Community Health Assessment Following
Mercaptan Spill: Eight Mile, Mobile County,
Alabama, September 2012
Behrooz Behbod, MB, ChB, MSc, ScD; Erin M. Parker, MA, PhD; Erin A. Jones, BS, MD;
Tesfaye Bayleyegn, MD; John Guarisco, PhD; Melissa Morrison, MPH; Mary G. McIntyre, MD, MPH, SSBB;
Monica Knight, MPH; Bert Eichold, MD, DrPH, FACP; Fuyuen Yip, PhD, MPH

rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr

Context: In 2008, a lightning strike caused a leak of tert-butyl
mercaptan from its storage tank at the Gulf South Natural Gas
Pumping Station in Prichard, Alabama. On July 27, 2012, the
Alabama Department of Public Health requested Centers for
Disease Control and Prevention epidemiologic assistance
investigating possible health effects resulting from airborne
exposure to mercaptan from a contaminated groundwater
spring, identified in January 2012. Objective: To assess the
self-reported health effects in the community, to determine the
scope of the reported medical services received, and to develop
recommendations for prevention and response to future
incidents. Design: In September 2012, we performed a
representative random sampling design survey of households,
comparing reported exposures and health effects among
residents living in 2 circular zones located within 1 and 2 miles
from the contaminated source. Setting: Eight Mile community,
Prichard, Alabama. Participants: We selected 204 adult
residents of each household (≥18 years) to speak for all
household members. Main Outcome Measures: Self-reported
mercaptan odor exposure, physical and mental health outcomes,
and medical-seeking practices, comparing residents in the 1and 2-mile zones. Results: In the past 6 months, 97.9% of
respondents in the 1-mile zone and 77.6% in the 2-mile zone
reported mercaptan odors. Odor severity was greater in the
1-mile zone, in which significantly more subjects reported
exposures aggravating their physical and mental health including
shortness of breath, eye irritations, and agitated behavior.
Overall, 36.5% sought medical care for odor-related symptoms.
J Public Health Management Practice, 2014, 20(6), 632–639
Copyright C 2014 Wolters Kluwer Health   Lippincott Williams & Wilkins

Conclusions: Long-term odorous mercaptan exposures were
reportedly associated with physical and psychological health
complaints. Communication messages should include strategies
to minimize exposures and advise those with cardiorespiratory
conditions to have medications readily available. Health care
practitioners should be provided information on mercaptan
health effects and approaches to prevent exacerbating existing
chronic diseases.
KEY WORDS: health effects, mercaptan, natural gas, odor, spill

Author Affiliations: Epidemic Intelligence Service (Drs Behbod and Parker),
National Center for Environmental Health, Division of Environmental Hazards and
Health Effects (Drs Behbod, Bayleyegn and Yip), National Center for Injury
Prevention and Control, Division of Unintentional Injury Prevention (Dr Parker),
and Office of Public Health Preparedness and Response, Office of Science and
Public Health Practice, Career Epidemiology Field Officer Program
(Ms Morrison), US Centers for Disease Control and Prevention, Atlanta, Georgia;
Boston University Medical School, Boston, Massachusetts (Dr Jones); Alabama
Department of Public Health, Montgomery, Alabama (Drs Guarisco, McIntyre
and Ms Morrison); and Mobile County Health Department, Mobile, Alabama (Ms
Knight and Dr Eichold).
The authors thank Dr Prabasaj Paul for his assistance with mapping the
sampling frame and Mr James Durant for creating the wind roses for Mobile
County. They also thank the members of the Air Pollution and Respiratory Health
Branch Epidemiology Team for their support in designing and reviewing our
household questionnaire.
The authors report no conflicts of interest.
Supplemental digital content is available for this article. Direct URL citation
appears in the printed text and is provided in the HTML and PDF versions of this
article on the journal’s Web site (www.JPHMP.com).
Correspondence: Behrooz Behbod, MB, ChB, MSc, ScD, Epidemic Intelligence
Service, Centers for Disease Control and Prevention, 1600 Clifton Rd, Mailstop
E-92, Atlanta, GA 30333 (bbehbod@post.harvard.edu).
DOI: 10.1097/PHH.0000000000000024

