Hg increased systolic and ~0.6 mm Hg increased diastolic blood pressure for blood lead between 1 and >40 µg/dL (Nawrot et al., 2002). The most recent review of this topic indicated there was sufficient evidence to infer a casual relationship of lead exposure with hypertension (Navas-Acien et al., 2007). However, further research is needed to determine: the precise dose–response relationship, the relative importance of short-term versus chronic lead effects, the relevant mechanisms at environmental levels of exposure, and whether the magnitude of the association is different in children or in other vulnerable population subgroups. lead exposure has also been associated with increased incidence of clinical cardiovascular outcomes such as coronary heart disease, stroke, and peripheral arterial disease. However, due to a limited number of studies, researchers §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 70 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 (Navas-Acien et al., 2007) concluded that the evidence is suggestive but not sufficient to infer a causal relationship for clinical cardiovascular endpoints in the general population. 5A.2.2Renal Several studies conducted in general population samples have reported an association between blood lead concentration and common biomarkers of renal function (serum creatinine and creatinine clearance) (Kosnett et al., 2007). Increased risk for nephrotoxicity has been observed at the lowest lead dose levels studied to date (U.S. EPA 2006). Specifically, blood lead ranged from 2.5 to 3.8 µg/dL in the first significant health effect category in Muntner et al. (2003), and associations between blood lead as a continuous variable and worse renal function were reported at a mean of 2.2 µg/dL (Akesson et al., 2005) (U.S. EPA, 2006). In addition, a large. NHANES III study (U.S. EPA, 2006), found alterations in urinary creatinine excretion rate (one indicator of possible renal dysfunction) in hypertensives at a mean blood lead of only 4.2 µg/dL. However, the data available to date are not sufficient to determine whether nephrotoxicity is related more to current blood lead levels, higher levels from past lead exposures, or both. In addition, while these general population studies are consistent with an adverse effect of low level lead exposure on renal function, the extent to which diminished renal function may itself result in increased body lead burden has not been fully elucidated (U.S. EPA, 2006). Overall, recent studies provide strong evidence that renal effects occur at much lower blood lead levels than previously recognized. At levels of exposure in the general U.S. population overall, lead combined with other risk factors, such as diabetes, hypertension, or chronic renal insufficiency from non-lead related causes, can result in clinically relevant effects (U.S. EPA 2006). The size of such susceptible populations is increasing in the United States due to obesity (U.S. EPA, 2006). In summary, renal effects may also be relevant to adults who experience RRP exposure while working in COFs—especially if those adults suffer from other kidney related risk factors. 5A.2.3Neurotoxic In the limited literature examining environmental lead exposure, mixed evidence exists regarding associations between lead and impaired cognitive performance in adults. Studies using concurrent blood lead levels as the marker for lead exposure (a relevant marker for adults experiencing COF RRP) found no association between cognitive performance and lead exposure. However, significant associations were seen in relation to bone lead concentrations, suggesting that long-term cumulative exposure may be crucial in contributing to neurocognitive deficits in adults (U.S. EPA 2006). Researchers recently report there is sufficient evidence to conclude that there is an association between lead dose and decrements in cognitive function in adults (Shih et al., 2007). Overall, while the association between blood lead levels and cognitive function is more pronounced in occupational groups with high current lead exposures, associations between bone lead levels and cognitive function are more evident in studies of older subjects with lower current blood lead levels, particularly in longitudinal studies of cognitive decline (Shih et al., 2007). 5A.2.4Reproductive and Developmental There are many potential reproductive and developmental endpoints that must be considered for men and women as well as the fetus. There are no adequate data to evaluate associations between lead exposure and female fertility. However, lead clearly crosses the placenta during all trimesters (U.S. EPA, 2006). §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 71 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Maternal exposure may result in fetal exposure and maternal blood lead levels in pregnancy have been associated with adverse impacts on postnatal neurobehavioral development (Kosnett et al., 2007). For many other outcomes (birth weight, fetal growth, preterm delivery, and congenital anomalies), the evidence suggest at most a small association with lead exposure (U.S. EPA, 2006). However, there may be populations that are highly susceptible to lead-related reproductive effects, especially if they have additional risk factors for these outcomes. With one exception, studies have not shown associations between spontaneous abortion and maternal or paternal lead exposure (U.S. EPA, 2006). The epidemiologic evidence suggests small associations between exposure to lead and male reproductive outcomes, including perturbed semen quality and increased time to pregnancy. However, these associations appear at blood lead levels >45 µg/dL, as most studies have only considered exposure in the occupational setting. 5A.2.5Other Health Endpoints Numerous other health endpoints and adverse responses have been identified in association with exposure to lead. These adverse effects include genotoxic, carcinogenic, immune, and hematopoietic systems. Overall, epidemiological evidence provides only very limited evidence suggestive of lead exposure associations with carcinogenic or genotoxic effects in humans (U.S. EPA 2006). The 2003 National Toxicology Program (NTP) and 2004 International Agency for Research on Cancer (IARC) reviews concluded that lead and lead compounds were probable carcinogens based on limited evidence in humans and sufficient evidence in animals (U.S EPA 2006). Several studies have examined possible associations between lead exposures and biomarkers of immune function. Findings from recent epidemiologic studies suggest that lead exposure may be associated with effects on cellular and humoral immunity (U.S. EPA 2006). lead exposure has also been associated with disruption of heme synthesis in both children and adults (U.S. EPA 2006). 5A.3 Ecological Effects When lead is released in an uncontrolled manner into the environment, whether indoors or outdoors, it ultimately becomes a part of the larger environment in which humans, animals and plants live. In the case of renovation work, the immediate concern is primarily with releases occurring outdoors due to exterior RRP events that do not follow the work practices specified in the proposed regulations. Interior releases, however, can eventually be washed, swept, or disposed of into outdoor environments. When lead-paint dust and debris are disposed of responsibly, the impact may be minimal. However, in the absence of the proposed regulations, it is anticipated that lead will ultimately enter the environment and thereby provide a source of exposure to aquatic or terrestrial receptors. Lead can affect biota in various ways. Plants are exposed to lead both through their foliage (through atmospheric deposits) and their roots. The extent to which plants uptake lead from the soil depends on various factors, including “cation exchange capacity, soil composition (e.g., organic matter component, calcium content)… [and] is favored at lower pH values and in soils with low organic carbon content” (U.S. EPA 2005b). In plants, exposure to lead can result in decreased photosynthesis, transpiration, water absorption and growth (U.S. EPA 1986, 2005b). Soil microorganisms (decomposers) appear to be more sensitive to lead than higher plants (U.S. EPA 1986). §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 72 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Avian and mammalian species are exposed to lead primarily through food. Exposure may take place through direct ingestion of soil or through the ingestion of lead-contaminated plants or prey. Effects of lead on avian and mammalian species include increased mortality, adverse effects on growth and reproduction, as well as developmental and behavioral changes. Effects on soil invertebrates include decreased growth and reproduction as well as increased mortality (U.S. EPA 1986). One benchmark indicator of the toxicity of lead to plants and animals is the Ecological Soil Screening Levels (Eco-SSLs) developed by the U.S. EPA’s Office of Emergency and Remedial Response. These contaminant concentration levels are intended for use in identifying contaminants of potential concern (COPCs) during the screening stage of an ecological risk assessment. Eco-SSLs are derived separately for plants, soil invertebrates, birds and mammals and represent conservative estimates of contaminant concentrations that may pose a risk to each group of species (U.S. EPA 2005b). Table 5A-1 presents current Eco-SSLs for lead (as of March 2005). Table 5A-1: Eco-SSLs for Lead (mg/kg dry weight in soil) Plants 120 Source: U.S. EPA 2005b 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Invertebrates 1,700 Wildlife Avian 11 Mammal 56 In addition to terrestrial species, lead is also highly toxic to aquatic life. The Agency has established both acute and chronic fresh and saltwater criteria for lead. Because the effects threshold increases with water hardness, these criteria vary depending upon the hardness of the water for a given area; the higher the hardness, the higher the criterion (U.S. EPA 1985). The impacts of lead on domestic animals have been well-documented. Lead-based paint is the most commonly reported source of lead exposure for household cats and dogs. According to Jill E. Madison, “Renovations of older houses involving sanding or scraping lead-based paint is believed to be the major origin of the lead-based paint in [paint ingestion] instances” (Madison 2005). Lead poisoning symptoms in dogs include GI abnormalities, such as anorexia or colic, and behavioral changes, including anxiety, salivation, blindness and muscular spasms (Merck & Co. 2003). Lead poisoning in dogs may also cause seizures (Thomas 2005). Cats are less likely to be exposed to lead because they are less likely to chew on lead-containing surfaces. Because of their grooming habits, however, they may ingest lead particles. Food avoidance is one of the major symptoms of lead poisoning in cats (Madison 2005). Similarly to dogs and cats, birds may also be exposed to lead through lead-based paint. According to Merck & Co., “in avian species, anorexia, ataxia, loss of condition, wing and leg weakness and anemia are the most notable signs” (Merck & Co. 2003). Although specific economic benefits to ecological components have not been quantified in this rule, it is reasonable to assume that proposed rule is likely to reduce the amount of lead released from renovation activities, thereby reducing the lead body burdens to plants and animals. 37 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 73 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Appendix 5B: Identifying and Characterizing Lead Loadings for Interior Renovation, Repair and Painting Tasks in Target Housing COFs 5B.1 Renovation, Repair, and Painting Tasks, Work Components, and Estimated Time Requirements Dust loading levels for RRP tasks can be estimated using the 1997 EPA Lead Exposure Associated with Renovation and Remodeling Activities: Environmental Field Sampling Study (EFSS) (U.S. EPA 1997a). The 1997 study measured lead levels in dust and air, primarily in the rooms where the renovation work was performed. The levels measured in the work area were extrapolated to adjacent rooms for this analysis because the lead dust levels in adjacent rooms provide an estimate for the average lead dust level in a home. In the 1997 study, lead loadings were quantified for specific work components (e.g., drilling, sawing, HVAC work) rather than for a task as defined by the AHS (e.g., replacing a sink). Consequently, it was necessary to match the work components described in the 1997 study to the renovation and remodeling tasks described in the AHS. Table 5B- 1Table 5B- 1 lists the work components for which lead loading data were provided in the renovation and remodeling study. The exhibit provides a brief description of the activity summarized from the renovation study (1997). The components are followed in parentheses by the units of measure (e.g., carpet one home, replace 3 windows, carry out one hour of work) used to evaluate lead loadings in the EPA study. These units are either time or activities. There were differences in the conditions under which different work components were carried out. Carpet removal and window replacement were each self-contained observational phases of the study, where only the types and locations of the samples were specified in the sampling design. Housing style, lead levels and other factors varied. The remainder of the work components was carried out in the “controlled, experimentally-designed (CED) phase,” with observations and measurements made for one hour of work in controlled environments. The type and conduct of the renovation activity, as well as the type and location of samples, were controlled (U.S. EPA 1997a, pg 8-45) in the CED phase. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 74 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5B- 1: Work Components Listed in the 1997 EPA Renovation Study with Lead Loading Dataa Activity Description Removal of 2 to 5 rooms of carpeting from 8 units (1 apartment, 1 duplex and 6 Carpet removal single-family homes) of various sizes (pgs. 8-6 and 8-7). The quantity of carpet A (1 house or removed ranged from 246 to 732 ft2 per unit and was not related to the size of the unit. apartment) The geometric mean (pg A-5) was used in this analysis. Removal and replacement of 3 windows each in 4 structures. Windows were selected Windows to represent all approaches to window replacement, all types of windows and the B (3 units) various lead levels that could be encountered in this type of work (pg 8-22). The geometric mean (pg. A-23) was used in this analysis. Paint In each unit, paint was removed in one area. Paint was removed in half of the area by C removal/sanding hand scraping and sanding and the other half by power sanding. No vacuum (time) attachments or dust reduction methods were used (pg 8-48). In each unit, three large structure removal activities were carried out involving the Demolition demolition of one or more walls, usually a single plaster wall. Removal involved D (time) removing the surface of the wall, and NOT the underlying structure, so that an exposed wall structure was available for installation of new drywall (pg 8-46). HVAC One HVAC repair/replacement activity was carried out in each unit, involving E (time) removal of several sections of the HVAC ductwork (pg 8-46). This work involved trimming wood from the edge of an existing door, sanding it Door modification F smooth and drilling a hole for installation of a doorknob, characterized as a small (time) surface disruption. The work involved primarily sawing and sanding (pgs 8-46-47). Power drilling G A lattice of holes was drilled into either plaster walls or wood doors or walls (pg 8-47) (time) Sawing A series of 15 parallel cuts, 5 feet in length, were made into either plaster walls or H (time) wood baseboards using either a rough or fine blade (pg 8-47). Component removal The following were removed in one unit: lead-painted wood trim, baseboards, doors, I (time) and door-jambs (pg 8-47). a. Page numbers where data are located in the 1997 renovation document are listed in parenthesis. 