RESTRAINT AND SECLUSION IN MAINE SCHOOLS Reviewing the First Four Years of Data Required by MDOE Rule Chapter 33 Disability Rights Maine 24 Stone St, Suite 204 Augusta, ME 04330 800.452.1948 ● advocate@drme.org April 2017 Table of Contents Introduction ..................................................................................................................................... 1 Why Limit the Use of Physical Restraint and Seclusion in Schools?............................................. 2 What Does the Available Chapter 33 Data Tell Us? ...................................................................... 4 Figure 1 ....................................................................................................................................... 5 Students with Disabilities are Significantly More Likely to be Restrained and Secluded. ............ 5 Figure 2 ....................................................................................................................................... 6 Figure 3 ....................................................................................................................................... 7 Figure 4 ....................................................................................................................................... 7 Figure 5 ....................................................................................................................................... 8 Chapter 33 Data is Incomplete ........................................................................................................ 9 Figure 6 ....................................................................................................................................... 9 Chapter 33 Data is Unreliable ....................................................................................................... 10 Schools Must Review Data Quarterly to Identify Ways to Reduce the Use of Seclusion and Restraint. ....................................................................................................................................... 11 Recommendations ......................................................................................................................... 12 INTRODUCTION In 2011, concerns raised by parents, advocates, and journalists regarding the use of restraint and seclusion in Maine schools led to the formation of a work group.1 The Chapter 33 ConsensusBased Rule Development Group was tasked with taking a fresh look at regulating the use of restraint and seclusion practices in schools.2 At the time, Maine schools were not required to report the number of restraints or seclusions to the Maine Department of Education (“MDOE”). The resulting language was submitted to the Joint Standing Committee on Education and Cultural Affairs.3 The final regulations, enacted in April 2012, are known as “Chapter 33.” 4 In addition to substantive and procedural requirements designed to ensure that restraint and seclusion are limited to emergency situations where there is a risk of injury or harm (and only after less intrusive interventions have failed), Chapter 33 requires schools to document each use of physical restraint or seclusion in an incident report. Chapter 33 also requires that on a quarterly basis, these incident reports be aggregated and reported to the superintendent or chief administrator, who then “shall review cumulative reports received…and identify those areas that can be addressed to reduce the future use of physical restraint and seclusion.”5 Further, Chapter 33 requires that each covered entity “shall submit to the Department of Education an annual report of the incidence of physical restraint and seclusion.”6 Four years of this data is available.7 Disability Rights Maine (“DRM”), Maine’s federally funded Protection and Advocacy agency for people with disabilities, undertook an analysis of the first four years of available Chapter 33 data on restraint and seclusion in Maine’s schools. In addition to reviewing publically available data reported to MDOE, DRM obtained a sample of the quarterly “building-level data” 1 Emily Parkhurst, Children held down: Families question use of therapeutic restraints in Maine public schools, The Forecaster (2010), http://www.theforecaster.net/children-held-down-families-question-useof-therapeutic-restraints-in-maine-public-schools/ (last visited Mar 16, 2017); Emily Parkhurst, ‘I thought I was the only one’: 3 kids, 3 years, nearly 100 school restraints, The Forecaster (2010), http://www.theforecaster.net/i-thought-i-was-the-only-one-3-kids-3-years-nearly-100-school-restraints/ (last visited Mar 16, 2017). 2 Maine Department of Education, The Consensus Based Rulemaking Team (CBR Team), Maine.gov (2011), http://www.maine.gov/education/rulechanges/chapter33/team.html (last visited Mar 16, 2017). 3 Maine Department of Education, DOE Rule Chapter 33 Review and Revision Project, Maine.gov (2007), http://www.maine.gov/education/rulechanges/chapter33/index.html (last visited Mar 16, 2017). 4 CMR 05–071 Ch. 33: Rule Governing Physical Restraint and Seclusion. Shortly after Chapter 33 went into effect, the Rule was brought before the Maine Legislature a second time, where several changes were made to address concerns raised by educators. 