632
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

 Community Health Assessment Following Mercaptan Spill

In 2008, a lightning strike caused a leak of tert-butyl
mercaptan (TBM) from a storage tank at the Gulf South
Natural Gas Pumping Station in Prichard, Alabama
(Figure 1). Tert-butyl mercaptan (also known as a mercaptan or thiol) is a sulfur-containing compound used
as an industrial additive in natural gas to detect leaks.
Since October 2011, residents have reported experiencing possible exposure and health symptoms including
nausea, dizziness, headaches, general weakness, and
respiratory and mucous membrane symptoms, such as
eye, nose, and throat irritation.
In January 2012, Mobile Gas Service collected a water sample at a groundwater spring located at a beaver
pond (Figure 1); the sample contained 460 µg/L TBM).
Another water sample collected in February 2012 measured 14 000 µg/L TBM. In April 2012, at the request
of the Alabama Department of Environmental Management, the US Environmental Protection Agency Region 4 collected ambient air samples from the residential neighborhoods surrounding the pumping station to
measure levels of TBM and other pollutants potentially
responsible for the health complaints.1 High concentrations of TBM (230 ppbV) were measured at the beaver
pond spring. While concentrations were above odor
thresholds for TBM,2-4 these levels did not exceed the
permissible exposure limits for most mercaptans (500
ppbV).5
FIGURE 1 ● Map Depicting the Locations of the Gas

Pumping Station, Beaver Pond Spring, and 1- and 2-Mile
Radius Study Sampling Frames in the Eight Mile
Community, Alabama
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq

(Acknowledgment: Prabasaj Paul; Source: Google, 2012)

❘ 633

Tert-butyl mercaptan is used to detect gas leaks
because of its skunk-like odor and low odor threshold
(0.1 ppb).3,6 Because of an estimated Koc of 49, TBM has
very high mobility if released into soil or water and
generally volatilizes into the ambient air.7 Short-term
exposure may cause dermal, eye and respiratory
irritation, lack of sense of smell, nausea, vomiting,
diarrhea, difficulty breathing, headache, drunk-like
symptoms, cyanosis, lung congestion, kidney damage,
convulsions, and even coma. Tert-butyl mercaptan
is not known to cause reproductive toxicity or to be
teratogenic or carcinogenic.6 Currently, no information
is available on adverse health effects associated with
long-term TBM exposure.
On July 27, 2012, the Alabama Department of Public Health requested Centers for Disease Control and
Prevention (CDC) epidemiologic assistance investigating possible health effects resulting from community
exposure to TBM from contaminated ground and surface water in the Eight Mile community of Prichard,
Alabama. Our study objectives were as follows: (1) assess the self-reported health effects in the community,
(2) determine the scope of the reported medical services received, and (3) develop recommendations for
prevention and response to future incidents.

● Methods
We used a representative random sampling design survey of households in the Eight Mile community to complete an in-person interview. We conducted household
questionnaires during September 12 to 14, 2012.

Exposure assessment
The Environmental Protection Agency report1 showed
that TBM pollutant concentrations varied substantially
both temporally and spatially. Measures of environmental TBM concentrations over a few days may not
represent the chronic, time-varying exposures. Moreover, assessing exposures based on community-level
measurements may not be suitable to estimate spatially
variable, individual-level exposures. We therefore decided to use self-reported or perceived exposures to
the TBM odor as a proxy for exposures among the residents of the Eight Mile community. As self-reported
exposures can be prone to recall bias, we also included an objective measure of exposure based on distance to the suspected source, the beaver pond spring
(Figure 1). We grouped households into 2 zones based
on their distance from the source.
To select the number, location, and shape of these
zones, we considered the pollutant characteristics, average wind speeds and directions, and the location and