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 This analysis estimated the number of work hours, or the number of alternative units (such as windows replaced) required for each component of each renovation, repair, and painting task expected to generate lead dust detailed in the POMS and AHS. The activities typically required in each RRP task were analyzed using a normative approach due to the wide variability in homes, materials and the way that work is carried out. This approach used assumptions about what a contractor would reasonably do in the process of carrying out a specific task in an average home (average homes are defined in Chapter 4). For example, replacing plumbing fixtures in a home is likely to require a small amount of two activities sawing and drilling into wood or plaster. The number of fixtures replaced is not specified in the AHS. The work was estimated to require an hour or less of each work component. Although the full work of replacing plumbing fixtures is likely to require more time than this, the work involves many other activities that do not cause lead disruption. For example, the estimated time to complete tasks did not include the actual plumbing activities that would not generate lead paint dust, such as sawing new copper pipe. The process of estimating time and work components for RRP tasks involves uncertainty. The AHS remodeling statistics report the types of renovation activities but not the extent of work. This analysis uses reasonable assumptions regarding task size. For example, EPA assumed that one-half of the §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 75 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 windows in an average size house would be replaced when the survey respondent said windows were replaced. As noted above, when plumbing fixtures were replaced in a home, it was assumed that two major fixtures were replaced (e.g., a toilet and sink), and EPA estimated that one hour each was spent on drilling and sawing. Time units were measured in hours with the smallest unit of time for any task being one hour. For example, tasks figured to involve drilling were estimated to generate at least one hour worth of lead (18 µg/ft2; see Table 5B- 3Table 5B- 3) in the room adjacent to the workroom. An argument could be made that of the work components included in the analysis, power drilling and sawing may not occur continuously for the duration of an hour. However, in calculating the amount of lead generated by a unit of either component, it was assumed that the work occurs uninterrupted for an entire hour. This could have resulted in an over estimate of the number of households with increased lead levels due to RRP. Households that conducted only one renovation or repair task comprised of power drilling were not included in the analysis since the expected lead generation was below the EPA floor lead dust hazard level of 40 µg/ft2. Conversely, households that conducted renovation or repair tasks comprised of power drilling and sawing or just sawing were included in the analysis as the expected lead generation was above the EPA floor lead dust hazard level. Some tasks posed more difficult estimation problems. For example, carpet replacement, which is a very common renovation activity, can be carried out in one or most rooms in a house. Carpet replacement was measured in the renovation study on a per-home basis, in homes ranging from a single apartment to a multi-story house. The number of rooms of carpet replaced and the square footage did not appear to be correlated to home size. The lack of correlation is not unexpected due to the very small sample size in the study (8 homes) and because the number of rooms needing carpet removal varies within a home and does not include the entire house. In the analysis, EPA assumed that the amount of carpet replaced in the study is representative of carpet reported in the 2003 AHS. This was based on a review of home types and sizes. Although carpet replacement in the renovation study cannot be specifically matched to its national occurrence and scope, the renovation study probably provides a reasonable approximation of the types of units, amount replaced, and resulting lead loadings from replacement, based on the information available for review. In addition, the amount of lead released during carpet replacement is quite high, so that if less carpet were replaced, the lead contamination threshold of 40 µg/ft2 would still likely be exceeded. Variability in the amount of work required for a particular task is an uncertainty for most tasks. The variability may be due to the type of materials involved, their age, their condition, etc, as well as the scope of work (e.g., number or size of units). For example, replacing a free-standing bathroom sink may involve minimal carpentry and focus solely on plumbing, while replacing a sink in a vanity may involve removing the cabinet, re-cutting the counter area, etc. The wood may have many coats of paint or none, and the materials may have other variations that alter the work component time required and the amount of lead paint released. Estimates of the work components carried out and the number of units of work (e.g., hours) required was made considering a pre-1978 housing stock, but work on individual homes will vary. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 76 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 5B.2 Work Component Lead Loading Levels The renovation study (U.S. EPA 1997a) provides both airborne (workers’ breathing zone) and floor lead loading measurements for each component, following a specific increment of work (e.g., 102 minutes of sanding) or a particular activity (e.g., replacing 3 windows). EPA chose to use floor lead loadings as the relevant contamination measure in this analysis because they could be matched to the selected benchmarks. Floor lead loading measurements ranged mostly from 1 to 6 feet away from the actual work site in most study evaluations. The 6-foot location was used for estimating loadings in this analysis. The relationship between lead loadings and distance from the worksite declines steeply over the first few feet and then tapers off more gradually. The goal in using these loadings is to estimate contamination levels throughout the work area. Consequently, the most distant measurement is more appropriate for estimating lead loadings in other locations. The workroom lead loadings at 6 feet were subsequently scaled to adjacent room lead loadings for this analysis. All floor measurements used in this analysis were taken one hour post-activity. These were available for all activities, unlike some of the longer post-activity measurements, and so provide consistency in deposition time. However, the use of these data will likely underestimate lead loadings because airborne lead can continue to settle for many hours after work is completed. This was illustrated in the 1997 renovation study, where some measurements were taken both one and two hours post-activity. The study found that deposition continued to occur through the second hour, although at a lower rate (pg A-31 in U.S. EPA 1997a shows lead loadings for the carpet removal work component for one and two hours postactivity). This observation indicates that floor loadings at one hour post-activity are a low estimate of potential lead loadings. Table 5B- 2Table 5B- 2 lists the contaminant levels measured in the 1997 renovation study for each renovation, repair, and painting task. For activities quantified on the basis of work duration, the contaminant levels are presented first as they were reported in the 1997 renovation report. This is followed by the levels standardized to one hour of work. The standardized value was calculated by proportionally scaling the contaminant levels to one hour of work. For example, it was assumed that if 150 µg/ft2 resulted from 1.5 hours of work, then 100 µg/ft2 would result from 1.0 hour of work. The 1997 EPA study reports a distribution of contamination levels for each task at the 6-foot distance (U.S. EPA 1997a). For purposes of this analysis, the 50th percentile values were chosen as most representative of the “average” levels that may occur as a result of the work. They are listed under “measured levels” in Table 5B- 2Table 5B- 2. Maximum or upper 95th percentile airborne lead contaminant values were not used in this analysis directly; however, they are included in the exhibit as an indication of the potential range of exposures that may occur as a result of RRP activities. As shown in Table 5B- 2Table 5B- 2, measurements of contamination resulting from two different types of materials (i.e., wood and plaster) were made for some work components and yielded different lead releases and lead loadings. Because the proportion of work to be carried out in a typical task on each type of material cannot be estimated, EPA calculated an average lead loading level for the two media. For paint removal, drilling, sawing and clean-up, one half hour of work on each type of media was assumed to occur and contamination from the two half hours were added to yield an average hourly lead loading. For example, floor lead loadings for drilling wood and plaster were 3.61 µg/ft2 and 0.1025 µg/ft2 per minute, §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 77 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 respectively. When these are each multiplied by 30 minutes and added together, the resulting average (18 µg/ft2) represents one half-hour of work on each media. This value was used in the analysis as a representative of the floor lead loading generated by one hour of drilling. Lead concentration in household paint varies considerably. The 1997 renovation analysis does not include a quantitative evaluation of the percent of lead in paint, the thickness of the paint or the number of layers of paint that was present in the structural materials. Data are also not available on those characteristics for the national housing stock. Because EPA designed its 1997 study to capture the spectrum of housing types and ages, the study housing units are likely to have similar lead concentrations to those throughout the country. However, the lack of information is a source of uncertainty in the analysis. 5B.3 Estimated Contamination Levels Throughout the TH-COF The 1997 renovation study included measurements of lead concentrations in air and dust floor loadings in work areas, as shown in Table 5B- 2Table 5B- 2Table 5B- 2Table 5B- 2. The measured levels in the work area are not a good measure of residents’ exposure for two reasons: • Clean-up always takes place in the work area after task completion, so that residents are not exposed to pre-clean-up levels (i.e., the primary measurement in the U.S. EPA 1997a study). Although some data on post-clean-up levels exist, they are limited. Contractors usually broomclean or vacuum (frequently using a shop vac) the work area and remove debris. • Lead dispersion is airborne and the heaviest particles settle within a short distance from the worksite. This causes adjacent rooms to have a lower lead loading than that found in the work area, and these adjacent rooms are a better measure of lead loadings occurring throughout the house. Lacking direct lead loading measurements in rooms adjacent to where most RRP activities occurred, EPA estimated these loadings. The estimates are based on the observed relationships between the workroom and adjacent room contamination for window replacement. Adjacent room levels were available only for airborne samples, not for floor samples. It is reasonable to use airborne levels because the lead that settles to the floor results from airborne particles moving through the home. The airborne lead levels in adjacent rooms contribute directly to the floor levels in those rooms. Of the two work components for which adjacent room data are available, window replacement and carpet removal, window replacement is most similar to other work components. It involves working with wood, plaster, and other structural materials, as do the other components. Carpet removal is a unique activity and carpeting is also a unique material known to be a “sink” for lead. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 78 5/17/2007 DRAFT—DELIBERATIVE Table 5B- 2: Renovation, Repair, and Painting Work Components and Related Floor Lead Loadings and Air Concentrations (Based on data from U.S. EPA, 1996. All measurements are in the work room unless noted as “adjacent.”) Floor Lead Loading Air Concentrations (µg/m3) (µg/ft2) a Time Average Average Renovation Task Average Average Maximum (minutes lead concentration (N=Sample Size) lead personal Maximumb concentratio or units) loading for one unit loading concentration n for one measured per unit or hour measured measured unit or hour or hour (See Col 2) Work room Rooms N/A because floor 8.4 8.4 221.3 221.3 Carpet (N=14) needing rereplaced carpeting Work room 130.4 130c 0.3 0.3 13.4 13.4 (see text) (N=16) Work room 3 windows 878.4 878d 7.5 7.5 44.3 44.3 (N=8) Window Adjacent room not not 1.2 not available 4.2 not available (N=8) available available Paint Removal via sanding Hand (N=6) 63 254 242 1,410 1,343 Power (N=3) 39 571 879 3,170 4,877 Both methods 102 15,500e 9,118 413 561 2,290 3,110 (1/2 hour each) Demolition (N=20) 61 1,530 1,505 108 107 403 396 HVAC (N=4) 48 414 518 50 63 66.6 83 Door (N=6) 68 6,700 5,912 590 521 4,480 3,953 Wood (N=7) 36 130 217 15 25 16.3 27 Plaster (N=6) 40 4.1 6 7 11 127 191 Drilling Both methods (1/2 hour 112 each) Wood (N=6) 41 7,900 11,561 546 799 1,020 1,493 Plaster (N=2) 19 480 1,516 110 347 681 2,151 Sawing Both methods 6,539 (1/2 hour each) Component Removal 160 1,464 549 344 129 370 139 (N=2) a. Measurements taken one hour after work was completed. No maximums were available for floor lead loadings other than window and carpet removal (see text). b. All but carpet and window data are 95th percentiles. See U.S. EPA 1997a for a description of the maximum levels estimated for carpet and window replacement. c. Maximum was reported as 6135 :g/ft2 for the floor in the adjacent room. d. Maximum was reported as 54.515 :g/ft2 for the floor in the adjacent room. e. A single value was reported for 102 minutes (hand and power sanding not measured separately). 