5 05-071 CMR Ch. 33, § 10.1. 6 05-071 CMR Ch. 33, § 10.2. 7 Data from School Year (“SY”) 12-13, SY 13-14, SY 14-15, and SY 15-16, have been collected by MDOE and made available on its website. See: http://www.maine.gov/doe/school-safety/restraints/ 1 (individual school buildings, as opposed to the district as a whole) directly from school districts.8 We also reviewed individual case files where restraint and seclusion were used and solicited similar anecdotal information from other attorneys and advocates who represent students. DRM found that students with disabilities are subjected to seclusion and restraint at rates that far exceed students without disabilities. In addition, DRM discovered that many covered entities are not reporting the use of restraint and seclusion, as required by Chapter 33. For the data that was reported, DRM identified serious concerns with reliability and consistency. Finally, DRM determined that schools may not be fulfilling their obligation to review building-level data quarterly as a means of identifying ways to reduce the future use of restraint and seclusion. WHY LIMIT THE USE OF PHYSICAL RESTRAINT AND SECLUSION IN SCHOOLS? Across the country, the use of restraint and seclusion in schools has come under increasing levels of scrutiny and study. At the federal level, the United States Department of Education (USDOE) has urged minimal reliance on these emergency interventions. A recently issued USDOE restraint and seclusion resource document emphasized that the “foundation of any discussion about the use of restraint and seclusion is that every effort should be made to structure environments and provide supports so that restraint and seclusion are unnecessary.” 9 USDOE reached this conclusion after determining that restraint and seclusion were potentially harmful, and ineffective.10 According to the USDOE, “[t]here is no evidence that using restraint or seclusion is effective in reducing the occurrence of the problem behaviors that frequently precipitate the use of such techniques.”11 Similarly, the Senate Health, Education, Labor and Pensions (HELP) Committee recently investigated the use of restraint and seclusion in schools and concluded that “[t]here is no evidence that physically restraining or putting children in unsupervised seclusion in the K-12 school system provides any therapeutic benefit to a child.”12 8 In September 2015, DRM requested building-level data on restraint and seclusion from 43 school districts. DRM received responses from 41 districts covering 12 counties and ranging in size from 200 – 7,000 students. 9 U.S. Department of Education, Restraint and Seclusion: Resource Document, Washington D.C., 2012, available at: https://ed.gov/policy/seclusion/restraints-and-seclusion-resources.pdf 10 Id. 11 Id. 12 Majority Committee Staff, U.S. Senate Health, Education, Labor, and Pensions Committee, Dangerous Use of Seclusion and Restraints in Schools Remains Widespread and Difficult to Remedy: A Review of Ten Cases (Feb. 12, 2014), available at http://www.help.senate.gov/imo/media/doc/Seclusion%20and%20Restraints%20Final%20Report.pdf 2 These interventions, which have no demonstrable therapeutic or educational benefit, are also potentially dangerous and harmful. According to the Substance Abuse and Mental Health Services Administration (SAMHSA), “studies have shown that the use of seclusion and restraint can result in psychological harm, physical injuries, and death to both the people subjected to and the staff applying these techniques.”13 For individuals with a history of trauma, the use of physical restraints can be particularly harmful.14 The recognition that the use of seclusion and restraint can lead to physical and psychological injury, coupled with the lack of evidence that these interventions are effective at reducing difficult behaviors, has caused states to begin to look more critically at the use of these interventions in schools. Maine took this step with Chapter 33. Chapter 33 is generally aligned with the recommendations from the USDOE and other entities that have addressed the use of restraint and seclusion in schools in recent years.15 Chapter 33 recognizes that restraint and seclusion: 1) do not have any therapeutic or educational value; and 2) are only properly understood as emergency interventions when there is a risk of injury or harm. Accordingly, Chapter 33 prioritizes the use of behavior intervention and strategies to proactively address problem behaviors through skill building and environmental modifications in order to avoid situations where physical intervention may become necessary. On a systemic level, Chapter 33 incorporates detailed documentation and reporting requirements to ensure that parents are notified of the use of these emergency interventions, as well as debriefing requirements designed to ensure that steps are taken to reduce the need for emergency intervention in the future. In order to achieve these goals, Chapter 33 requires schools to look at aggregate data quarterly in order to identify areas where they could reduce the use of future emergency interventions. But the implementation of Chapter 33 has, for many students, fallen far short of these lofty goals. Prior to this report, a review of special education cases handled by DRM and other advocates around the state yielded many troubling examples of highly repetitive uses of supposedly “emergency” interventions. For example:  A 10 year old student with autism was secluded over 75 times by a special purpose private school over a 3-month period. Five of these lasted longer than 45 minutes. 