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

 634 ❘ Journal of Public Health Management and Practice
density of at-risk households. The chemicals in question typically vaporize, stay close to the ground surface, and do not travel far from the source. Assuming
only 1 source, we chose 1- and 2-mile zones to stay in
proximity to the source as well as to meet the sample
size required for sufficient statistical power. The prevailing wind direction for the Eight Mile community is
south-easterly from the Gulf of Mexico, but the winds
also come from other directions over the course of the
year. As we were interested in long-term, year-round
exposures, all households around the source may be at
risk. We therefore generated a circular sampling area,
centered at the beaver pond spring.
We determined that we would need to complete surveys for at least 97 households per zone so that the margin of error (95% confidence interval) would be ±10%
if the prevalence of each outcome of interest was about
50%. We increased the sample size to 110 per zone
(220 total) to account for potential nonresponse. We
sampled 2 circular zones (1-mile and 2-mile radii),
centered at the beaver pond spring (Figure 1).

Outcome assessment
We surveyed residents of the randomly selected households in each of the 2 zones in the Eight Mile community about self-reported health outcomes and health
care-seeking practices.

Field work
We used the 2010 Census8 and randomly selected
households with an equal probability of selection. To
identify the Census blocks from which the 220 households would be selected in the field, we applied the
Community Assessment for Public Health Emergency
Response (CASPER) tool developed within ArcMap
10.0 software (ESRI, Redlands, California). On the basis
of the cumulative house number per block (Census 2010
data8 ) and the fact that we needed 110 housing units
per zone, we selected blocks from zones 1 and 2, respectively. The number of housing units to be sampled in
each Census block was proportional to the number of
housing units per block; therefore, more homes would
be interviewed from more populated blocks.
Members of each study team were provided information on the total number of households in each Census block, which was divided by the number of homes
to be surveyed. For example, if 30 homes were in the
team’s designated area, of which team members were
to survey 3, they would visit 1 in 10 homes. Team members would each randomly select a first household, systematically survey every 10th household, and then visit
every 10th household within their Census block until
they completed 3 interviews. This method ensured an

approximated random sample, with each household
having an equal probability of selection.
After team members gained verbal consent, they
asked 1 adult (≥18 years) from each selected household to respond to the survey. Any adult, regardless
of gender, race, ethnicity, or religion, was eligible to
participate in the assessment.

Statistical methods
Data were entered into EpiInfo 7 (CDC, Atlanta,
Georgia; http://wwwn.cdc.gov/epiinfo/) and analyzed with SAS 9.3 (SAS Institute Inc, Cary, North
Carolina). Unweighted descriptive statistics of the
community demographics and the prevalence of
reported exposures to mercaptan-like odors, health
complaints, and medical services sought or received
are presented. We compared residents living within 1
(zone 1) and 2 miles (zone 2) of the beaver pond spring.
As a secondary analysis, we used univariate logistic regression to evaluate the association between the
objective measure of residential zone and the odds
of reporting odor exposure in the past 6 months. We
also checked to see if this association was independent
of other potential predictors of reported odor or confounders. Covariates that we considered included demographics, employment within the Eight Mile area,
household type, home characteristics, and environmental coexposures. If any association had a P value
of less than .05, we included them together in a multiple regression model.
This investigation was a response to a public health
threat. In accordance with federal human subject protection regulations, it was not considered to be human
subject research. As such, the study was designated
as public health practice and not research by the CDC
National Center for Environmental Health/Agency for
Toxic Substances and Disease Registry institutional review board.

● Results
We completed 97 (88% of the n = 110 goal) and 107
(97% of the n = 110 goal) surveys in zones 1 and 2,
respectively. Table 1 presents the demographic characteristics of our entire study sample. The respondents
were representative of the population within the sampling frame with respect to gender, race, and ethnicity
according to the 2010 Census8 (results not shown). The
2 zones did not vary with respect to demographic characteristics, with the exception of employment status;
more unemployed respondents were in zone 1 (25.8%)
than in zone 2 (12.2%), while more retired respondents
were in zone 2 (50.5%) than in zone 1 (36.1%).