1 2 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 79 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 The airborne window replacement lead concentrations shown in Table 5B- 2Table 5B- 2Table 5B- 2Table 5B- 2 for the same room (7.5 µg/m3) and the adjacent room (1.2 µg/m3) were used to obtain a ratio that is expected to be characteristic of the “workroom-adjacent room” floor lead loadings relationship for other work components. This ratio between the workroom and the adjacent room was calculated as: 1.2 µg/m3 / 7.5 µg/m3 = 0.16 (a dimensionless value) EPA multiplied this ratio (0.16) times the workroom floor lead loadings for each activity shown in Table 5B- 3Table 5B- 3 to estimate the floor lead loadings in an adjacent room for one unit of work. For example, one hour of demolition results in 1,505 µg/ft2 in the workroom. This is multiplied by 0.16 to obtain an estimate of the floor lead loadings in the adjacent rooms of 241 µg/ft2. Table 5B- 3 contains the estimated floor lead loadings in adjacent rooms for all the work components after applying the 0.16 scaling factor to each task shown in Table 5B- 2Table 5B- 2. Table 5B- 3: Work Components, Unit/Time of Measurement, Measured and Estimated Floor Lead Loadings and a Comparison to the Proposed EPA Floor Standard (40 µg/ft2) Measured Level Estimated Levels in One Unit Exceeds in Work Room Adjacent Rooms Proposed EPA Standard Work Componenta I.D. Codeb (µg/ft2)c (µg/ft2)d A Not available 130e Yes Carpet removal B 878 141 Yes Window replacement C 9118 1459 Yes Paint removal D 1505 241 Yes Demolition E 518 83 Yes HVAC F 5912 946 Yes Door removal G 112 18 No Drilling H 6539 1046 Yes Sawing I 549 88 Yes Component removal a. See definitions in Table 5B- 1. b. Identification codes that were used to assign specific work components to each task are listed in Table 5B4Table 5B-1Table 5B-1Table 5B-4 and Table 5B- 5Table 5B- 2. c. Measurements in the work room are taken from d. A multiplier of 0.16 was used to estimate floor lead loading in adjacent rooms. See text for discussion. e. The value for carpet replacement is an actual measurement, as opposed to an estimated value. 16 17 18 19 20 21 22 23 24 25 26 Adjacent room loadings are assumed to be reasonably representative of lead loadings throughout the house. However, lead loadings will vary in a home and also vary over time. In the absence of clean-up, the lead loadings in rooms nearest the workroom will initially be the highest. The levels will decline as activities and physical processes cause lead to be moved to other rooms. Consequently, although the loading estimates in this analysis are relevant for an extended period after the RRP work, the specific dynamics of the very long-term loadings are not known. Actual levels in homes depend on a variety of factors including home size, airflow, resident behavior, the paint’s lead concentration and thickness, and other factors. In a very large house, there would likely be some rooms with lower lead loadings than those near the workroom. In the absence of specialized clean-up, however, the lead would eventually be distributed to all rooms in the house. Based on available data, it was not possible to determine the §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 80 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 distribution of lead loadings in the various rooms of a home over the long term. The most relevant rooms are those where residents (especially children) spend most of their time. 5B.4 Renovation, Repair, and Painting Tasks and Lead Loading Estimates This analysis uses the 1995 Property Owners and Managers Survey (POMS) (U.S. Census 1995) and the 1997 and 2003 American Housing Survey (AHS) (U.S. Census 1997 and 2003) to characterize the different RRP tasks that Target Housing COFs perform and the lead loadings that result from performing these tasks. The AHS and the POMS provide data on the type of housing unit (e.g., single-family homes, apartment buildings, the age of the structure) and other critical features of the housing stock (e.g., number of rooms, type of interior RRP). The AHS also describes the residents of homes in which these tasks take place, including whether there are children under the age of 6, the size of families and other demographic features. In 1997, the AHS provided detailed survey information on renovation and repair tasks performed in homes from a panel of residents living in over 50,000 housing units. In 2003, the AHS continued to cover renovation and repair activities, but reported the data in less detail, combining the 1997 AHS tasks into a smaller number of tasks. As a result, EPA uses the 2003 AHS for characterizing some RRP tasks and 1997 AHS data to characterize RRP tasks where the level of detail in the 2003 survey is not sufficient. The survey results are extrapolated to the national population using weighting factors that adjusted the survey housing stock to match the national housing stock.29 From the three data sources mentioned above, those tasks likely to generate some lead dust in homes built prior to 1978 were identified. Because it is not known which respondents to the AHS live in structures that actually have lead paint, data from the 2000 HUD National Survey of Lead and Allergens in Housing was used to estimate the number of RRP events occurring in homes that contain lead-based paint and to estimate the number of homes with a pre-existing lead hazard. To determine the estimated exposure to lead dust resulting from an RRP event, EPA estimated the total floor lead loadings for the RRP tasks where painted surfaces are likely to be disturbed. Table 5B- 4Table 5B-1Table 5B-1 shows the RRP tasks that are reported in the 2003 AHS (owner-occupied RRP) and the 1995 POMS (rental unit RRP), the work components associated with each task, and the estimated household dust loadings (outside work area) resulting from the work components. Since some tasks are too general to assign work components to, 1997 AHS data are used to estimate the work components and resulting COF lead loadings for these tasks. In these cases, Table 5B- 4Table 5B-1Table 5B-1 lists the 1997 tasks and Table 5B- 5Table 5B- 2 shows the work components and lead loadings associated with these tasks. An average weighted by the relative frequencies observed in the 1997 AHS data is used to estimate the lead loading associated with the more general task used in the analysis. For example, the lead loading associated with 2003 AHS task 45, Added/Replaced Doors Or Windows In Home (1,991 µg/ft2) is estimated as the weighted average of 5,492 (loading for 1997 AHS Task 45) and 1,599 (loading for 1997 AHS Task 46); 1997 tasks 45 and 46 are weighted based on the relative frequency of their occurrence in the 1997 AHS. 29 For this analysis, sample weights developed by the Harvard University Joint Center for Housing were used to develop national estimates. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 81 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 Using the task definitions in the AHS, EPA estimated the amount of time each component would typically be employed for each task. While the actual amount of each component used may vary widely within a task, data are not available to provide a distribution of values. Further data collection would be necessary to improve this effort. The number of units of each work component was multiplied by the expected household dust loading resulting from one unit of each activity. The totals for each component were summed to get an overall estimate of the floor loading outside the work area for each task. For example, replacing plumbing fixtures in a home (task #47 listed in Table 5B- 4Table 5B-1) was estimated to require one hour each of drilling and sawing. Drilling results in 18 µg/ft2 and sawing 1,046 µg/ft2 (from Table 5B- 3Table 5B- 3) with a summed estimate of 1,064 µg/ft2. All tasks shown exceed EPA’s proposed standard of 40 µg/ft2 of lead on floors. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 82 5/17/2007 DRAFT—DELIBERATIVE Table 5B- 4Table 5B-1: RRP Tasks and Their Associated Lead Loadings 2003 AHS Task Description (Owner-Occupied Units) Resulting Lead Loading 1997 AHS Task ID or Outside Work 2003 AHS Task POMS Task Description a 2003 Work Components ID (Rental Units) Areab (µg/ft2) 1997 Task 16 1997 Task 17 Remodeled Bathroom 71 Renovation Of Bathroom 1997 Task 18 1997 Task 19 5,701 1997 Task 21 1997 Task 25 1997 Task 26 1997 Task 27 1997 Task 29 Remodeled Kitchen 72 Replacement Of Kitchen 1997 Task 30 10,874 1997 Task 32 1997 Task 33 1997 Task 34 Added Room, Or Room Created Through Structural Changes 8, 9, 10, 26, 35, 36 N.A. Added/Replaced Internal Water Pipes In Home 40 Upgrading Plumbing System Added/Replaced Electrical Wiring To Home 42 Unit Rewired Added/Replaced Doors Or Windows In Home 45 N.A. Added/Replaced Plumbing Fixtures In Home 47 N.A. Installed Paneling Or Ceiling Tiles 55 N.A. 4D, 2G, 2H, 4I, F, 4C 1997 Task 40 1997 Task 41 1997 Task 42 1997 Task 43 1997 Task 45 1997 Task 46 G, H 1997 Task 55 1997 Task 56 10,226 288 746 1,991 1,064 3,433 Added/Replaced Central Air 57 2G, 2H, 8E 2,792 Upgrading Heating System; Conditioning Heating/AC Unit Repaired; Added/Replaced Built-In Heating Add Or Upgrade AC.c G, H, 2I 1,240 58 Equipment Other Major Improvements Or Other Major Repairs Or 2G, 2H, 2C, 2I 5,222 64 Repairs Inside Home Capital Improvement Added/Replaced Security System In G, C 1,477 74 Addition of Security System Home 4C 5,836 Interior Painting N.A. Interior Painting a. See Table 5C-5 for the work component key, 1997 task descriptions, and lead loadings associated with the 1997 tasks. b. Loadings shown are the estimated loadings for the room adjacent to the work area. Loading estimated based on 1997 tasks are calculated as the weighted average loading according to the 1997 AHS. c. The loading associated with rental units reporting any of these tasks is estimated as the weighted average loading observed for owner-occupants reporting task 57 and/or task 58 (a loading of 3,226). Note that these tasks are frequently performed together. Source: EPA Calculations 1 2 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 83 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5B- 5Table 5B- 225: Work Components Assigned to each Interior Renovation Task and Expected Lead Loadings AHS Work Components (listed in Estimated adjacent Task Task Name Key) and number of units of room levels (µg/ft2) ID each component (see text). 16 Moved Walls In Bathroom 2D, 2I, G, H, F, 4C 8,504 17 Added/Replaced Cabinets In Bathroom G, H, I, C 2,611 18 Added/Replaced Flooring In Bathroom I 88 19 Added/Replaced Counter Tops In Bathroom G, H, I, C 2,611 21 Added/Replaced Tub/Shower In Bathroom G, C 1,477 25 Painted/Papered/Wall Tiles Bathroom 2C 2,918 26 Moved Walls In Kitchen 4D, 2G, 2H, 4I, F, 4C 10,226 27 Added/Replaced Cabinets In Kitchen 2G, 2H, 2I, 2C 5,222 29 Added/Replaced Counter Tops In Kitchen 2G, 2H, 2I, C 3,763 Added/Replaced Other Built-In Appliances In 30 G, H, I 1,152 Kitchen 32 Added/Replaced Lighting Fixtures In Kitchen G, C 1,477 33 Added/Replaced Other Electrical Items In Kitchen G, H, I 1,152 34 Painted/Papered/Wall Tiles Kitchen 4C 5,836 40 Added Internal Water Pipes To Home 2G, 2H 2,128 41 Replaced Internal Water Pipes In Home 3G 54 42 Added Electrical Wiring To Home G, C 1,477 Completely Rewired The Electrical Wiring In 43 4G, I 160 Home 45 Added Doors/Windows To Home 2H, 2D, 2C 5,492 46 Replaced Door/Windows To Home B, C 1,599 55 Installed New Paneling/Ceiling Tiles G, C 1,477 56 Replaced Existing Paneling/Ceiling Tiles 8I, 2C, G, H 4,686 Key: A = carpet removal C = paint removal E = HVAC G = drilling I = remove trim (See text) (1 hour) (1 hour) (1 hour) (1 hour) B = replace windows D = interior demolitions F = door removal H = sawing (3 windows) (1 hour) (1 hour) (1 hour) 2 3 4 5 6 7 8 9 10 11 12 5B.5 Interior Painting 5B.5.1 Number of Homes Painted This analysis identified the number of homes performing an interior painting event using the 1997 AHS. This is described in more detail in Chapter 4 of this document. 5B.6 Background Lead Loadings in Homes §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 84 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 Background lead levels existing in homes prior to RRP work have not been added to the loading levels resulting from RRP events in this analysis, although they may have an impact on one task: carpet removal.30 The lead loadings reported in the renovation study (U.S. EPA 1997a) include only postactivity levels on floors for most tasks and involved a sampling method that started with a clean, smooth floor surface at the outset of the task (window and carpet replacement did not take this approach and include pre-existing levels). In actual practice any RRP event will add to the background level. 7 30 The high lead levels generated by carpet replacement are assumed to be due to the presence of accumulated lead (Yiin 2002). The levels reported in EPA’s renovation study reflect carpet removal from a home with lead-based paint. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 85 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Appendix 5C: Identifying and Characterizing Lead Loadings for Interior Renovation, Repair and Painting Tasks in COFs in Public or Commercial buildings. Dust Loading Calculation Dust loadings for COFs in public or commercial buildings are estimated in the same way as target housing-wide dust loadings are estimated (i.e., as described in Appendix 5B)—with two exceptions: (1) instead of assuming an average work area size and the adjacent area size for each event type, this analysis assumes that all classrooms are renovated at the same time during scheduled maintenance, so that the work area is the size of all areas that contain LBP31 on disturbed surfaces and the adjacent area is either the same size as the work area or the remaining space in the COF, whichever is smaller;32 and (2) instead of assuming one average dust loading, a distribution is generated of dust loadings which are a function of the work area and adjacent area sizes as a share of the COF size. More precisely, the analysis estimates each possible COF-wide dust loading as follows: • • • Assume that all the rooms in a COF are renovated at the same time when scheduled maintenance occurs. Assume the work area for a large RRP event is the size of the rooms with LBP, and that the work area for a small event is the area along one wall of a room.33 Assume the adjacent area is the size of the work area, or the size of the remainder of the COF, whichever is smaller. Thus, COF-wide loading is calculated as: [work area loading] * [percentage size of work area] + [adjacent area loading ]* minimum of {[percentage size of work area], 1 – [percentage size of work area ]}, where the adjacent area loading is assumed to be 16 percent of the work area loading (which is the same assumption used for target housing). The estimated distribution of loadings are presented in Appendix 5D. Dust Loading Calculation 31 COFs in public or commercial buildings can have LBP in some classrooms and not in others. The likelihood that one or more classrooms contain LBP was estimated using data from HUD 2003. 