13 National Center for Trauma-Informed Care and Alternatives to Seclusion and Restraint (last updated Sept. 16, 2015), available at https://www.samhsa.gov/trauma-violence/seclusion 14 Id. 15 To be clear, DRM believes that there are problems with Chapter 33. For example, it is DRM’s position that seclusion should be a prohibited practice in all educational and therapeutic settings. But a full exploration of these concerns is outside the scope of this report. 3    A 6 year old in a public day treatment program was secluded at least once per day for 25 out of 46 school days, for a total of 8 hours spent in seclusion during a 9 week period. A student with autism placed at a special purpose private school was subjected to 98 uses of restraint and seclusion from grades 4 through 6 over a three-year period. In a public day treatment program, a student was placed in seclusion for 45 minutes while staff stood outside the door and counted to two hundred and eleven – the number of times the student hit his head on the metal door frame. In cases like these, DRM believes that the repeated use of restraint and seclusion represents a programmatic failure. That is, the failure to make necessary environmental modifications or provide appropriate services and supports to address behaviors before they escalate to a point where physical intervention may be justified. While there will be emergent situations, it is incumbent on professionals to respond to those situations in a way that reduces the likelihood of the need for physical intervention in the future. This approach is required by Chapter 33, in keeping with the recognition that these interventions, which have no demonstrated therapeutic or educational benefit, are potentially dangerous and harmful. In its analysis, DRM sought to look beyond anecdotal information, and directly examine the data schools are required to collect on emergency interventions pursuant to Chapter 33. As this report makes clear, there appear to be a lot of “emergencies” taking place in Maine schools. Over the past four years, the average documented and reported use of emergency interventions was over 13,000 per year. As discussed in further detail below, failure to make required reports and inaccuracies in reporting suggest that the actual number is no doubt higher. WHAT DOES THE AVAILABLE CHAPTER 33 DATA TELL US? DRM initially reviewed four years of Chapter 33 data with the goal of determining whether the use of physical restraint was increasing or decreasing in Maine schools, or whether additional conclusions on the use of emergency interventions could be drawn. Using data provided to MDOE, reported instances of physical restraint and seclusion over time is illustrated in Figure 1: 4 Figure 1 Restraints and Seclusions Reported to MDOE 18000 16000 14000 12000 10000 8000 6000 4000 2000 0 Total # Restraints Used Total # Seclusions Used 2013 2014 2015 2016 The data in Figure 1 shows that the total uses of restraint and seclusion reported to MDOE ranged between 11,600 and 16,500 during the four reporting years. When compared to the total student population reported (average of 166,000 students per year), there were 83 uses of restraint or seclusion for every 1,000 students. Based on the data provided, only three of the state’s thirty largest districts reported a decrease in total uses of restraint/seclusion each year for 2013-2016. Meanwhile, two of the thirty largest districts reported yearly increases in total uses of restraint/seclusion during the four-year period. The remaining twenty-five districts reported numbers that rose and fell over the years. Drawing conclusions from this data, however, is problematic. As discussed later in this report, the available data is incomplete and unreliable, and cannot be used to identify trends or draw conclusions. Despite the clear reporting requirements under Chapter 33, we still do not know how many students are subjected to physical restraint or seclusion in Maine each year. STUDENTS WITH DISABILITIES ARE SIGNIFICANTLY MORE LIKELY TO BE RESTRAINED AND SECLUDED. According to the most recent national data, students with disabilities who are served under the Individuals with Disabilities Education Act (“IDEA”) were subjected to 67% of the uses of physical restraint and seclusion.16 To put that number into context - these students represent only 16 OCR, 2013-2014 Civil Rights Data Collection: A First Look, 4 (updated Oct. 28, 2016), available at https://www2.ed.gov/about/offices/list/ocr/docs/2013-14-first-look.pdf 5 12% of the K-12 population.17 As indicated in Figure 2, this disparity is significantly more pronounced in Maine, where 86% of the reported uses of physical restraint and seclusion involved a student receiving services under the IDEA.18 This number is sobering, considering that students with disabilities make up approximately 16% of Maine’s total student population.19 In addition, Maine schools have significantly underreported data on seclusion and restraint to the federal government, which masks the true extent of the grossly disparate rates of seclusion and restraint for students with disabilities. 