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

 Community Health Assessment Following Mercaptan Spill
TABLE 1 ● Demographics of Respondents in the Eight

Mile Community, Alabama, September 2012
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq

Age characteristics, mean (range)
Age of adult respondents, y
Categorical characteristics, % (95% CI)
Age of all household members, y
≤2
3-5
6-9
10-17
18-25
26-40
41-65
66-75
≥76
Gender
Male
Female
Don’t know/refused
Race/ethnicity
White, Non-Hispanic
Black, Non-Hispanic
Other
Don’t know/refused
Annual household income, $
0-<5000
5000-<10 000
10 000-<15 000
15 000-<20 000
20 000-<25 000
25 000-<35 000
35 000-<50 000
50 000-<75 000
>75 000
Don’t know/refused
Length of time lived in Eight Mile, y
≤1
2-4
≥5
Employment status
Employed
Unemployed
Student
Retired
Don’t know/refused
Work within Eight Mile (among employed)?
Yes
No
Don’t know/refused

Overall (n = 204)
52.5 (18.0-94.0)

5.2 (3.5, 6.9)
5.2 (3.5, 6.9)
5.7 (4.0, 7.5)
10.4 (8.1, 12.7)
11.3 (8.9, 13.7)
15.7 (12.9, 18.4)
32.5 (29.0, 36.0)
9.8 (7.5, 12.0)
4.2 (2.7, 5.7)
36.3 (30.0, 43.1)
58.3 (51.5, 64.9)
5.4 (3.0, 9.4)
17.7 (13.0, 23.5)
80.9 (74.9, 85.7)
1.0 (0.3, 3.5)
0.5 (0.1, 2.7)
11.8 (8.0, 16.9)
6.4 (3.8, 10.6)
5.9 (3.4, 10.0)
9.3 (6.0, 14.1)
12.3 (8.4, 17.5)
7.4 (4.5, 11.8)
5.9 (3.4, 10.0)
9.8 (6.4, 14.7)
5.4 (3.0, 9.4)
26.0 (20.5, 32.4)
4.9 (2.7, 8.8)
9.8 (6.4, 14.7)
85.3 (79.8, 89.5)
33.8 (27.7, 40.6)
18.6a (13.9, 24.5)
3.4 (1.7, 6.9)
43.6a (37.0, 50.5)
0.5 (0.1, 2.7)
23.2 (14.8, 34.4)
75.4 (64.0, 84.0)
1.5 (0.3, 7.8)

a The 2 zones did not vary with respect to demographic characteristics, with the
exception of employment status; more unemployed respondents were in zone 1
(25.8%) than in zone 2 (12.2%), while more retired respondents were in zone 2
(50.5%) than in zone 1 (36.1%).