32 For example, for a large event in a COF with 3 classrooms where 1 has LBP, the adjacent area is assumed to be 1 classroom (the same as the work area). But for a large event in a COF with 3 classrooms where 2 have LBP, the adjacent area is assumed to be 1 classroom (the remaining space in the COF), since assuming the adjacent area is the same size as the work area would exceed the size of the COF (i.e., 2 classrooms in work area + 2 classrooms in adjacent area > total size of 3 classrooms). For large scheduled maintenance events where all classrooms contain LBP (i.e., the entire COF qualifies as the work area), the analysis does not incorporate adjacent area loadings. 33 Using the same classroom size for all RRP events differs from the analysis of target housing, which used different average room sizes depending on the RRP event being performed (since the average room sizes differs for bathrooms, kitchens, additions, etc.). §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 86 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Dust loadings for COFs in public or commercial buildings are estimated in the same way as target housing-wide dust loadings are estimated (i.e., as described in Appendix 5B)—with two exceptions: (1) instead of assuming that the work area size and the adjacent room size are each 16% of the housing unit,34 this analysis assumes that all classrooms are renovated at the same time during scheduled maintenance, so that the work area is the size of all rooms that contain LBP on disturbed surfaces and the adjacent room is either the same size as the work area or the remaining space in the COF, whichever is smaller; and (2) instead of assuming one average dust loading, a distribution is generated of dust loadings which are a function of the work area and adjacent room area sizes as a share of the COF size. More precisely, the analysis estimates each possible COF-wide dust loading as follows: Assume that all the rooms in a COF are renovated at the same time when scheduled maintenance occurs. Assume the work area is the size of the rooms with LBP (or in the case of a small unscheduled maintenance event, the area along one wall of a room, 135 square feet). oNote: assumed to be 16% of the house in the target housing. Assume the adjacent room is the size of the work area, or the size of the remainder of the COF, whichever is smaller. oNote: assumed to be 16% of the house in target housing. Thus, COF-wide loading is calculated as: [work area loading] * [percentage size of work area] + [adjacent area loading ]* min{[percentage size of work area], 1 – [{[percentage size of work area ]}, The estimated distribution of loadings are presented in Appendix 5D. 34 For cost estimation purposes, large interior painting jobs involved 25% of the unit and small interior painting jobs involved 7% of the unit. The midpoint, 16% of the unit, was the work area size assumed for the purposes of benefits estimation. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 87 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 Appendix 5D: Distributions of Inputs and Results As described in section 5.5, the estimation of benefits uses a Monte Carlo approach to reflect the uncertainties inherent in the available data. Several of the inputs used in the analysis are described as distributions, and the results are expressed as a distribution of estimates. This appendix supplements section 5.5 by providing additional details on the various distributions. 5D.1 Custom Distributions for Input Parameters Four of the input parameters are in the form of custom distributions. These are: Soil Lead Concentrations, Frequency of COF Cleaning, Percent of COF Represented by Work Room Area, and Initial Interior Lead Levels. The distribution of soil lead concentrations due to exterior painting events is shown in Table 5D- 1Error! Reference source not found.Table 5D-1. There are five paint-removal techniques. Within a housing type, the likelihood of occurrence for each of the five techniques is assumed to be the same. But the number of exterior paint removal events varies across housing types based on the number of housing units of each type. Thus the percentage of exterior paint removal events by technique varies across housing types. Also, the lead concentration for a particular paint removal approach varies across housing types because the relationship of the perimeter area to the rest of the yard area also varies across housing types.35 For COFs in target housing, the analysis assumes that one quarter of the exterior paint removal events are large, including the entire exterior of the home. One quarter are assumed to be small, including only one wall of the home. And the remaining half are in between large and small events (including the average of four walls and one wall). Similarly, the analysis assumes that one quarter of exterior paint removal events for COFs in public or commercial buildings are large, one quarter are small, and half are in between. The quarter of exterior paint removal events that are small are assumed to take place along one exterior wall regardless of the type of COF. In the quarter that are large, the number of walls where work is assumed to take place differs for daycare centers and schools (i.e., kindergartens/pre-kindergartens). For COFs in public or commercial buildings, the rule’s requirements for exterior events only apply to the exterior sides of the building that are immediately adjacent to the COF or the common areas routinely used by children under the age of six. Large exterior paint removal events in daycare centers are assumed to include the entire exterior of the building (four exterior walls). However, only part of a school’s exterior may be covered by the rule. Large exterior paint removal events in schools are only assumed to include two exterior walls. (It is assumed that the remaining walls of the school building are not adjacent to classrooms and common areas routinely used by children under the age of six.) 35 For each housing type, the analysis assumes that one-quarter of the exterior paint removal events encompass the entire exterior, one-quarter encompass only one side of the home and the other one-half encompass some thing in between. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 88 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 For all types of COFs in public or commercial buildings, the remaining half of exterior paint removal events include the average of large and small events (i.e., the average of one wall and four walls for daycare centers, and the average of one wall and two walls for schools). Table 5D- 1: Custom Distribution Values for Input Parameter: Soil Lead Concentration Outdoors without Rule Probability of Occurrence /Percent Soil Conc. (µg/g) of Households Single Family Target Housing COF – Small Job (1 exterior wall) 10% Exterior paint – alkaline 512 10% Exterior paint - heat gun 659 10% Exterior paint - paint shaver 511 10% Exterior paint - safe stripper 623 10% Exterior paint – wet scrape 613 Single Family Target Housing COF – Large Job (4 exterior walls) 577 Exterior paint – alkaline 10% 10% 1,166 Exterior paint - heat gun 10% 575 Exterior paint - paint shaver 10% 1,023 Exterior paint - safe stripper 10% 983 Exterior paint – wet scrape Public or Commercial Building COF – Small Job (1 exterior wall) 10% Exterior paint – alkaline 104 10% Exterior paint - heat gun 251 10% Exterior paint - paint shaver 103 10% Exterior paint - safe stripper 215 10% Exterior paint – wet scrape 205 Public or Commercial Building COF; Kindergarten or Pre-Kindergarten – Large Job (2 exterior walls) 10% Exterior paint – alkaline 134 10% Exterior paint - heat gun 428 10% Exterior paint - paint shaver 133 10% Exterior paint - safe stripper 357 10% Exterior paint – wet scrape 337 Public or Commercial Building COF; Daycare Center – Large Job (4 exterior walls) 10% Exterior paint – alkaline 194 10% Exterior paint - heat gun 783 10% Exterior paint - paint shaver 192 10% Exterior paint - safe stripper 640 10% Exterior paint – wet scrape 600 Source: See Section 5.2.5 for development of the lead concentrations. See Section 4.2.5 for development of probabilities. 6 7 8 9 The second custom distribution is represents the frequency of household cleaning. In estimating benefits, households are assumed to clean their house immediately following the initial cleaning by the contractor §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 89 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 (or in-house staff performing RRP) at the end of the RRP project. After that cleaning, the COF is assumed to clean the house on a routine basis, with the frequency drawn from the distribution shown in Table 5D- 2:Table 5D- 2:Table 5D- 2. Centers are assumed to clean on a daily basis. Table 5D- 2:: Custom Distribution Values for Input Parameter: Frequency of Household Cleaning per Year Frequency of Probability of Cleaning Per Year Occurrence Greater than Weekly 104 40% Weekly 52 45% Less than Weekly 26 15% Source: Simcox, 1995 6 7 8 9 10 11 12 13 The house-wide average interior lead loadings are calculated using the lead loadings in the workroom and in the adjacent room, weighted by the percent of house the workroom and the adjacent room represent. These percentages are shown in Table 5D- 3:Table 5D- 3:Table 5D-3. These values are not used by them selves, but are combined with the lead loading likelihood values, as shown in Table 5D-4Table 5D-4 below. Table 5D- 3: Work Area Size by Event Type (Percent of Housing Unit) Work Area Event Type Size Kitchen 6% Bathroom 3% Addition 5% Non-Room-Specific 30% Interior Painting 16% Average Household Work Area 24% Source: Calculated from the American Housing Survey, see Chapter 4 for details. 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Each combination of RRP events yields a specific lead loading in the workroom. A “combination of RRP events” refers to the event or events performed in a single housing unit in one year. This may be a single event (e.g., renovating a kitchen), or it may be a several events involving multiple rooms (e.g., a kitchen and a bathroom). In addition, RRP events where the type of activity is reported but not the location with in the home. Examples are: added or replaced doors or windows, internal water pipes or electrical wiring, other major improvements or repairs. These events are referred to as Non-Room Specific events. Associated with each of the workroom lead loadings is an adjacent room lead loading, equal to 16 percent of the workroom lead load. Each combination of RRP events has a particular probability of occurring, based on the American Housing Survey data and likelihood of lead. Because the category called Non-Room Specific RRP events contains so many different RRP activities (See Table 4-2 in EPA 2006b) and each combination of activities has its own estimated lead loading, the most commonly reported Percent of Home that is Work §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 90 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 Area is 29.9 percent. The second to last column Table 5D-4Table 5D-4 presents the probability that the particular RRP event and associated lead loadings will occur in COF RRP events performed by outside contractors in target housing. The most commonly occurring single RRP event is interior painting. Combinations shown with a 0.0% probability on the table have a positive but very small likelihood (less than 0.05%) of occurring. Table 5D-5Table 5D-5Table 5D-5 presents similar distributional information for DIY COF RRP events in target housing. Table 5D-6Table 5D-6Table 5D-6, Table 5D-7Table 5D7Table 5D-7, and Table 5D-8Table 5D-8Table 5D-8 present this distributional data for COF RRP events in daycare Centers, in Kindergartens, and in COFs that contain both Pre-Kindergarten and Kindergarten space, respectively. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 91 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 36,475 12,444 32,638 67,963 103,594 20,163 6,650 110,244 69,113 35,631 104,438 100,600 42,281 4,663 68,269 43,125 136,231 56,638 74,613 140,069 72,106 48,919 9,231 114,906 48,075 80,406 19,094 88,125 146,719 39,288 116,038 17,106 142,881 55,794 41,138 Adjacent Room Loading (µg/ft2) 5,836 1,991 5,222 10,874 16,575 3,226 1,064 17,639 11,058 5,701 16,710 16,096 6,765 746 10,923 6,900 21,797 9,062 11,938 22,411 11,537 7,827 1,477 18,385 7,692 12,865 3,055 14,100 23,475 6,286 18,566 2,737 22,861 8,927 6,582 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 16.0% 29.9% 29.9% 5.6% 8.9% 29.9% 29.9% 29.9% 29.9% 3.3% 21.6% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 24.9% 19.3% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 23.21182% 8.44208% 6.93930% 5.47977% 3.98626% 3.23017% 3.01093% 2.29407% 2.17626% 1.83906% 1.70055% 1.36442% 1.32739% 1.27051% 1.13491% 1.12594% 1.12186% 1.07959% 1.07334% 1.06467% 1.04518% 0.97507% 0.82562% 0.72357% 0.71737% 0.66076% 0.56257% 0.55348% 0.52073% 0.49111% 0.47829% 0.47008% 0.44736% 0.42480% 0.40266% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z q p y n mn r w mnw qy m nq ny mw v my qw mny qr nw mnq mq pq z mnvw mp np pw nr mnqw wy mnp pv mnwy mr qv 92 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 40,294 108,256 151,381 74,919 32,606 137,075 172,706 52,800 78,756 46,944 130,406 111,088 21,675 104,744 123,756 147,544 26,813 41,869 140,894 62,444 54,725 122,688 37,300 75,763 1,800 63,913 144,731 77,194 92,269 76,769 21,456 119,475 184,019 79,581 84,550 11,313 Adjacent Room Loading (µg/ft2) 6,447 17,321 24,221 11,987 5,217 21,932 27,633 8,448 12,601 7,511 20,865 17,774 3,468 16,759 19,801 23,607 4,290 6,699 22,543 9,991 8,756 19,630 5,968 12,122 288 10,226 23,157 12,351 14,763 12,283 3,433 19,116 29,443 12,733 13,528 1,810 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 5.2% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.40085% 0.39253% 0.37524% 0.36630% 0.35356% 0.34520% 0.33260% 0.32440% 0.31252% 0.31110% 0.29752% 0.29668% 0.27039% 0.27038% 0.26885% 0.26414% 0.24901% 0.23211% 0.22553% 0.20696% 0.20300% 0.19774% 0.19579% 0.18375% 0.18249% 0.18168% 0.18117% 0.18110% 0.17562% 0.17111% 0.16134% 0.16047% 0.15572% 0.15222% 0.15137% 0.15109% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z mv mnv mnqvw mwy pr nqy mnqy ry mqw mvw mnrw nqw pz mqy mnr mnvwy rw yz mnvy mrw mpw mnpw vy qwy u o mnqv nz mqr mqv x mnwz mnqvwy mvwy mpq vw 93 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 177,369 152,513 116,881 33,900 107,250 179,356 89,275 88,431 83,419 111,394 127,350 63,288 59,388 73,775 94,775 38,563 52,738 124,600 79,275 120,700 44,863 120,763 156,394 68,238 72,931 80,425 135,069 72,625 45,706 23,756 47,788 43,950 109,831 145,463 87,056 55,569 Adjacent Room Loading (µg/ft2) 28,379 24,402 18,701 5,424 17,160 28,697 