20 Figure 2 Maine Students Restrained or Secluded: USDOE Data 2013-2014 All Other Students 14% IDEA-Eligible Students 86% While we have concluded, as discussed below, that the aggregate data is unreliable in several respects and that it is not possible to say whether restraints and seclusions are increasing or decreasing, the data does make clear that students with disabilities are subjected to restraint and seclusion at significantly higher rates. Although Chapter 33 does not require aggregate reporting of restraint and seclusion to be disaggregated by disability status, DRM reached this conclusion by reviewing the OCR data referenced above, as well as data from schools and programs that only serve students with disabilities. Students with disabilities may occasionally be placed outside of their school district in a special purpose private school (“SPPS”). These schools are established specifically to serve children 17 Id. U.S. Dept. of Education Office for Civil Rights, Civil Rights Data Collection, School / District Search, available at http://ocrdata.ed.gov/DistrictSchoolSearch (last visited Mar 28, 2017). 19 Id. 20 A comparison of 2014 data reported to MDOE with data reported to OCR revealed that only 33% of the seclusions and restraints reported to MDOE were also reported to the federal government. There could be several reasons for this. But the most likely appears to be that many of the programs where seclusion and restraint are used most often (special purpose private schools) do not report directly to OCR. It would appear that the districts placing the students in these programs did not include them in federal reporting. 18 6 with disabilities, and all students in these schools are eligible for and receiving special education services under IDEA. According to reported data, an average of 826 students are educated in special purpose private schools in Maine each year. Though SPPS students made up just 0.50% of all 166,000 students (average) reported for the four years, they endured 55% of the total restraints/seclusions.21 This disparity is shown in the following charts (Figures 3 and 4). Figure 3 Uses of Restraint and Seclusion: SPPS vs. All Other Schools All Other 45% SPPS 55% Figure 4 Enrollment Data: SPPS vs. All Other Schools SPPS 0.50% All Other 99.50% Imagine a class of 25 students. If the students in that class were subjected to restraint and seclusion at the rate reported by the SPPS who reported to MDOE, 224 restraints/seclusions would be used over the course of a school year. But if the students were subjected to restraint and seclusion at the rate reported by all other covered entities, there would only be 1 use of 21 Total restraint/seclusion for all students 2013-2016 was 53,926 versus total restraint/seclusion for SPPS students was 29,622. 7 restraint or seclusion over the course of that same school year. This disparity is demonstrated in Figure 5, below. Figure 5 In addition, DRM requested quarterly building-level data from school districts across the state. And the data confirmed that some school based day treatment programs, which serve only students with disabilities, had significantly disproportionate rates of seclusion and restraint. In one school district, the day treatment program served about 1% of the students district-wide, but was the site for 97% of the reported uses of seclusion and 80% of reported restraints. In another district, a day treatment program served less than 1% of the students district-wide but was the site of 99% of the uses of seclusion and 86% of the restraints in the district. In another district, a small special education program (likely less than 0.5% of the district population), reported 37% of the restraints and 51% of the seclusions district-wide. Given the fact that the Chapter 33 data we have is limited as described above, and given that aggregate reporting does not require data to be disaggregated by disability status, it is not possible to identify the extent to which students with disabilities are subjected to disproportionate uses of restraint and seclusion. Based on the data reported to OCR, the analysis of rates of restraint at programs that only serve students with disabilities, and a review of a sampling of quarterly building-level data, the disparities are profound. 8 CHAPTER 33 DATA IS INCOMPLETE All entities that own, operate, or control a school or educational program funded by MDOE, referred to as “covered entities” in Chapter 33, are required to report the use of restraint and seclusion to MDOE each year. 22 To date, 30% of the required annual reports from covered entities to MDOE have not been made.23 This missing data accounts for at least 10% of Maine’s students. Some covered entities have failed to report any data for any of the four years where reporting was required. Even Child Development Services (CDS), an entity run by MDOE, only reported restraint and seclusion data for one year, 2013.24 The failure of covered entities to report required data to MDOE was identified as an area of concern when the first round of data was released.25 And while reporting rates have improved, as indicated in Figure 6, dozens of covered entities continue to fail to report required data. Figure 6 Percentage of Covered Entities Reporting to MDOE 100% 88% 80% 76% 60% 40% 64% 54% 20% 0% 2013 2014 2015 2016 Chapter 33 defines the term: “Covered Entity means an entity that owns, operates or controls a school or educational program that receives public funds from the Maine Department of Education…” 05-071 CMR Ch. 33, § 2(4). 