❘ 635

When examining the mercaptan odor characteristics reported by respondents, we found that 97.9% of
respondents in zone 1 and 77.6% in zone 2 reported
experiencing odors in the past 6 months (Table 2). In
addition, 47.4% of residents in zone 1 and 58.9% in zone
2 reported experiencing odors for 4 years or less; 28.9%
in zone 1 and 7.5% in zone 2 reported odors before
the start of the mercaptan spill in 2008 (≥5 years). In
the majority of households in both zones, respondents
described the odors as either gas or rotten eggs.
Respondents primarily reported smelling the odors
both inside and outside their homes, followed by the
beaver pond vicinity (results not shown). Odor severity was greater in zone 1 (mean score 9.1/10) relative to
zone 2 (mean score 7.8/10). In the univariate analysis,
respondents living in the 1-mile zone were more likely
than those in the 2-mile zone to report odors at home
(odds ratio [OR] = 4.4 [95% confidence interval (CI):
2.1, 9.5]; P < .001; N = 204). Since male gender (OR =
0.4 [95% CI: 0.2, 0.8]; P < .001; N = 204) and the use
of a gas stove for cooking (OR = 1.9 [95% CI: 1.0, 3.9];
P = .055; N = 204) were also associated with reported
odor in the univariate analysis, we included them in the
adjusted model. However, in the multiple regression
model, gender (OR = 0.4 [95% CI: 0.2, 0.8]; P < .001;
N = 204) remained an independent predictor of reported odor, but the use of a gas stove (OR = 1.5 [95%
CI: 0.7, 3.1]; P = .292; N = 204) was no longer significant.
Nevertheless, the association between residential zone
and odor was not confounded in the adjusted model
(OR = 4.7 [95% CI: 2.1, 10.4]; P < .001; N = 204).
Of households reporting experiencing odors, the
majority (88.7% in zone 1 and 63.6% in zone 2) reported
that odors were worst outdoors, around their home,
especially in the early morning (6-8 AM) and evening
(6-10 PM) hours. A variety of weather conditions appeared to affect odor severity in both zones, though
25.8% of residents in zone 1 and 13.1% in zone 2 reported severities not being affected by weather. The
majority of respondents in both zones reported odor
severity not changing since the smell first became noticeable (42.7%). However, among those who reported
a change, more subjects reported increased severity in
zone 1 (30.9%), whereas more subjects reported decreased severity over time in zone 2 (26.2%). As a result of experiencing odors, more respondents in zone 1
than in zone 2 reported decreasing time spent outdoors
(59.8% vs 39.3%) and being less frequently able to open
their windows at home (33.0% vs 19.6%).
Compared to respondents in zone 2, more respondents in zone 1 reported odors aggravated their physical (41.2% vs 21.5%) and mental (17.5% vs 12.2%)
health. In addition, more households in zone 1 than
in zone 2 complained of nasal congestion, dyspnea, cough, wheezing, loss of appetite, eye irritations,

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

 636 ❘ Journal of Public Health Management and Practice
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq

TABLE 2 ● Odor Characteristics Reported by Respondents in the Eight Mile Community, Alabama, September 2012

1-Mile Zone (n = 97),
% (95% CI)

2-Mile Zone (n = 107),
% (95% CI)

Overall (n = 204),
% (95% CI)

Odor smelled in past 6 mo
“In the past 6 mo, have you smelled a suspicious or strange gas odor in or near the Eight Mile community?”
Yes
97.9 (92.8, 99.4)
77.6 (68.8, 84.4)
No
2.1 (0.6, 7.2)
22.4 (15.6, 31.2)

87.3 (82.0, 91.2)
12.8 (8.9, 18.0)

Odor duration, y
“If yes, when did you first notice this odor or smell?”
≤1
26.8 (19.0, 36.4)
2-4
20.6 (13.8, 29.7)
≥5
28.9 (20.8, 38.6)
Don’t know / refused
23.7 (16.4, 33.1)

47.7 (38.5, 57.0)
11.2 (6.5, 18.6)
7.5 (3.8, 14.1)
33.6 (25.4, 43.0)

37.8 (31.4, 44.6)
15.7 (11.3, 21.3)
17.7 (13.0, 23.5)
28.9 (23.1, 35.5)

Odor description
“How would you describe the smell of the odor?”
Gas
54.6 (44.7, 64.2)
Skunk
7.2 (3.5, 14.2)
Rotten eggs
19.6 (12.9, 28.6)
Onions
4.1 (1.6, 10.1)
Other
22.7 (15.5, 32.0)

29.9 (22.1, 39.2)
7.5 (3.8, 14.1)
30.8 (22.9, 40.1)
5.6 (2.6, 11.7)
13.1 (8.0, 20.8)