14,284 14,149 13,347 17,823 20,376 10,126 9,502 11,804 15,164 6,170 8,438 19,936 12,684 19,312 7,178 19,322 25,023 10,918 11,669 12,868 21,611 11,620 7,313 3,801 7,646 7,032 17,573 23,274 13,929 8,891 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.15036% 0.14857% 0.14765% 0.14369% 0.14272% 0.14125% 0.13696% 0.12952% 0.12940% 0.11471% 0.11394% 0.11090% 0.10498% 0.10199% 0.10169% 0.10069% 0.09789% 0.09769% 0.09464% 0.09400% 0.09392% 0.09217% 0.09013% 0.08769% 0.08650% 0.08256% 0.08241% 0.08094% 0.07943% 0.07646% 0.07534% 0.07416% 0.07158% 0.06987% 0.06969% 0.06966% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z mnqvy mnpq npq px nwy mnqwy qry mry mqvw mqwy mnpvw qrw mpvw qvy nrw pvx mpv nqr nvw mnpv mz nry mnry mpr mvy qvwy mnrvw nv qz pvw qvw vwy nyz mnyz npw pqw 94 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 109,406 163,825 69,081 25,556 25,200 116,056 78,344 157,175 160,231 111,913 24,825 98,919 171,544 61,300 91,200 45,081 159,163 163,044 113,669 109,100 174,838 143,725 110,925 105,263 53,581 157,238 39,256 100,569 17,450 152,113 124,906 112,825 6,463 166,881 91,719 59,450 Adjacent Room Loading (µg/ft2) 17,505 26,212 11,053 4,089 4,032 18,569 12,535 25,148 25,637 17,906 3,972 15,827 27,447 9,808 14,592 7,213 25,466 26,087 18,187 17,456 27,974 22,996 17,748 16,842 8,573 25,158 6,281 16,091 2,792 24,338 19,985 18,052 1,034 26,701 14,675 9,512 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.06923% 0.06913% 0.06866% 0.06816% 0.06728% 0.06420% 0.06366% 0.06266% 0.06250% 0.05827% 0.05706% 0.05540% 0.05494% 0.05447% 0.05444% 0.05407% 0.05338% 0.05160% 0.05154% 0.05103% 0.05094% 0.04831% 0.04712% 0.04530% 0.04517% 0.04415% 0.04407% 0.04394% 0.04369% 0.04362% 0.04343% 0.04239% 0.04209% 0.04186% 0.04068% 0.04045% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z mqvy mnpqvw pqr puvw st mqvwy qyz mnpqv mnqr nvwy rv mqrw mnqrvw qrv mpqw py mnpqw mnrwy nqz nqv opuvwx nqwy opuvwx nvy pqv nqry prw npr s mnwyz mqry mnz uv mnqrw npvw rwy 95 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 131,919 30,688 95,863 85,588 136,200 50,369 161,056 83,844 192,869 102,475 14,244 89,213 125,269 210,469 13,894 278,431 15,881 160,594 156,775 60,231 167,706 181,938 293,600 58,150 221,344 99,544 409,031 114,494 149,300 61,188 111,988 68,575 139,638 76,413 29,394 106,194 Adjacent Room Loading (µg/ft2) 21,107 4,910 15,338 13,694 21,792 8,059 25,769 13,415 30,859 16,396 2,279 14,274 20,043 33,675 2,223 44,549 2,541 25,695 25,084 9,637 26,833 29,110 46,976 9,304 35,415 15,927 65,445 18,319 23,888 9,790 17,918 10,972 22,342 12,226 4,703 16,991 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 8.5% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.03920% 0.03887% 0.03866% 0.03821% 0.03744% 0.03721% 0.03713% 0.03699% 0.03665% 0.03646% 0.03636% 0.03619% 0.03607% 0.03506% 0.03467% 0.03344% 0.03309% 0.03267% 0.03224% 0.03208% 0.03195% 0.03149% 0.03139% 0.03067% 0.03014% 0.02992% 0.02986% 0.02949% 0.02826% 0.02824% 0.02761% 0.02756% 0.02714% 0.02697% 0.02684% 0.02678% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z mnpwz xz mpqvw opz mnpr qvz mnrvy nwz mnqry opvx pu mpqv mnpz mopuvwx vz mnopuvwx wz ovwx mnvwyz pqvw mnrvwy mnqyz nopx pqz nopuvw mo opuvw nvyz mnqz mpuvw mop ov mnrwz nuw rz mow 96 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 104,713 148,388 7,750 26,119 123,531 37,644 115,750 89,638 186,600 48,519 13,113 197,531 178,888 238,681 45,213 67,950 51,513 95,081 137,044 95,925 100,950 131,250 28,325 84,994 193,250 199,519 204,181 96,288 85,069 142,850 57,306 31,475 93,094 140,863 238,750 127,825 Adjacent Room Loading (µg/ft2) 16,754 23,742 1,240 4,179 19,765 6,023 18,520 14,342 29,856 7,763 2,098 31,605 28,622 38,189 7,234 10,872 8,242 15,213 21,927 15,348 16,152 21,000 4,532 13,599 30,920 31,923 32,669 15,406 13,611 22,856 9,169 5,036 14,895 22,538 38,200 20,452 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 10.4% Chapter 5 0.02613% 0.02608% 0.02580% 0.02579% 0.02571% 0.02555% 0.02531% 0.02520% 0.02511% 0.02492% 0.02492% 0.02475% 0.02443% 0.02402% 0.02394% 0.02367% 0.02275% 0.02273% 0.02255% 0.02237% 0.02228% 0.02224% 0.02220% 0.02209% 0.02149% 0.02138% 0.02130% 0.02112% 0.02103% 0.02074% 0.02061% 0.02024% 0.01972% 0.01944% 0.01903% 0.01877% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z mpqr nqvwy t vx npqw pst nqvw npz mnopw wyz uvw mnqrvy nopuvwx movw pvwx qrvw mwz mrwy npqr qrwy npvwz nqrw pwz qwyz mnqvwyz mnqrwy mnqrvwy npwz npv mnprw mpz rvw mrvy mnprv opuvwx o 97 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 128,419 66,538 89,419 172,675 117,488 147,513 32,769 200,275 67,106 129,569 77,500 28,106 325,713 72,900 146,306 300,513 117,744 74,888 23,256 20,894 70,531 91,875 138,163 20,544 52,356 339,131 177,338 144,144 113,975 164,644 155,950 192,288 366,575 133,500 42,094 105,606 Adjacent Room Loading (µg/ft2) 20,547 10,646 14,307 27,628 18,798 23,602 5,243 32,044 10,737 20,731 12,400 4,497 52,114 11,664 23,409 48,082 18,839 11,982 3,721 3,343 11,285 14,700 22,106 3,287 8,377 54,261 28,374 23,063 18,236 26,343 24,952 30,766 58,652 21,360 6,735 16,897 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.01847% 0.01830% 0.01819% 0.01711% 0.01684% 0.01675% 0.01643% 0.01607% 0.01595% 0.01575% 0.01563% 0.01545% 0.01533% 0.01531% 0.01530% 0.01491% 0.01487% 0.01433% 0.01424% 0.01412% 0.01412% 0.01400% 0.01346% 0.01335% 0.01334% 0.01303% 0.01296% 0.01294% 0.01255% 0.01235% 0.01213% 0.01206% 0.01200% 0.01188% 0.01185% 0.01179% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z mnrv pxy nx mnpqr mnvz mnprvw vwx mnovwx mrvw mqrvy myz wx mostz mprv nqyz moz npwxz mprw ux puw qrvz mpuvwxz mnuvwx vwz qwz mnouvwxz mnpqrv mnpwx mqyz novwx mnqwz opsuvwx opuvwx mnuwx muv npst 98 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 191,263 48,744 74,425 209,856 153,963 183,988 136,581 94,731 75,731 161,744 69,531 94,300 109,375 31,850 97,356 93,938 58,888 179,325 100,588 34,056 103,581 62,031 186,625 141,738 164,894 128,194 62,663 37,269 188,588 146,556 57,463 338,294 172,425 132,313 208,056 172,275 Adjacent Room Loading (µg/ft2) 30,602 7,799 11,908 33,577 24,634 29,438 21,853 15,157 12,117 25,879 11,125 15,088 17,500 5,096 15,577 15,030 9,422 28,692 16,094 5,449 16,573 9,925 29,860 22,678 26,383 20,511 10,026 5,963 30,174 23,449 9,194 54,127 27,588 21,170 33,289 27,564 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.01176% 0.01171% 0.01168% 0.01131% 0.01122% 0.01117% 0.01080% 0.01071% 0.01057% 0.01055% 0.01042% 0.01038% 0.01021% 0.01014% 0.01000% 0.00992% 0.00968% 0.00946% 0.00946% 0.00942% 0.00914% 0.00912% 0.00885% 0.00877% 0.00875% 0.00873% 0.00872% 0.00870% 0.00840% 0.00819% 0.00813% 0.00805% 0.00781% 0.00773% 0.00773% 0.00764% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z mnopvw muvw nuv mnopuwx mnqvz mnpqrvw mnpvwz mpstx pqrw mnpqz mpx npvz mpqrv stw nrz qrvy mux mnpqrw qrvwy rvz mqrvw ryz mopvwxyz nqvy mnqrv npqvw psvwx prv mnqwyz mopuvwx rvy mopuvwx nopwx movwx mnopwx mnrwyz 99 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 165,625 269,981 225,638 53,313 132,988 64,800 43,919 173,056 62,538 41,838 24,100 37,338 57,838 15,694 63,738 327,863 127,675 104,006 302,663 99,438 38,906 64,113 127,413 127,650 11,031 121,275 263,950 18,906 29,863 123,681 26,338 98,481 41,081 108,325 126,113 119,694 Adjacent Room Loading (µg/ft2) 26,500 43,197 36,102 8,530 21,278 10,368 7,027 27,689 10,006 6,694 3,856 5,974 9,254 2,511 10,198 52,458 20,428 16,641 48,426 15,910 6,225 10,258 20,386 20,424 1,765 19,404 42,232 3,025 4,778 19,789 4,214 15,757 6,573 17,332 20,178 19,151 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.00759% 0.00758% 0.00757% 0.00755% 0.00742% 0.00740% 0.00727% 0.00720% 0.00701% 0.00692% 0.00684% 0.00682% 0.00677% 0.00677% 0.00674% 0.00661% 0.00653% 0.00648% 0.00643% 0.00632% 0.00631% 0.00629% 0.00607% 0.00586% 0.00573% 0.00564% 0.00563% 0.00561% 0.00553% 0.00549% 0.00537% 0.00534% 0.00530% 0.00526% 0.00512% 0.00507% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z mnryz mnopvx opx muwz mnrz pqwz prvw mnpqvwz stwxz prz sw wxz sty uvz mwx opstuvwx opstvx nrwz opuvwx nrvw sx rvwy nrwy mowx uz mnuwz opstuvwx puv stv ostuvwx pvz ouvwx stwz npuvx npqz npwy 100 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 73,744 70,563 169,463 80,394 93,163 135,913 43,006 19,250 93,781 8,450 70,763 34,569 71,675 200,038 89,113 73,144 118,675 46,656 59,769 42,306 145,100 132,075 56,013 68,681 70,413 256,200 137,650 121,544 60,456 20,194 168,394 9,550 69,763 88,013 97,663 34,438 Adjacent Room Loading (µg/ft2) 11,799 11,290 27,114 12,863 14,906 21,746 6,881 3,080 15,005 1,352 11,322 5,531 11,468 32,006 14,258 11,703 18,988 7,465 9,563 6,769 23,216 21,132 8,962 10,989 11,266 40,992 22,024 19,447 9,673 3,231 26,943 1,528 11,162 14,082 15,626 5,510 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.00502% 0.00500% 0.00495% 0.00485% 0.00469% 0.00443% 0.00440% 0.00433% 0.00420% 0.00418% 0.00413% 0.00400% 0.00398% 0.00392% 0.00390% 0.00372% 0.00354% 0.00348% 0.00332% 0.00326% 0.00321% 0.00308% 0.00307% 0.00303% 0.00299% 0.00297% 0.00295% 0.00294% 0.00285% 0.00282% 0.00281% 0.00278% 0.00278% 0.00274% 0.00274% 0.00261% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z pqrv ow mnqrz pqrvw nst nqrvw psuvw su mpqz uw msuwz uvwx mrwz opstuvwx ost oz optuvwx stx stuvwx pstv moswx nrvwy psvx rwyz pstvwx opsuvwx mnrvz npqv mrv pt mnpqwz tu nu osw mryz uy 101 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 29,906 152,081 145,431 55,888 202,100 110,225 72,519 48,488 92,788 96,931 46,875 134,138 125,425 168,550 176,938 107,219 176,113 99,744 60,488 127,731 102,531 33,306 174,125 82,363 29,894 133,831 123,850 14,400 35,350 125,931 246,431 121,044 98,650 100,731 105,156 45,775 Adjacent Room Loading (µg/ft2) 4,785 24,333 23,269 8,942 32,336 17,636 11,603 7,758 14,846 15,509 7,500 21,462 20,068 26,968 28,310 17,155 28,178 15,959 9,678 20,437 16,405 5,329 27,860 13,178 4,783 21,413 19,816 2,304 5,656 20,149 39,429 19,367 15,784 16,117 16,825 7,324 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.00257% 0.00256% 0.00254% 0.00248% 0.00242% 0.00238% 0.00234% 0.00230% 0.00229% 0.00229% 0.00228% 0.00220% 0.00219% 0.00216% 0.00216% 0.00214% 0.00212% 0.00205% 0.00204% 0.00203% 0.00198% 0.00197% 0.00197% 0.00196% 0.00196% 0.00195% 0.00192% 0.00192% 0.00191% 0.00190% 0.00190% 0.00188% 0.00186% 0.00185% 0.00185% 0.00180% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z uwx mnprwz mnprz stxz mnqryz optvx qrwz prwz nrv mqrv pstz mqryz nrvy nqrvwy mnrvwyz nprw mnqrwz mrvwy mptv nstuvwx nuvwx ptuvw mnqrvz ntw ps nqrz nstxz tw vxz nstvwx motvw mns nxz nvwx qrwyz pswz 102 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 65,869 38,344 131,556 54,094 12,413 136,219 129,263 287,431 36,044 161,900 24,856 208,750 213,413 85,413 92,063 81,363 128,375 151,894 99,625 51,319 77,469 111,881 81,188 35,856 49,844 217,606 51,350 28,644 79,550 181,906 105,231 66,638 113,944 148,356 156,744 29,425 Adjacent Room Loading (µg/ft2) 10,539 6,135 21,049 8,655 1,986 21,795 20,682 45,989 5,767 25,904 3,977 33,400 34,146 13,666 14,730 13,018 20,540 24,303 15,940 8,211 12,395 17,901 12,990 5,737 7,975 34,817 8,216 4,583 12,728 29,105 16,837 10,662 18,231 23,737 25,079 4,708 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.00180% 0.00179% 0.00174% 0.00170% 0.00170% 0.00165% 0.00161% 0.00156% 0.00155% 0.00154% 0.00153% 0.00147% 0.00144% 0.00137% 0.00136% 0.00132% 0.00131% 0.00118% 0.00111% 0.00110% 0.00110% 0.00108% 0.00106% 0.00104% 0.00094% 0.00089% 0.00089% 0.00086% 0.00083% 0.00081% 0.00077% 0.00077% 0.00074% 0.00072% 0.00072% 0.00065% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z qrz psuw mqrwy xy tv mqrvwy nqrv mnopsvx rwz nqrvy ptv mnqrwyz mnqrvwyz ns nsw os opsuvwx mnptwx mptuvwxz stvx mprz nprvw mswx twx mtuv mnoptuwx psx ptuw mprvw mnpqrz nprv mtux mpqrz npqrvw mnprvwz ptz 103 5/17/2007 DRAFT—DELIBERATIVE Table 5D-4: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area Contractor Target Housing that is a work area. RRP Task(s) Performed: Work Area Loading (µg/ft2) Adjacent Room Loading (µg/ft2) Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z 246,500 39,440 29.9% 0.00064% ptuvwx 84,963 13,594 29.9% 0.00062% pqrwz 208,025 33,284 29.9% 0.00061% mnotvwx 233,388 37,342 29.9% 0.00060% ptx 29,206 4,673 29.9% 0.00059% tx 111,363 17,818 29.9% 0.00058% mpqrw 116,025 18,564 29.9% 0.00055% mpqrvw 141,706 22,673 29.9% 0.00051% npqrv 188,556 30,169 29.9% 0.00049% mnpqrwz 193,219 30,915 29.9% 0.00048% mnpqrvwz 16,981 2,717 29.9% 0.00031% tz 22,113 3,538 29.9% 0.00027% sv 43,381 6,941 29.9% 0.00024% mt 42,188 6,750 29.9% 0.00007% tuy 45,050 7,208 29.9% 0.00006% tvy 20,863 3,338 29.9% 0.00005% tuvw Source: See Appendix 5B for a description of the development of work area and adjacent room loadings for specific RRP tasks. See Table 5D-3 for percent of home that is work area. Distribution of RRP tasks (probability of occurrence and type/combinations) estimated using the U.S. Census American Housing Survey and Property Owners and Managers Survey, with sample sizes of about 55 thousand and 16 thousand, respectively. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 104 5/17/2007 DRAFT—DELIBERATIVE 1 Table 5D-5: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the DIY Target Housing that is a Work Area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 36,475 12,700 32,638 75,113 114,788 20,163 6,650 121,438 69,113 39,675 111,588 46,325 107,750 81,763 5,994 43,125 72,313 147,425 56,638 151,263 76,150 49,175 52,375 127,431 9,231 87,813 19,350 95,275 127,488 157,913 39,288 154,075 59,838 42,469 45,669 163,906 78,963 32,863 120,781 Adjacent Room Loading (µg/ft2) 5,836 2,032 5,222 12,018 18,366 3,226 1,064 19,430 11,058 6,348 17,854 7,412 17,240 13,082 959 6,900 11,570 23,588 9,062 24,202 12,184 7,868 8,380 20,389 1,477 14,050 3,096 15,244 20,398 25,266 6,286 24,652 9,574 6,795 7,307 26,225 12,634 5,258 19,325 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 16.0% 29.9% 29.9% 5.6% 8.9% 29.9% 29.9% 29.9% 29.9% 3.3% 21.6% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 24.9% 19.3% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 22.3990% 8.2649% 6.8112% 5.5101% 3.9592% 3.1802% 3.1382% 2.2753% 2.1426% 1.9779% 1.6742% 1.4063% 1.3468% 1.1441% 1.1379% 1.1085% 1.1074% 1.1045% 1.0629% 1.0482% 1.0290% 0.9600% 0.7917% 0.7124% 0.6888% 0.6686% 0.6531% 0.5449% 0.5202% 0.5127% 0.4902% 0.4404% 0.4182% 0.3964% 0.3943% 0.3694% 0.3606% 0.3481% 0.3458% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z q p y n mn r w mnw qy m nq mw ny nw v qw my mny qr mnq mq pq mp mnvw z np pw nr mnp mnqw wy mnwy mr qv mv mnqvw mwy pr mnv 105 5/17/2007 DRAFT—DELIBERATIVE Table 5D-5: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the DIY Target Housing that is a Work Area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 144,225 52,319 183,900 52,800 82,800 21,344 59,025 141,600 118,238 63,913 18,694 108,788 134,950 134,138 160,069 21,931 26,813 41,869 153,419 66,488 38,631 34,044 75,763 157,256 67,669 96,313 12,644 82,144 196,544 130,669 84,956 88,850 84,344 189,894 163,963 124,288 114,400 190,550 85,256 89,275 Adjacent Room Loading (µg/ft2) 23,076 8,371 29,424 8,448 13,248 3,415 9,444 22,656 18,918 10,226 2,991 17,406 21,592 21,462 25,611 3,509 4,290 6,699 24,547 10,638 6,181 5,447 12,122 25,161 10,827 15,410 2,023 13,143 31,447 20,907 13,593 14,216 13,495 30,383 26,234 19,886 18,304 30,488 13,641 14,284 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 5.2% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.3399% 0.3292% 0.3274% 0.3194% 0.3077% 0.3037% 0.3030% 0.2929% 0.2921% 0.2901% 0.2718% 0.2662% 0.2647% 0.2636% 0.2601% 0.2537% 0.2452% 0.2270% 0.2220% 0.2038% 0.1920% 0.1853% 0.1809% 0.1784% 0.1739% 0.1729% 0.1691% 0.1685% 0.1533% 0.1526% 0.1499% 0.1490% 0.1485% 0.1480% 0.1463% 0.1454% 0.1405% 0.1391% 0.1370% 0.1348% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z nqy mvw mnqy ry mqw x mpw mnrw nqw o pv mqy mnr mnpw mnvwy pz rw yz mnvy mrw vy px qwy mnqv mwx mqr vw mqv mnqvwy mnwz mvwy mpq nz mnqvy mnpq npq nwy mnqwy ox qry 106 5/17/2007 DRAFT—DELIBERATIVE Table 5D-5: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the DIY Target Housing that is a Work Area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 92,475 88,794 115,438 140,131 63,288 1,775 65,019 127,825 75,106 101,925 131,750 133,481 48,906 127,913 167,588 72,538 78,306 40,038 21,125 94,463 219,394 81,756 147,594 25,344 27,338 45,706 81,106 58,369 49,119 69,906 45,281 91,906 116,981 151,725 156,656 55,825 114,781 176,606 69,338 40,694 Adjacent Room Loading (µg/ft2) 14,796 14,207 18,470 22,421 10,126 284 10,403 20,452 12,017 16,308 21,080 21,357 7,825 20,466 26,814 11,606 12,529 6,406 3,380 15,114 35,103 13,081 23,615 4,055 4,374 7,313 12,977 9,339 7,859 11,185 7,245 14,705 18,717 24,276 25,065 8,932 18,365 28,257 11,094 6,511 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 10.4% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.1275% 0.1274% 0.1129% 0.1122% 0.1092% 0.1053% 0.1034% 0.1031% 0.1004% 0.1001% 0.0962% 0.0925% 0.0925% 0.0907% 0.0887% 0.0863% 0.0852% 0.0849% 0.0844% 0.0841% 0.0816% 0.0813% 0.0811% 0.0804% 0.0794% 0.0782% 0.0764% 0.0746% 0.0742% 0.0739% 0.0730% 0.0720% 0.0705% 0.0694% 0.0688% 0.0686% 0.0682% 0.0681% 0.0676% 0.0649% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z mry mqvw mqwy mnpvw qrw u mpvw o qvy nrw nqr mnpv mz nry mnry mpr mvy pvx puw npw mnopwx qvwy mnrvw pvw vx qz nv mpv qvw ov vwy owx nyz nop mnyz pqw mqvy mnpqvw pqr pwx 107 5/17/2007 DRAFT—DELIBERATIVE Table 5D-5: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the DIY Target Housing that is a Work Area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 100,456 121,431 78,344 169,956 171,425 120,394 26,156 123,575 102,963 184,069 62,631 95,500 83,538 170,613 174,238 120,819 117,581 28,581 150,875 113,744 55,169 14,475 163,250 82,144 164,388 39,513 107,975 163,306 128,950 124,019 178,075 8,425 61,019 59,450 110,600 144,556 157,256 124,931 101,494 103,106 Adjacent Room Loading (µg/ft2) 16,073 19,429 12,535 27,193 27,428 19,263 4,185 19,772 16,474 29,451 10,021 15,280 13,366 27,298 27,878 19,331 18,813 4,573 24,140 18,199 8,827 2,316 26,120 13,143 26,302 6,322 17,276 26,129 20,632 19,843 28,492 1,348 9,763 9,512 17,696 23,129 25,161 19,989 16,239 16,497 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.0642% 0.0632% 0.0627% 0.0617% 0.0615% 0.0574% 0.0562% 0.0546% 0.0545% 0.0541% 0.0536% 0.0536% 0.0533% 0.0526% 0.0508% 0.0507% 0.0502% 0.0484% 0.0476% 0.0446% 0.0445% 0.0442% 0.0438% 0.0435% 0.0435% 0.0434% 0.0433% 0.0429% 0.0428% 0.0417% 0.0412% 0.0411% 0.0401% 0.0398% 0.0395% 0.0391% 0.0389% 0.0382% 0.0381% 0.0380% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z npvw mqvwy qyz mnpqv mnqr nvwy rv npuvwx mqrw mnqrvw qrv mpqw nuw mnpqw mnrwy nqz nqv pwz nqwy nvy pqv pu mnpuvwx mpuwx nqry prw npr mnwyz mqry mnz mnqrw uw mx rwy opvwx mnuwx mnpuwx mox mpqvw nwx 108 5/17/2007 DRAFT—DELIBERATIVE Table 5D-5: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the DIY Target Housing that is a Work Area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 87,756 269,731 147,650 51,700 173,581 90,994 204,063 97,956 94,844 69,444 27,994 455,788 170,294 188,844 121,800 123,213 169,300 61,819 180,231 215,925 193,131 7,769 214,188 214,463 70,563 58,406 103,588 186,850 79,325 240,200 122,975 285,081 45,338 137,906 160,494 88,138 91,838 136,131 15,881 131,581 Adjacent Room Loading (µg/ft2) 14,041 43,157 23,624 8,272 27,773 14,559 32,650 15,673 15,175 11,111 4,479 72,926 27,247 30,215 19,488 19,714 27,088 9,891 28,837 34,548 30,901 1,243 34,270 34,314 11,290 9,345 16,574 29,896 12,692 38,432 19,676 45,613 7,254 22,065 25,679 14,102 14,694 21,781 2,541 21,053 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 8.5% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.0372% 0.0370% 0.0369% 0.0366% 0.0366% 0.0364% 0.0361% 0.0357% 0.0356% 0.0339% 0.0334% 0.0331% 0.0329% 0.0328% 0.0327% 0.0324% 0.0317% 0.0316% 0.0315% 0.0311% 0.0310% 0.0308% 0.0305% 0.0304% 0.0303% 0.0302% 0.0298% 0.0296% 0.0296% 0.0295% 0.0290% 0.0288% 0.0288% 0.0279% 0.0278% 0.0278% 0.0276% 0.0276% 0.0275% 0.0272% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z nvw mnopuvw mnpr qvz mnrvy nwz mnqry opx mpqv muwx wx moy opuwx mox npvwx mnuw mnvwyz pqvw mnrvwy mnowxz mnqyz uv mopvwx mnouvwx ow pqz mo mopw pvwxy opuvwx nvyz mnopuwx py mnux mnqz mpuvwx opvz mnx wz mowx 109 5/17/2007 DRAFT—DELIBERATIVE Table 5D-5: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the DIY Target Housing that is a Work Area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 109,156 154,944 150,831 47,113 29,394 109,013 156,869 353,213 54,150 272,381 130,938 124,231 97,044 48,519 15,225 210,056 155,481 124,713 178,700 69,281 136,719 55,556 99,125 91,031 144,450 95,925 138,400 84,994 205,775 210,713 216,706 153,025 154,300 156,600 32,806 161,788 98,469 164,550 35,819 153,644 Adjacent Room Loading (µg/ft2) 17,465 24,791 24,133 7,538 4,703 17,442 25,099 56,514 8,664 43,581 20,950 19,877 15,527 7,763 2,436 33,609 24,877 19,954 28,592 11,085 21,875 8,889 15,860 14,565 23,112 15,348 22,144 13,599 32,924 33,714 34,673 24,484 24,688 25,056 5,249 25,886 15,755 26,328 5,731 24,583 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 14.2% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.0270% 0.0268% 0.0267% 0.0264% 0.0264% 0.0257% 0.0257% 0.0255% 0.0255% 0.0254% 0.0253% 0.0249% 0.0248% 0.0245% 0.0245% 0.0244% 0.0241% 0.0240% 0.0237% 0.0233% 0.0230% 0.0224% 0.0224% 0.0223% 0.0222% 0.0220% 0.0219% 0.0217% 0.0212% 0.0210% 0.0210% 0.0204% 0.0204% 0.0200% 0.0199% 0.0194% 0.0194% 0.0192% 0.0191% 0.0191% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z npx opuvw mnrwz puy rz mpqr nqvwy mnopvwx mpu mnouwx npqw nqvw npz wyz vz mnqrvy mnpwx mopuw mno qrvw mnpz mwz mrwy opuvw npqr qrwy nqrw qwyz mnqvwyz mnqrwy mnqrvwy ost mnprw mnpuvx rvw opstvxy mrvy mostuvwx pux mnprv 110 5/17/2007 DRAFT—DELIBERATIVE Table 5D-5: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the DIY Target Housing that is a Work Area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 140,944 177,250 118,388 184,125 289,638 130,013 160,294 76,613 72,481 93,806 41,450 134,944 60,800 81,544 17,450 78,531 153,456 143,369 79,188 71,863 52,356 203,319 48,100 190,119 72,338 53,981 117,906 118,019 167,144 145,275 30,150 86,363 36,350 307,088 139,025 162,144 166,488 196,769 136,588 149,363 Adjacent Room Loading (µg/ft2) 22,551 28,360 18,942 29,460 46,342 20,802 25,647 12,258 11,597 15,009 6,632 21,591 9,728 13,047 2,792 12,565 24,553 22,939 12,670 11,498 8,377 32,531 7,696 30,419 11,574 8,637 18,865 18,883 26,743 23,244 4,824 13,818 5,816 49,134 22,244 25,943 26,638 31,483 21,854 23,898 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 10.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.0182% 0.0169% 0.0169% 0.0168% 0.0168% 0.0166% 0.0165% 0.0160% 0.0157% 0.0157% 0.0157% 0.0155% 0.0154% 0.0154% 0.0152% 0.0151% 0.0151% 0.0144% 0.0141% 0.0139% 0.0131% 0.0130% 0.0129% 0.0128% 0.0127% 0.0125% 0.0124% 0.0124% 0.0119% 0.0119% 0.0118% 0.0118% 0.0118% 0.0117% 0.0113% 0.0111% 0.0111% 0.0110% 0.0107% 0.0106% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z mnrv mopstuvwx npxz mnpqr ovwx mnvz mnprvw op mrvw npv mu mqrvy mpuw myz s mprv nqyz mnpwz mprw qrvz qwz mopux muw mnpqrv ouw xy npsvw mqyz mnqwz os ps mpvwx puvwz osvwx no noux mnqvz mnpqrvw opvxy mnpvwz 111 5/17/2007 DRAFT—DELIBERATIVE Table 5D-5: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the DIY Target Housing that is a Work Area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 139,350 75,988 173,194 103,038 115,006 104,506 95,269 83,263 154,825 81,363 152,050 190,775 101,919 35,388 20,469 219,075 20,450 108,956 62,031 150,219 177,419 136,931 38,856 112,013 112,375 199,781 35,763 193,119 58,794 103,950 179,063 122,556 7,750 200,044 183,469 158,375 176,819 46,688 93,025 144,181 Adjacent Room Loading (µg/ft2) 22,296 12,158 27,711 16,486 18,401 16,721 15,243 13,322 24,772 13,018 24,328 30,524 16,307 5,662 3,275 35,052 3,272 17,433 9,925 24,035 28,387 21,909 6,217 17,922 17,980 31,965 5,722 30,899 9,407 16,632 28,650 19,609 1,240 32,007 29,355 25,340 28,291 7,470 14,884 23,069 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.0106% 0.0104% 0.0104% 0.0102% 0.0101% 0.0098% 0.0098% 0.0095% 0.0095% 0.0095% 0.0093% 0.0093% 0.0093% 0.0093% 0.0092% 0.0090% 0.0090% 0.0090% 0.0090% 0.0086% 0.0086% 0.0086% 0.0086% 0.0083% 0.0083% 0.0083% 0.0081% 0.0080% 0.0080% 0.0079% 0.0079% 0.0078% 0.0077% 0.0076% 0.0075% 0.0075% 0.0075% 0.0074% 0.0074% 0.0073% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z mouvwx pqrw mnpqz npvz mpqrv nrz qrvy opw mnpvx os mopuvwx mnpqrw qrvwy rvz puv ovx pt mqrvw ryz nqvy mnqrv npqvw prv mouw opuvwx mnqwyz uvwx mnouvw rvy opvx nopvx mnuv t mnox mnrwyz nopw mnryz pvwx ouvx mnrz 112 5/17/2007 DRAFT—DELIBERATIVE Table 5D-5: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the DIY Target Housing that is a Work Area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 65,056 45,506 29,094 185,838 95,563 42,094 66,119 48,463 33,988 77,194 111,156 107,919 61,606 65,444 129,206 134,563 75,419 47,788 38,794 69,375 184,694 80,369 148,831 163,588 133,519 164,056 75,331 180,656 303,300 81,981 47,425 116,925 228,625 144,394 154,438 98,081 75,719 181,038 242,613 137,631 Adjacent Room Loading (µg/ft2) 10,409 7,281 4,655 29,734 15,290 6,735 10,579 7,754 5,438 12,351 17,785 17,267 9,857 10,471 20,673 21,530 12,067 7,646 6,207 11,100 29,551 12,859 23,813 26,174 21,363 26,249 12,053 28,905 48,528 13,117 7,588 18,708 36,580 23,103 24,710 15,693 12,115 28,966 38,818 22,021 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 19.4% 29.9% Chapter 5 0.0073% 0.0072% 0.0071% 0.0071% 0.0070% 0.0068% 0.0068% 0.0067% 0.0064% 0.0064% 0.0064% 0.0062% 0.0062% 0.0062% 0.0060% 0.0060% 0.0058% 0.0058% 0.0057% 0.0057% 0.0054% 0.0052% 0.0051% 0.0051% 0.0050% 0.0050% 0.0049% 0.0049% 0.0048% 0.0048% 0.0048% 0.0047% 0.0044% 0.0044% 0.0042% 0.0041% 0.0039% 0.0035% 0.0035% 0.0034% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z pqwz prvw tx mnpqvwz npt prz mptv puvwx vwx mtuwx nrwz nrvw mpz rvwy mnuvw nrwy mtwx ptvx sx mpuvz nopstuv mpwx mnpx ouvwx npqz mnpvxz pqrv mnqrz mopstvwx pqrvw mt npuvx mnopwxz nqrvw npvwxy mpqz mrwz osuvwx mno mopx 113 5/17/2007 DRAFT—DELIBERATIVE Table 5D-5: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the DIY Target Housing that is a Work Area. RRP Task(s) Performed: Work Area Loading (µg/ft2) 96,456 27,119 140,556 68,681 159,494 104,881 150,175 130,281 65,831 179,844 142,781 101,706 278,100 163,531 156,881 213,294 72,519 24,100 48,744 101,269 102,306 138,181 133,906 177,031 189,463 114,625 187,306 105,119 63,775 186,650 140,981 105,156 65,869 135,600 57,125 141,594 137,744 36,044 170,381 219,944 Adjacent Room Loading (µg/ft2) 15,433 4,339 22,489 10,989 25,519 16,781 24,028 20,845 10,533 28,775 22,845 16,273 44,496 26,165 25,101 34,127 11,603 3,856 7,799 16,203 16,369 22,109 21,425 28,325 30,314 18,340 29,969 16,819 10,204 29,864 22,557 16,825 10,539 21,696 9,140 22,655 22,039 5,767 27,261 35,191 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% 29.9% Chapter 5 0.0033% 0.0032% 0.0030% 0.0030% 0.0030% 0.0029% 0.0029% 0.0029% 0.0028% 0.0028% 0.0028% 0.0027% 0.0027% 0.0025% 0.0025% 0.0024% 0.0023% 0.0023% 0.0023% 0.0023% 0.0023% 0.0022% 0.0022% 0.0021% 0.0021% 0.0021% 0.0021% 0.0020% 0.0020% 0.0019% 0.0019% 0.0018% 0.0018% 0.0017% 0.0017% 0.0016% 0.0016% 0.0015% 0.0015% 0.0014% Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z nx puvw nrvwy rwyz nopuv nuwx mnrvz npqv mrv mnpqwz mnwx mryz mopvwx mnprwz mnprz mnqryz qrwz sw prwz nrv mqrv mqryz nrvy nqrvwy mnrvwyz nprw mnqrwz mrvwy msw mnqrvz nqrz qrwyz qrz mqrwy ms mqrvwy nqrv rwz nqrvy mnqrwyz 114 5/17/2007 DRAFT—DELIBERATIVE Table 5D-5: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the DIY Target Housing that is a Work Area. RRP Task(s) Performed: Work Area Loading (µg/ft2) Adjacent Room Loading (µg/ft2) Probability of Percent of Home that Occurrence of RRP is Work Area Task(s) Bathroom=m, Kitchen=n, Addition=o, Window/Door=p, Interior Painting=q, HVAC=r, Heating=s, Cooling=t, internal water pipes=u, wiring=v plumbing fixtures=w, paneling/ceiling=x, Other=y, Security=z 225,938 36,150 29.9% 0.0014% mnqrvwyz 162,188 25,950 29.9% 0.0013% nptvwxy 23,119 3,699 29.9% 0.0013% ux 171,000 27,360 29.9% 0.0013% mnptuvwx 25,200 4,032 29.9% 0.0012% st 81,769 13,083 29.9% 0.0011% mprz 120,619 19,299 29.9% 0.0011% nprvw 82,863 13,258 29.9% 0.0010% nt 112,631 18,021 29.9% 0.0010% ntuwx 85,181 13,629 29.9% 0.0008% mprvw 193,356 30,937 29.9% 0.0008% mnpqrz 54,075 8,652 29.9% 0.0008% mtw 113,969 18,235 29.9% 0.0008% nprv 118,244 18,919 29.9% 0.0007% mpqrz 157,094 25,135 29.9% 0.0007% npqrvw 169,525 27,124 29.9% 0.0007% mnprvwz 85,219 13,635 29.9% 0.0006% pqrwz 115,663 18,506 29.9% 0.0006% mpqrw 121,656 19,465 29.9% 0.0005% mpqrvw 150,444 24,071 29.9% 0.0005% npqrv 200,006 32,001 29.9% 0.0005% mnpqrwz 206,000 32,960 29.9% 0.0005% mnpqrvwz 152,306 24,369 29.9% 0.0003% mntuwx Source: See Appendix 5B for a description of the development of work area and adjacent room loadings for specific RRP tasks. See Table 5D-3 for percent of home that is work area. Distribution of RRP tasks (probability of occurrence and type/combinations) estimated using the U.S. Census’ American Housing Survey, which has a sample size of about 55 thousand. 