23 This is a conservative estimate. Although MDOE indicates in its 2016 public report that it operates with a list of 256 covered entities (259 minus 3 that make up a single SPPS), DRM chose to use a more conservative number of covered entities, 235, as the average for 2013-2015. This number is taken from MDOE’s Directory of Schools, found at: http://www.maine.gov/doe/schools/directories.html. Public schools under a common district and SPPS under common ownership are treated as a single entity. 24 DRM has received reports that in some cases, CDS personnel do not believe their programs are subject to Chapter 33 requirements. But Chapter 33 is clear that CDS and all early childhood programs that receive funding from the state are covered entities if they serve children age 3 and above. 25 Christopher Cousins, Data from schools show widespread use of restraint and seclusion, but validity of numbers debated, The Bangor Daily News (2013), https://bangordailynews.com/2013/09/26/education/data-from-schools-show-widespread-use-of-restraintand-seclusion-but-validity-of-numbers-debated/ (last visited Mar 16, 2017) (the article notes that in 2013, 65 of more than 190 school districts failed to report the required data). 22 9 CHAPTER 33 DATA IS UNRELIABLE In addition to the failure of covered entities to fulfill their reporting obligations, we identified several areas in which the numbers that have been reported are unreliable. To reach this conclusion, we compared the statewide data reported to MDOE with building-level data obtained by DRM, as well as data on restraint reported to the federal government. DRM also compared verified uses of restraint and seclusion in individual case files to the numbers reported to MDOE. Examples of underreporting identified by this analysis include:26    One school district reported fewer than 150 uses of restraint and seclusion to MDOE in one year. But data provided directly to DRM indicated that a program for students with disabilities had been excluded from those numbers. When that program was included, the actual restraint and seclusion numbers exceeded 1,500 - ten times higher than the number reported to MDOE. Another school district reported approximately 20 uses of restraint and seclusion to MDOE in one year. But data obtained by DRM indicated the exclusion of one of the elementary schools, accounting for over 50 additional uses of restraint and seclusion. DRM reviewed the restraint and seclusion reports for a student who was secluded 250 times in one school year. Comparing that file with the totals reported to MDOE, DRM concluded that the district underreported its seclusion numbers. This was true even when making the assumption that all other students subjected to seclusion were only secluded once. DRM also reached out to other advocates and attorneys who represent students to identify additional potential areas of concern with reporting under Chapter 33. The information provided confirmed anecdotal data from DRM’s own cases and raised concerns about underreporting in two areas: 1) the failure to consistently report the involuntary movement of a student from one place to another as a restraint; and 2) the apparently widespread belief that using an adult to block a student from leaving a room means that student is not being secluded.27 In these areas, some schools are utilizing interventions that meet the definition of seclusion and restraint, but do 26 Note: We have not used the names of specific school districts or programs in this report. And we have used language designed to limit connecting a specific district to these examples, i.e. “over 50” and “approximately 20”. We made this decision after reaching conclusions about the reliability of the data as a whole and wanted to avoid any focus on a particular school or program that was complying with its duty to report under Chapter 33, while an entity that did not report at all would escape scrutiny. 27 In addition to the belief that using an adult as a door to prevent a student from leaving a seclusion room means a student is not alone and thus not secluded, DRM has recently received a report that at least one school is placing a staff member in a seclusion room with a student for the sole purpose of circumventing the seclusion reporting requirements. In that case, two staff members were outside holding the door to the seclusion room shut while the student was pushing against the door and struggling to get out. The parent was told it was not seclusion because a staff member had been placed in the room with the student to observe. If Chapter 33 does not adequately prohibit this type of intervention, then the Chapter 33 seclusion definition should be revisited. 