41.7 (35.1, 48.5)
7.4 (4.5, 11.8)
25.5 (20.0, 31.9)
4.9 (2.7, 8.8)
14.2 (10.1, 19.7)

headache, dizziness, agitated behavior, difficulty concentrating, and worsening hypertension (Figure 2; full
table of health complaints available in Supplementary Table 1 at http://links.lww.com/JPHMP/A57).
A higher proportion of respondents in zone 1 than
in zone 2 reported 14 or more out of the past
30 days as being physically and mentally unhealthy,
in addition to being unable to do normal activities
(eg, self-care, work, recreation) because of poor physical/mental health (Supplementary Table 2, available at
http://links.lww.com/JPHMP/A58).
Most households reported having health insurance,
with access to a regular physician and transportation
to receive medical care (Table 3).
FIGURE 2 ● Household-Level Health Complaints That

Significantlya Vary by Zone, Eight Mile, Alabama,
September 2012
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq

● Discussion
To our knowledge, this study is the first epidemiological investigation in published literature describing the
long-term community health in an area where a mercaptan spill occurred. In a representative random sampling design survey of households in the Eight Mile
community of Prichard, Alabama, we found a higher
proportion of residents in the 1-mile than in the 2-mile
zone reporting odors that aggravated their physical
and mental health. While our study was not designed
to evaluate a causal relationship between mercaptan
exposure and health effects, we did find a significant association between residential zone and reported odors.
As a variation on the traditional CASPER cluster survey methodology, we have demonstrated a rapid and
relatively low-cost response tool that may be used for
future assessments of long-term environmental exposures and health in small communities. To complement
the self-reported odor exposure, we used an objective
measure of proximity to the source. A 1989 California investigation employed a similar approach, using
proximity as a marker of exposure to study the acute
health effects from community exposure to n-propyl
mercaptan from a pesticide-treated potato field.9 Notably, that study reported that odor perception, rather
than proximity to the potato field, may have been responsible for the observed health effects.
Because the perception of an odor, rather than an
actual chemical exposure, may have a role in the development of health complaints, the properties of the

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

 Community Health Assessment Following Mercaptan Spill

❘ 637

qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq

TABLE 3 ● Household-Level Health Care Sought by Respondents in the Eight Mile Community, AL, September 2012

1-Mile Zone (n = 97),
% (95% CI)

Insured (health)
“Do you have health insurance?”
Yes
78.4 (69.2, 85.4)
No
20.6 (13.8, 29.7)
Don’t know/refused
2.1 (0.6, 7.4)
Access to regular physician
“Do you have a regular physician?”
Yes
81.4 (72.6, 87.9)
No
16.5 (10.4, 25.1)
Don’t know/refused
2.1 (0.6, 7.2)
Access to transportation to receive medical care
“Do you have transportation to receive medical care?”
Yes
94.9 (88.5, 97.8)
No
5.2 (2.2, 11.5)
Don’t know/refused
...

2-Mile Zone (n = 107),
% (95% CI)

Overall (n = 204),
% (95% CI)

76.6 (67.8, 83.6)
23.4 (16.4, 32.2)
...

77.5 (71.2, 82.7)
22.1 (16.9, 28.2)
0.5 (0.1, 2.7)

77.6 (68.8, 84.4)
22.4 (15.6, 31.2)
...

79.4 (73.3, 84.4)
19.6 (14.7, 25.6)
1.0 (0.3, 3.5)

93.5 (87.1, 96.8)
3.7 (1.5, 9.2)
2.8 (1, 7.9)

94.1 (90.0, 96.6)
4.4 (2.3, 8.2)
1.5 (0.5, 4.2)