1 2 3 4 5 6 7 8 9 10 11 12 13 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 115 5/17/2007 DRAFT—DELIBERATIVE 1 2 Table 5D-6: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the Center that is a Work Area. Percent of Work Area Adjacent Room Center that is Probability of Work Area Loading (ug/ft2) Loading (ug/ft2) Occurrence Event Type 351 56 7.40% 11.70% Wall Disturbing Event 351 56 7.90% 5.30% Wall Disturbing Event 351 56 8.20% 6.20% Wall Disturbing Event 351 56 8.90% 5.10% Wall Disturbing Event 351 56 9.10% 2.70% Wall Disturbing Event 351 56 8.80% 3.40% Wall Disturbing Event 351 56 5.10% 47.90% Wall Disturbing Event 12,444 1,991 1.50% 0.30% Window/Door Replacement 12,444 1,991 2.00% 0.20% Window/Door Replacement 12,444 1,991 2.70% 0.20% Window/Door Replacement 12,444 1,991 2.80% 0.20% Window/Door Replacement 12,444 1,991 3.50% 0.30% Window/Door Replacement 12,444 1,991 3.60% 0.10% Window/Door Replacement 12,444 1,991 8.20% 0.20% Window/Door Replacement 12,444 1,991 8.90% 0.10% Window/Door Replacement 12,444 1,991 9.30% 0.30% Window/Door Replacement 12,444 1,991 13.90% 0.20% Window/Door Replacement 12,444 1,991 14.00% 0.20% Window/Door Replacement 12,444 1,991 15.00% 0.20% Window/Door Replacement 12,444 1,991 17.90% 0.40% Window/Door Replacement 12,444 1,991 18.70% 0.40% Window/Door Replacement 12,444 1,991 25.80% 0.40% Window/Door Replacement 12,444 1,991 26.70% 0.40% Window/Door Replacement 12,444 1,991 31.00% 0.40% Window/Door Replacement 12,444 1,991 33.00% 0.60% Window/Door Replacement 12,444 1,991 33.90% 0.60% Window/Door Replacement 12,444 1,991 35.20% 0.10% Window/Door Replacement 12,444 1,991 35.80% 0.50% Window/Door Replacement 12,444 1,991 24.50% 0.10% Window/Door Replacement 36,475 5,836 10.30% 0.50% Interior Painting 36,475 5,836 24.30% 0.60% Interior Painting 36,475 5,836 24.40% 0.30% Interior Painting 36,475 5,836 29.90% 0.20% Interior Painting 36,475 5,836 31.40% 0.30% Interior Painting 36,475 5,836 34.40% 0.40% Interior Painting 36,475 5,836 42.50% 0.30% Interior Painting 36,475 5,836 48.70% 0.40% Interior Painting 36,475 5,836 48.10% 0.70% Interior Painting 36,475 5,836 44.50% 0.20% Interior Painting 36,475 5,836 42.80% 0.50% Interior Painting 36,475 5,836 41.70% 0.40% Interior Painting 36,475 5,836 32.50% 0.60% Interior Painting 36,475 5,836 15.00% 0.40% Interior Painting 36,475 5,836 100.00% 5.50% Interior Painting §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 116 5/17/2007 DRAFT—DELIBERATIVE Source: See Appendix 5B for development of lead loadings. Distribution of tasks estimated using the U.S. Department of Housing and Urban Development’s First National Environmental Health Survey of Child Care Centers, with a sample size of 168 centers. 1 2 3 4 Table 5D-7: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the Kindergarten that is a Work Area Percent of Work Area Adjacent Room Center that is Probability of Loading (ug/ft2) Loading (ug/ft2) Work Area Occurrence Event Type 351.05 56.168 6.80% 11.70% Wall Disturbing 351.05 56.168 7.30% 5.30% Wall Disturbing 351.05 56.168 7.60% 6.20% Wall Disturbing 351.05 56.168 8.30% 5.10% Wall Disturbing 351.05 56.168 8.50% 2.70% Wall Disturbing 351.05 56.168 8.20% 3.40% Wall Disturbing 351.05 56.168 4.50% 47.90% Wall Disturbing 12443.75 1991 1.50% 0.30% Window/Door Replacement 12443.75 1991 2.00% 0.20% Window/Door Replacement 12443.75 1991 2.70% 0.20% Window/Door Replacement 12443.75 1991 2.80% 0.20% Window/Door Replacement 12443.75 1991 3.50% 0.30% Window/Door Replacement 12443.75 1991 3.60% 0.10% Window/Door Replacement 12443.75 1991 8.20% 0.20% Window/Door Replacement 12443.75 1991 8.90% 0.10% Window/Door Replacement 12443.75 1991 9.30% 0.30% Window/Door Replacement 12443.75 1991 13.90% 0.20% Window/Door Replacement 12443.75 1991 14.00% 0.20% Window/Door Replacement 12443.75 1991 15.00% 0.20% Window/Door Replacement 12443.75 1991 17.90% 0.40% Window/Door Replacement 12443.75 1991 18.70% 0.40% Window/Door Replacement 12443.75 1991 25.80% 0.40% Window/Door Replacement 12443.75 1991 26.70% 0.40% Window/Door Replacement 12443.75 1991 31.00% 0.40% Window/Door Replacement 12443.75 1991 33.00% 0.60% Window/Door Replacement 12443.75 1991 33.90% 0.60% Window/Door Replacement 12443.75 1991 35.20% 0.10% Window/Door Replacement 12443.75 1991 35.80% 0.50% Window/Door Replacement 12443.75 1991 24.50% 0.10% Window/Door Replacement 36475 5836 10.30% 0.50% Interior Painting 36475 5836 24.30% 0.60% Interior Painting 36475 5836 24.40% 0.30% Interior Painting 36475 5836 29.90% 0.20% Interior Painting 36475 5836 31.40% 0.30% Interior Painting 36475 5836 34.40% 0.40% Interior Painting 36475 5836 42.50% 0.30% Interior Painting 36475 5836 48.70% 0.40% Interior Painting 36475 5836 48.10% 0.70% Interior Painting 36475 5836 44.50% 0.20% Interior Painting 36475 5836 42.80% 0.50% Interior Painting 36475 5836 41.70% 0.40% Interior Painting §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 117 5/17/2007 DRAFT—DELIBERATIVE 36475 36475 36475 5836 5836 5836 32.50% 15.00% 100.00% 0.60% 0.40% 5.50% Interior Painting Interior Painting Interior Painting Source: See Appendix 5B for development of lead loadings. Distribution of tasks estimated using the U.S. Department of Housing and Urban Development’s First National Environmental Health Survey of Child Care Centers, with a sample size of 168 centers. 1 2 3 4 Table 5D-8: Custom Distribution Values for Input Parameters: Initial Lead Dust Level in Work Area After Indoor Work, and Percentage of Area of the PKKG that is a Work Area Percent of Work Area Adjacent Room Center that is Probability of Loading (ug/ft2) Loading (ug/ft2) Work Area Occurrence Event Type 351.05 56.168 4.90% 11.70% Wall Disturbing 351.05 56.168 5.40% 5.30% Wall Disturbing 351.05 56.168 5.60% 6.20% Wall Disturbing 351.05 56.168 6.40% 5.10% Wall Disturbing 351.05 56.168 6.60% 2.70% Wall Disturbing 351.05 56.168 6.30% 3.40% Wall Disturbing 351.05 56.168 2.50% 47.90% Wall Disturbing 12443.75 1991 1.50% 0.30% Window/Door Replacement 12443.75 1991 2.00% 0.20% Window/Door Replacement 12443.75 1991 2.70% 0.20% Window/Door Replacement 12443.75 1991 2.80% 0.20% Window/Door Replacement 12443.75 1991 3.50% 0.30% Window/Door Replacement 12443.75 1991 3.60% 0.10% Window/Door Replacement 12443.75 1991 8.20% 0.20% Window/Door Replacement 12443.75 1991 8.90% 0.10% Window/Door Replacement 12443.75 1991 9.30% 0.30% Window/Door Replacement 12443.75 1991 13.90% 0.20% Window/Door Replacement 12443.75 1991 14.00% 0.20% Window/Door Replacement 12443.75 1991 15.00% 0.20% Window/Door Replacement 12443.75 1991 17.90% 0.40% Window/Door Replacement 12443.75 1991 18.70% 0.40% Window/Door Replacement 12443.75 1991 25.80% 0.40% Window/Door Replacement 12443.75 1991 26.70% 0.40% Window/Door Replacement 12443.75 1991 31.00% 0.40% Window/Door Replacement 12443.75 1991 33.00% 0.60% Window/Door Replacement 12443.75 1991 33.90% 0.60% Window/Door Replacement 12443.75 1991 35.20% 0.10% Window/Door Replacement 12443.75 1991 35.80% 0.50% Window/Door Replacement 12443.75 1991 24.50% 0.10% Window/Door Replacement 36475 5836 10.30% 0.50% Interior Painting 36475 5836 24.30% 0.60% Interior Painting 36475 5836 24.40% 0.30% Interior Painting 36475 5836 29.90% 0.20% Interior Painting 36475 5836 31.40% 0.30% Interior Painting 36475 5836 34.40% 0.40% Interior Painting 36475 5836 42.50% 0.30% Interior Painting 36475 5836 48.70% 0.40% Interior Painting 36475 5836 48.10% 0.70% Interior Painting §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 118 5/17/2007 DRAFT—DELIBERATIVE 36475 36475 36475 36475 36475 36475 5836 5836 5836 5836 5836 5836 44.50% 42.80% 41.70% 32.50% 15.00% 100.00% 0.20% 0.50% 0.40% 0.60% 0.40% 5.50% Interior Painting Interior Painting Interior Painting Interior Painting Interior Painting Interior Painting Source: See Appendix 5B for development of lead loadings. Distribution of tasks estimated using the U.S. Department of Housing and Urban Development’s First National Environmental Health Survey of Child Care Centers, with a sample size of 168 centers. 1 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 119 5/17/2007 DRAFT—DELIBERATIVE 1 Appendix 5E: Monte Carlo Analysis Inputs 2 Table 5E-1: Monte Carlo Analysis Inputs Category Units Value or Distribution Reference or Report Section Population Assumptions Single value for each age (0-5) in Number of children by age in COFs Number of People centers, KG, covered by the rule. PKKG, and Target Housing Fraction of child awake time in COF by Single value for Unit-less fraction age. each age Single value for Hours per week in COF Hours/week each age Single value for Duration of school year Weeks/year each age and school setting Chapter 2 Section 5.3.1 Section 5.3.1 Section 5.3.1 Lead Loading Assumptions Work Area Custom Distribution Unit-less fraction (Range = 1.5% to 100%) Custom Percentage of area of the commercial or Distribution public building that is a work area Unit-less fraction (Range = 1.5% to (Kindergartens) 100%) Custom Percentage of area of the commercial or Distribution public building that is a work area Unit-less fraction (Range = 1.5% to (PKKG) 100%) Custom Percentage of area of the TH-COF that is Distribution Unit-less fraction a work area (Range = 3% to 30%) Percentage of area of the commercial or public building that is a work area (Centers) Appendix 5C, 5D Appendix 5C, 5D Appendix 5C, 5D Appendix 5D Work Area Load Initial lead dust level outside of work area and adjacent room for interior work (all COFs). Initial lead dust level in work area after indoor work—COFs in public or commercial buildings (Centers) §402(c) COF Economic Analysis Draft - Do Not Cite or Quote µg/ft2 0 Section 5.2.2 µg/ft2 Custom Distribution (Range = 351 to 36,475) Appendix 5C, 5D Chapter 5 120 5/17/2007 DRAFT—DELIBERATIVE Table 5E-1: Monte Carlo Analysis Inputs Category Units Initial lead dust level in work area after indoor work—COFs in public or commercial buildings (Kindergartens) µg/ft2 Initial lead dust level in work area after indoor work—COFs in public or commercial buildings (PKKG)) µg/ft2 Initial lead dust level in work area after indoor work—TH-contractor µg/ft2 Initial lead dust level in work area after indoor work—TH-DIY µg/ft2 Value or Distribution Custom Distribution (Range = 351 to 36,475) Custom Distribution (Range = 351 to 36,475) Custom Distribution (Range = 1,800 to 409,031) Custom Distribution (Range 1,775 to 455,788) Reference or Report Section Appendix 5C, 5D Appendix 5C, 5D Appendix 5B, 5D Appendix 5B, 5D Adjacent Area Load Percentage of initial dust loading in rooms adjacent to the work area—COF Unit-less fraction in public or commercial buildings (same for Centers, Kindergartens, PKKG) Percentage of initial dust loading in Unit-less fraction rooms adjacent to the work area—TH COF (same for TH-contractor and DIY) Initial lead dust level in rooms adjacent to work area after indoor work—COFs µg/ft2 in public or commercial buildings (Centers) Initial lead dust level in rooms adjacent to work area after indoor work—COFs µg/ft2 in public or commercial buildings (Kindergartens) Initial lead dust level in rooms adjacent to work area after indoor work—COFs µg/ft2 in public or commercial buildings (PKKG) Initial lead dust level in adjacent area after indoor work—TH-contractor µg/ft2 Initial lead dust level in adjacent area after indoor work—TH-DIY µg/ft2 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 16% Appendix 5C, 5D 16% Derived from U.S. EPA, 1997a Custom Distribution (Range = 56 to 5,836) Custom Distribution (Range = 56 to 5,836) Custom Distribution (Range = 56 to 5,836) Custom Distribution (Range =288 to 65,445) Custom Distribution (Range = 284 to 72,926) Appendix 5C, 5D Appendix 5C, 5D Appendix 5C, 5D Appendix 5B, 5D Appendix 5B, 5D 121 5/17/2007 DRAFT—DELIBERATIVE Table 5E-1: Monte Carlo Analysis Inputs Category Lead dust level indoors after contractor cleaning with rule Units µg/ft2 Value or Reference or Distribution Report Section 40 maximum or Toxic Substances concentration Control Act resulting from (TSCA) Section cleaning, 403 Rule whichever is lower Soil Loadings Soil Lead Concentration Outdoors without Rule-TH COF mg/kg Soil Lead Concentration Outdoors without Rule- COFs in public or commercial buildings mg/kg Soil Lead Concentration Outdoors with Rule—TH COF Soil Lead Concentration Outdoors with Rule—COF in public or commercial buildings Custom Appendix 5D, Distribution UIUC, 2002; HUD, (Range = 511 to 2003 1,166) Custom Appendix 5D, Distribution UIUC, 2002; HUD, (Range = 103 to 2003 783) mg/kg 490 HUD, 2000 mg/kg 74 HUD, 2003 Indoor Baseline Cleaning Assumptions –Scenario 1 Percentage of Floor CarpetedCommercial or Public Buildings Percentage of Floors Carpeted-Target Housing Unit-less fraction 0.338 Unit-less fraction 0.36 Uniform Distribution (Range = 14% 37%) Uniform Distribution Unit-less fraction (Range = 94.8% 98.5%) Uniform Distribution Unit-less fraction (Range = 14% 37%) Uniform Distribution Unit-less fraction (Range = 94.8% 98.5%) Custom Distribution Integer (Range from 26 to 104) Initial Cleaning Efficiency on Carpet (by contractor or in-house staff performing Unit-less fraction RRP) Initial Cleaning Efficiency on NonCarpet (by contractor or in-house staff performing RRP) Routine COF Cleaning Efficiency on Carpet Routine COF Cleaning Efficiency on Non-Carpet Frequency of TH-COF Cleaning per year §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 Derived from HUD, 2003 Derived from HUD, 2000 Yin, 2002 CETL, 2001 Yin, 2002 CETL, 2001 Simcox, 1995 122 5/17/2007 DRAFT—DELIBERATIVE Table 5E-1: Monte Carlo Analysis Inputs Category Frequency of COFs in Commercial or Public Buildings Cleaning per year Value or Distribution 260 (or daily, 5 days/week) Reference or Report Section Unit-less fraction 15 % Section 5.2.4 Unit-less fraction Same as first carpet cleaning efficiency Section 5.2.4 Unit-less fraction 70 % Section 5.2.4 Unit-less fraction 10 % Section 5.2.4 Units Integer HUD 2003 Indoor Baseline Cleaning Assumptions –Scenario 2 Initial cleaning Percent of RRP Events where Dust Reductions beyond those of Initial Cleaning in Work Area are Achieved (by contractor or in-house staff performing RRP) Additional Cleaning Efficiency on Carpet where Dust Reductions beyond those of Initial Cleaning are Achieved (by contractor or in-house staff performing RRP) Additional Cleaning Efficiency on NonCarpet where Dust Reductions beyond those of Initial Cleaning are Achieved (by contractor or in-house staff performing RRP) Adjacent room cleaning Percent of RRP Events where Adjacent Room is Cleaned (by contractor or inhouse staff performing RRP) Percent of RRP Events where Dust Reductions beyond those of Initial Cleaning in the Adjacent Room are Achieved (by contractor or in-house staff performing RRP) 15 % of the 10% of Adjacent Rooms Unit-less fraction that are Cleaned by Contractors (= 1.5%) Section 5.2.4 Unit-less fraction 15 % Section 5.2.4 Unit-less fraction Same as first carpet cleaning efficiency Section 5.2.4 Unit-less fraction 70 % Section 5.2.4 Routine Cleaning Percent of RRP Events where COF Achieve Dust Reductions beyond those of Routine Cleaning Additional COF Cleaning Efficiency on Carpet where COFs Achieve Dust Reductions beyond those of Routine Cleaning Additional COF Cleaning Efficiency on Non-Carpet where COFs Achieve Dust Reductions beyond those of Routine Cleaning §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 123 5/17/2007 DRAFT—DELIBERATIVE Table 5E-1: Monte Carlo Analysis Inputs Units Value or Distribution Reference or Report Section IEUBK Lead Dust to Blood Lead Relationship µg/dL per µg/ft2 Linear regression based on IEUBK model results. U.S. EPA, 2005 Empirical Lead to Blood Lead Relationship µg/dL per µg/ft2 See Section 5.3.1 U.S. EPA, 2000a IQ Reduction Parameter (peak) IQ points per 1 µg/dL IQ Reduction Parameter (average) IQ points per 1 µg/dL Category Child Dose-Response Assumptions Economic Assumptions Discount Rate IQ cost per point lost Unit-less fraction Dollars Normal Distribution (mean = 0.25, s.e. = 0.04) Normal Distribution (mean = 0.46, s.e. = 0.15) 0.03 and 0.07 12,953 Schwartz, 1994 Canfield, 2003 Salkever, 1995 1 §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 124 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 References Agency for Toxic Substances and Disease Registry (ATSDR). 1999. Toxicological profile for lead. U.S. Department of Health and Human Services, Public Health Service. http://www.atsdr.cdc.gov/toxprofiles/tp13.html Akesson, A.; Lundh, T.; Vahter, M.; Bjellerup, P.; Lidfeldt, J.; Nerbrand, C.; Goran, S.; Strömberg, U.; Skerfving, S. (2005) Tubular and glomerular kidney effects in Swedish women with low environmental cadmium exposure. Environ. Health Perspect. 113: 1627-1631. Battelle. 2005. Using the IEUBK Model to Predict Geometric Mean Blood-Lead Concentration in Children Aged 1-2 Years as a Function of Hand Dust-Lead Concentration. Bellinger, D. C., et al. 1992. Low-Level Lead Exposure, Intelligence and Academic Achievement: A Long-term Follow-up Study. Pediatrics, 90(6): 855-861. Canadian Mortgage and Housing Corporation (CMHC). 1992. Effectiveness of Clean-up Techniques for Leaded Paint Dust. Canadian Mortgage and Housing Corporation, Research Division. Research Highlights Technical; Series Number 92-203. http://www.cmhc.ca/publications/en/rh-pr/tech/92203.pdf Canfield, R. L., et al. 2003. Intellectual Impairment in Children with Blood Lead Concentrations below 10µg per Deciliter. New England Journal of Medicine, 348(16): 1517-1526. Centers for Disease Control (CDC). 1997. Children with Elevated Blood Lead Levels Attributed to Home Renovation and Remodeling Activities – New York, 1993-1994, Morbidity and Mortality Weekly Report, 45(51-52): 1120-3. http://www.cdc.gov/mmwr/PDF/wk/mm4551.pdf Centers for Disease Control and Prevention (CDC). 2005a. Lead Poisoning in Young Children: A Statement by the Centers for Disease Control and Prevention, Appendix, August 2005. http://www.cdc.gov/nceh/lead/Publications/PrevLeadPoisoning.pdf Chen, A., et al. 2005. IQ and blood lead from 2 to 7 years of age: Are the effects in older children the residual of high blood lead concentrations in 2-year-olds? Environmental Health Perspectives, 113(5): 597-601. Clark, CS, Bornschein, RL, Succop, P, Que Hee, SS, Hammond, PB, and Peace, B. (1985) Condition and type of housing as an indicator of potential environmental lead exposure and pediatric blood-lead levels. Environmental Research. 38:46-53. Clemson Environmental Technologies Laboratory (CETL). 2001. A Comparison of Post-Renovation and Remodeling Surface Cleaning Techniques (Unpublished grant report). EPA Grant Agreement Reference No. 82844901. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 125 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Dietrich, K. M., et al. 1993. The Development Consequences of Low to Moderate Prenatal and Postnatal Lead Exposure: Intellectual Attainment in the Cincinnati Lead Study Cohort Following School Entry. Neurotoxicology and Teratology, 15: 37-44. Dietrich, K. M. 1995. A higher level of analysis: Bellinger’s, Interpreting the literature on lead and child development (Commentary). Neurotoxicology and Teratology, 17(3): 223-225. Dietrich, K. M., et al. 2004. Effect of Chelation Therapy on the Neuropsychology and Behavioral Development of Lead-Exposed Children After School Entry. Pediatrics, 114(1): 19-26. Dixon, S., et al. 1999. Achieving dust lead clearance standards after lead hazard control projects: An evaluation of the HUD-recommended cleaning procedure and an abbreviated alternative. Applied Occupational and Environmental Hygiene, 14: 339–344. Ettinger, A. S., et al. 2002. Assessment of Cleaning to Control Lead Dust in Homes of Children with Moderate Lead Poisoning: Treatment of Lead-Exposed Children Trial. Environmental Health Perspectives, 110(12): A773-779. Ewers, L., et al. 1994. Clean-up of Lead in Household Carpet and Floor Dust. American Industrial Hygiene Association Journal, 55(7): 650-657. Hilts S. R., et al. 1998. Effect of Interventions on Children’s Blood Lead Levels. Environmental Health Perspectives, 106(2): 79-83. IRIS Database for Risk Assessment. Updated July 8, 2004. U.S. Environmental Protection Agency, Office of Research and Development, National Center for Environmental Assessment. http://www.epa.gov/iriswebp/iris/index.html Jacobs, D. E., et al. 2002. The prevalence of lead-based paint hazards in U.S. housing. Environmental Health Perspectives, 110(10): A599-606. Kosnett MJ, Wedeen RP, Rothenberg SJ, Hipkins KL, Materna BL, Schwartz BS, Hu H, Woolf A. Recommendations for Medical Management of Adult Lead Exposure. Environ Health Perspect, 2007; 115(3): 463-71. Lanphear, B. P., 1996. Lead-Contaminated House Dust and Urban Children’s Blood Lead Levels, American Journal of Public Health, 86(10): 1416-1421. Lanphear, B. P., et al. 1998a. Environmental Exposures to Lead and Urban Children’s Blood Lead Levels. Environmental Research, Section A, 76: 120-130. Lanphear, B. P., et al. 2005. Low-Level Environmental Lead Exposure and Children’s Intellectual Function: An International Pooled Analysis. Environmental Health Perspectives, 113(7): 894899. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 126 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Liu, X., et al. 2002. Do Children with Falling Blood Lead Levels Have Improved Cognition? Pediatrics, 110(4): 787-791. Madison, Jill E. and Christine G. Hawk. 2005. Lead Poisoned Pets and Your Family. Lead Advisory Service Australia. http://www.lead.org.au/fs/fst9.html. Merck & Co. 2003. Lead Poisoning: Introduction. The Merck Veterinary Manual. http://www.merckvetmanual.com/mvm/index.jsp?cfile=htm/bc/211700.htm Mulligan, G.M., Brimhall, D., and West, J. (2005). Child Care and Early Education Arrangements of Infants,Toddlers, and Preschoolers: 2001 (NCES 2006-039). U.S. Department of Education, National Center for Education Statistics. Washington, DC: U.S. Government Printing Office. Muntner, P.; He, J.; Vupputuri, S.; Coresh, J.; Batuman, V. (2003) Blood lead and chronic kidney disease in the vgeneral United States population: results from NHANES III. Kidney Int. 63: 1044-1050. Navas-Acien, A.; Guallar, E.; Silbergeld, E.K.; Rothenberg, S.J. Lead Exposure and Cardiovascular Disease—A Systematic Review. Environ Health Perspect, 2007; 115(3): 472-482. Nawrot TS, Thijs L, Den Hond EM, Roels HA, Staessen JA. An Epidemiological re-appraisal of the association between blood pressure and blood lead: a meta-analysis. Journal of Human Hypertension, 2002; 16: 123-31. NCHH. An Evaluation of the Efficacy of the Lead Hazard Reduction Treatments Prescribed in Maryland Environmental Article 6-8. Final Report. Submitted to Baltimore City Healthy Start, Inc., Baltimore City Health Department – Lead Abatement Action Program, Baltimore, Maryland. Prepared by National Center for Healthy Housing. April 30, 2002. Needleman, H. L.; Riess, J. A.; Tobin, M. J.; Biesecker, G. E.; Greenhouse, J. B. (1996) Bone Pb levels and delinquent behavior. JAMA J. Am. Med. Assoc. 275: 363-369. Needleman, H. L.; McFarland, C.; Ness, R. B.; Fienberg, S. E.; Tobin, M. J. (2002) Bone Pb levels in adjudicated delinquents. A case control study. Neurotoxicol. Teratol. 24: 711-717. Potula V. et al. 2001. Blood Lead Levels in Relation to Paint and Dust Lead Levels: The Lead-Safe Cambridge Program. American Journal of Public Health, 91(12): 1973-1974. Reissman, D. et al. 2002. Is Home Renovation or Repair a Risk Factor for Exposure to Lead Among Children Residing in New York City? Journal of Urban Health: Bulletin of the New York Academy of Medicine, 79(4): 502-511. Rich, D. Q. 2002. Comparison of Home Lead Dust Reduction Techniques on Hard Surfaces: The New Jersey Assessment of Cleaning Techniques Trial. Environmental Health Perspectives, 110(9): 889-893. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 127 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 Rogan, W. J., et al. 2001. The effect of chelation therapy with succimer on neuropsychological development in children exposed to lead. New England Journal of Medicine, 344(19): 14211426. Ruff, H. A., et al. 1993. Declining blood lead levels and cognitive changes in moderately lead-poisoned children. The Journal of the American Medical Association, 269(13): 1641-1646. Salkever, D. S. 1995. Updated estimates of earnings benefits from reduced exposure of children to environmental lead. Environmental Research, 70(1): 1-6. Schwartz, J.; Angle, C.; Pitcher, H. (1986) Relationship between childhood blood lead and stature. Pediatrics 77: 281-288. Schwartz, J. 1994. Low-level lead exposure and children’s IQ: a meta-analysis and search for a threshold. Environmental Research, 65(1): 42-55. Shih RA, Hu H, Weisskopf MG, Schwartz BS. Cumulative Lead Dose and Cognitive Function in Adults: A Review of Studies That Measured Both Blood Lead and Bone Lead. Environ Health Perspect, 2007; 115(3): 483-92. Shukla, R.; Bornschein, R. L.; Dietrich, K. N.; Buncher, C. R.; Berger, O.G.; Hammond, P. B.; Succop, P. A. (1989). Fetal and infant lead exposure: effects on growth in stature. Pediatrics 84: 604-612. Shukla R, Dietrick KN, Bornschein RL, Berger O, Hammond PB (1992). Lead exposure and growth in the early preschool child: a follow-up report from the Cincinnati Lead Study. Pediatrics, 88(5): 886-92. Simcox, N. J., et al. 1995. Pesticides in Household Dust and Soil: Exposure Pathways for Children of Agricultural Families. Environmental Health Perspectives, 103(12): 1126-1134. Stokes, L., et al. 1998. Neurotoxicity in young adults 20 years after childhood exposure to lead: the Bunker Hill experience. Occupational Environmental Medicine, 55: 507-516. Thomas, WB. 2005. Lead Poisoning and Seizures. Canine Epilepsy Resource Center. http://www.canine-epilepsy.com/Lead.html. Tong, S., et al. 1996. Lifetime exposure to environmental lead and children’s intelligence at 11-13 years: the Port Pirie cohort study. British Medical Journal, 312: 1569-1575. Topping 2005. Write-up of Methodology to Predict Geometric Mean Blood-Lead Concentration in Children Aged 1-2 Years as a Function of Hand Dust-Lead Concentration. Memorandum from Dave Topping, National Program Chemicals Division, U.S. EPA, to William Silagi, Economics, Exposure, and Technology Division, U.S. EPA. December 19, 2005. U.S. Census Bureau. 1995. Property Owners and Managers Survey (POMS). http://www.census.gov/hhes/www/housing/poms/poms.html §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 128 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 U.S. Census Bureau. 1997 and 2003. American Housing Survey (AHS). http://www.census.gov/hhes/www/housing/ahs/ahs.html U.S. Department of Energy (DOE). 2003. Commercial Buildings Energy Consumption Survey (CBECS). U.S. Department of Housing and Urban Development (HUD). 2000. National Survey of Lead and Allergens in Housing. http://www.hud.gov/offices/lead/techstudies/survey.cfm#natsurvey U.S. Department of Housing and Urban Development (HUD). 2003. First National Environmental Health Survey of Child Care Centers. U.S. Department of Housing and Urban Development (HUD). 2004. Evaluation of Effectiveness of Lead Safe Work Practices (LSWP) Training. March 23, 2004. U.S. Environmental Protection Agency (U.S. EPA). 1985. Ambient water quality criteria for lead - 1984. Washington, DC: Office of Water Regulations and Standards, Criteria and Standards Division; report no. EPA/440/5-84/027. Available: http://www.epa.gov/npdes/pubs/owm586.pdf [25October, 2005] U.S. Environmental Protection Agency (U.S. EPA). 1986. Air Quality Criteria for Lead, Volumes 1-4 and Supplement. Research Triangle Park, N.C., EPA 600/8-83-028AF. http://cfpub.epa.gov/ncea/cfm/nceapublication.cfm?ActType=PublicationTopics&detype=DOCU MENT&subject=LEAD&subjtype=TITLE&excCol=Archive U.S. Environmental Protection Agency (U.S. EPA). 1997a. Lead Exposure Associated with Renovation and Remodeling Activities: Environmental Field Sampling Study (EFSS). EPA 747-R-96-007, May 1997. http://www.epa.gov/lead/pubs/r96-007.pdf U.S. Environmental Protection Agency (U.S. EPA). 2000. Economic Analysis of Toxic Substances Control Act Section 403: Lead-Based Paint Hazard Standards. December 2000. http://www.epa.gov/lead/pubs/403_ea_d21.pdf U.S. Environmental Protection Agency (U.S. EPA). 2000a. Hazard Standard Risk Analysis Supplement – TSCA Section 403. Risk Analysis to Support Standards for Lead in Paint, Dust, and Soil, Supplemental Report, Volume 1. EPA 747-R-00-004. December 2000. http://www.epa.gov/lead/403risksupp.htm U.S. Environmental Protection Agency (U.S. EPA). 2001. Economic Analysis of Toxic Substances Control Act Section 403: Lead-Based Paint Hazard Standards. USEPA, Office of Pollution, Prevention and Toxics. http://www.epa.gov/lead/403_ea_d21.pdf U.S. Environmental Protection Agency (U.S. EPA). 2004. IRIS Database for Risk Assessment. Office of Research and Development, National Center for Environmental Assessment. Updated July 8, 2004. http://www.epa.gov/iriswebp/iris/index.html §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 129 5/17/2007 DRAFT—DELIBERATIVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 U.S. Environmental Protection Agency (U.S. EPA). 2005. Integrated Exposure Uptake Biokinetic (IEUBK) Model for Lead in Children, Windows® version (IEUBKwin v1.0 build 262) (September, 2005) 32-bit version. http://www.epa.gov/superfund/programs/lead/products.htm U.S. Environmental Protection Agency (U.S. EPA). 2005b. Ecological Soil Screening Levels for Lead. Interim Final. OSWER Directive 9285.7-70. Available at: http://www.epa.gov/ecotox/ecossl/pdf/eco-ssl_lead.pdf. U.S. Environmental Protection Agency (EPA). 2006. Air Quality Criteria for Lead (Final). EPA/600/R-5/144aF. October. http://cfpub2.epa.gov/ncea/cfm/recordisplay.cfm?deid=32647 U.S. EPA. 2006b. Economic Analysis for the Renovation, Repair, and Painting Program Proposed Rule. February 2006. University of Illinois at Urbana-Champaign. 2002. Comparative Analysis of Exterior Paint Removal Methods: Lead Exposures and Production When Preparing Exterior Clapboard Siding. University of Illinois at Urbana-Champaign, Building Research Council, School of Architecture. EPA Grant Agreement. (Unpublished grant report). University of Rochester School of Medicine (URSM) and National Center for Lead-Safe Housing (NCLSH). 1995. The Relationship of Lead-Contaminated House Dust and Blood-Lead Levels Among Urban Children. Prepared for the U.S. Department of Housing and Urban Development. Grant # MLDP T0001-93. November 15, 1994. WebMD. How Much Sleep Do Children Need? Edited by Cynthia Haines, MD, February 2006. http://www.webmd.com/content/article/122/114620.htm Wirt, J., Choy, S., Rooney, P., Provasnik, S., Sen, A., and Tobin, R. (2004). The Condition of Education 2004(NCES 2004-077). U.S. Department of Education, National Center for Education Statistics. Washington, DC: U.S. Government Printing Office. Yiin, L. M., et al. 2002. Comparison of Techniques to Reduce Residential Lead Dust on Carpet and Upholstery: The New Jersey Assessment of Cleaning Techniques Trial. Environmental Health Perspectives, 110(12): 1233-1237. Yiin, L. M., et al. 2004. Evaluation of cleaning methods applied in home environments after renovation and remodeling activities. Environmental Research, 96(2): 156-62. §402(c) COF Economic Analysis Draft - Do Not Cite or Quote Chapter 5 130 5/17/2007