Among those with symptoms in past 30 days
“Did you or any member of your household seek help for any of the above physical or mental health conditions at any of the following locations?”
Health care sought?
Yes
69.5 (59.6, 77.8)
68.2 (57.9, 77)
68.9 (61.8, 75.1)
No
28.4 (20.3, 38.2)
30.7 (22, 41)
29.5 (23.4, 36.5)
Don’t know/refused
2.1 (0.6, 7.4)
1.1 (0.2, 6.2)
1.6 (0.6, 4.7)
Type of health care sought
Community health center
10.5 (5.8, 18.3)
17.1 (10.6, 26.2)
13.7 (9.4, 19.4)
Mental health clinic
...
...
...
Family doctor
49.5 (39.6, 59.4)
42.1 (32.3, 52.5)
45.9 (38.8, 53.1)
ER
6.3 (2.9, 13.1)
2.3 (0.6, 7.9)
4.4 (2.2, 8.4)
Social Services
...
...
...
Urgent care center
2.1 (0.6, 7.4)
2.3 (0.6, 7.9)
2.2 (0.9, 5.5)
Free clinic
2.1 (0.6, 7.4)
1.1 (0.2, 6.2)
1.6 (0.6, 4.7)
Other
6.3 (2.9, 13.1)
5.7 (2.5, 12.6)
6.0 (3.4, 10.4)
Among those who reported odor at any time
“Did you or any member of your household seek help for any of physical or mental health conditions possibly resulting from the odor? If so,
which of the following locations did you visit?”
Health care sought?
Yes
41.1 (31.7, 51.1)
31.3 (22.4, 41.9)
36.5 (29.8, 43.8)
No
54.7 (44.7, 64.4)
62.7 (51.9, 72.3)
58.4 (51.1, 65.4)
Don’t know/refused
4.2 (1.7, 10.3)
6.0 (2.6, 13.3)
5.1 (2.7, 9.3)
Type of health care sought
Community health center
9.5 (5.1, 17)
7.2 (3.4, 14.9)
8.4 (5.2, 13.4)
Mental health clinic
...
...
...
Family doctor
29.5 (21.3, 39.3)
20.5 (13.2, 30.4)
25.3 (19.5, 32.1)
ER
4.2 (1.7, 10.3)
1.2 (0.2, 6.5)
2.8 (1.2, 6.4)
Social Services
...
...
...
Urgent care center
2.1 (0.6, 7.4)
...
1.1 (0.3, 4.0)
Free clinic
3.2 (1.1, 8.9)
...
1.7 (0.6, 4.8)
Other
5.3 (2.3, 11.7)
2.4 (0.7, 8.4)
3.9 (1.9, 7.9)

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

 638 ❘ Journal of Public Health Management and Practice
specific chemical agent must be considered, with
emphasis on the ambient concentration required for olfaction (ie, odor threshold) versus sensory irritation.10
Because mercaptans typically have very low odor
thresholds, while their irritation thresholds may be
a few orders of magnitude higher, their odors per se
may be responsible for the genesis of the reported
symptoms, as opposed to their toxicological effect.11
Conversely, potent irritant compounds, such as methyl
isothiocyanate that have irritation thresholds at lower
ambient concentrations than their respective odor
thresholds, may cause health effects without any
perceived odor.12
In addition to the chemical characteristics, odor perception may also be affected by a variety of factors,13
including top-down processing,12 such as individuals’
beliefs regarding the potential health effects from
odorants,14 their coping style,15 personality,16 environmental worries,11 and psychological stress.17 Moreover,
nasal irritation may be mediated by gender, allergies,
smoking, age, and self-reported pollutant sensitivity.18
A combination of complementary subjective and
objective measures of exposure can therefore be very
valuable in community assessments to measure the
plausibility of reported odors and compare their
characteristics with the expected features of a true
underlying environmental contaminant.
While we were limited by the self-reported odor exposure and health outcomes and therefore prone to
recall bias, the strength of this investigation was the objective measure of using distance to source as a marker
of exposure. Respondents were blind to their classification to zone 1 or 2, and interviewers’ questionnaires
were standardized to minimize potential for bias. In addition, we sampled 2 zones that were comparable with
respect to demographic characteristics while also being
representative of the population. Although mercaptan
odor perception may be prone to olfactory accommodation, one would not expect this to vary between zones
1 and 2.

● Conclusions
Proximity to the contaminated beaver pond was associated with persistent odorous mercaptan exposures as
well as physical and psychological health complaints
in the Eight Mile community. On the basis of our findings, we identified some key areas with the state and
local health department that we believe would be helpful in preparation for potential future incidents. Health
departments might prepare public health communication messages in advance to include strategies to minimize exposures (eg, limit outdoor activity and keep
windows closed in the evening and overnight hours)

and to advise those with chronic respiratory and cardiovascular conditions to have their medications readily available. We suggest that health care practitioners
be provided information on potential health effects of
mercaptan exposures and approaches to prevent and
manage exacerbations of existing chronic diseases, potentially resulting from mercaptan exposures.
REFERENCES
1. United States Environmental Protection Agency, Region 4.
Eight Mile/Prichard Air Study, Prichard, Mobile County,
Alabama, April 24-26, 2012. SESD Project Identification
Number: 12-0375. http://www.epa.gov/region4/sesd/
reports/2012-0375/2012-0375.pdf. Accessed May 3, 2013.
2. Ruth JH. Odor Thresholds and irritation levels of several chemical substances: a review. Am Ind Hyg Assn J.
1986;47(3):A142-A151.
3. Agency for Toxic Substances and Disease Registry (ATSDR).
Toxicological Profile for Methyl Mercaptan. Atlanta, GA: US Department of Health and Human Services, Public Health Service; 1992.
4. Arkema Inc. t-Butyl Mercaptan [Material Safety Data Sheet].
Product Code: 002404, Version 1.1. King of Prussia, Pennsylvania: Arkema Inc.
5. Occupational Safety and Health Administration. Chemical
sampling information: butyl mercaptan. http://www
.osha.gov/dts/chemicalsampling/data/CH_223600.html.
Accessed May 3, 2013.
6. National Institute for Occupational Safety and Health
(NIOSH). NIOSH Pocket Guide to Chemical Hazards: n-butyl mercaptan. http://www.cdc.gov/niosh/npg/
npgd0083.html. Accessed May 3, 2013.
7. US National Library of Medicine. TOXNET: Toxicology data network. http://toxnet.nlm.nih.gov/cgi-bin/sis/
search/a?dbs+hsdb:@term+@DOCNO+1611. Accessed May
3, 2013.
8. US Census Bureau. 2010 Population Estimates, Census
2010. http://factfinder2.census.gov/faces/nav/jsf/pages/
index.xhtml. Accessed May 3, 2013.
9. Ames RG, Stratton JW. Acute health effects from community exposure to n-propyl mercaptan from an Ethoprop (Mocap R )-treated potato field in Siskiyou County,
California. Arch Environ Health. 1991;46(4):213-217.
10. Dalton P. Odor, irritation and perception of health risk. Int
Arch Occup Environ Health. 2002;75:283-290.
11. Claeson AS, Liden E, Nordin M, Nordin S. The role of perceived pollution and health risk perception in annoyance and
health symptoms: a population-based study of odorous air
pollution. Int Occ Environ Health. 2013;86:367-374.
12. Shusterman D. Odor-associated health complaints: competing explanatory models. Chem Senses. 2001;26:339-343.
13. Greenberb MI, Curtis JA, Vearrier D. The perception of odor is
not a surrogate maker for chemical exposure: a review of factors influencing human odor perception. Clin Tox. 2013;57:7076.
14. Bulsing PJ, Smeets MAM, Van den Hout MA. The implicit association between odors and illness. Chem Senses.
2009;34:111-119.

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

 Community Health Assessment Following Mercaptan Spill

15. Cavalini PM, Koeter-Kemmerling LG, Pulles MP. Coping
with odour annoyance and odour concentrations: three field
studies. J Environ Psychol. 1991;11:123-142.
16. Smeets MAM, Dalton PH. Evaluating the human response to
chemicals: odor, irritation and non-sensory factors. Environ
Toxicol Pharmacol. 2005;19:581-588.

❘ 639

17. Dayal HH, Baranowski T, Li Y, Morris R. Hazardous chemicals: psychological dimensions of the health sequelae of a
community exposure in Texas. J Epidemiol Community Health.
1994;48:560-568.
18. Shusterman D. Individual factors in nasal chemesthesis.
Chem Senses. 2002;27:551-564.

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.