10 not call them seclusion or restraint, and are therefore not complying with Chapter 33 reporting requirements. Based on DRM’s analysis, the data reported to MDOE does not accurately reflect the number of emergency interventions being used in educational programs across the state. There is clear evidence of underreporting. SCHOOLS MUST REVIEW DATA QUARTERLY TO IDENTIFY WAYS TO REDUCE THE USE OF SECLUSION AND RESTRAINT. Chapter 33 requires that restraint and seclusion incident reports are aggregated and reported, by building, to the superintendent or chief administrator on a quarterly basis, who “shall review cumulative reports received…and identify those areas that can be addressed to reduce the future use of physical restraint and seclusion.”28 This provision seeks to ensure that the use of emergency interventions is analyzed by school administration in order to provide support to schools or programs with a high number of emergencies. Properly implemented, this provision would appropriately treat repeated uses of emergency interventions as an indicator of the need for changes in programming and support, as well as support and training. It does not appear that districts are taking seriously their responsibility to analyze the cumulative data. DRM specifically requested any and all documents related to the administrative review of quarterly building-level data and to any efforts to identify those areas that can be addressed to reduce the future use of physical restraint and seclusion. This request went to 43 school districts across the state, but only one document was produced and it was not responsive to the request.29 There were some narrative responses that included: an expression of intent to speak with building principals; the hope that recently hired behavioral interventionists would reduce the use of emergency interventions; a statement that administrators are asked to review data and to discuss hypotheses regarding the data and any warranted next steps; assertions that increased training helped staff become better at recognizing warning signs; and several identical responses that “with respect to subsection (d) of your request, the School District complies with its obligations under Chapter 33.” DRM believes that if schools were taking this requirement seriously, there would be some documentary evidence of those efforts. 28 05-071 CMR Ch. 33, § 10.1. That document was related to a staff training on the requirements of Chapter 33 and unrelated to the building-level administrative analysis required by Chapter 33. 29 11 RECOMMENDATIONS Based on our analysis, DRM makes the following recommendations: 1) MDOE should continue its efforts to ensure that 100% of all covered entities submit annual data as required by Chapter 33 and that there are uniform standards for reporting so comparisons can be made between covered entities. 30 Related to this, MDOE should ensure that students placed in special purpose private schools are represented in the federal reporting to OCR because current reporting to OCR appears inadequate. 2) MDOE should ensure that schools across the state are provided support in adopting positive behavioral interventions and supports and in taking other steps aimed at reducing the need for the use of emergency interventions. And MDOE should provide targeted and intensive support to schools and programs with high rates of seclusion and restraint to ensure that appropriate efforts are taken to reduce reliance on emergency interventions. 3) All covered entities should review building-level data quarterly and develop plans to reduce the use of emergency interventions, as required by Chapter 33. 4) MDOE should consider exercising its existing authority pursuant to Chapter 33 to request quarterly building-level data from entities with high rates of seclusion and restraint in order to better target professional development and other state level resources. 5) To facilitate access and analysis, MDOE should add restraint and seclusion data to MDOE’s “Data Warehouse,” alongside the numbers for discipline, enrollment, program enrollment, special education, and student attendance. 6) Given the significantly disparate rates of seclusion and restraint for students with disabilities, Chapter 33 should be amended to require that the quarterly data and the annual data be disaggregated for disability status (whether the student has an IEP or a 504 plan). This is already a requirement for the Chapter 33 incident reports. And it is already 30 DRM would like to note that Dr. Robert Hasson, the Commissioner of the Maine Department of Education, reached out to set up a meeting in March 2017, after a preliminary analysis of the Chapter 33 data was released on our website. MDOE acknowledged past issues with reporting under Chapter 33 and made clear that efforts were underway and ongoing to address this. MDOE indicated that although the position responsible for supporting the implementation of Chapter 33 had been vacant for an extended period of time, that position has now been filled, allowing for a renewed focus on these issues. In fact, just as we were about to release this report, MDOE provided updated reporting numbers for 2016 as a result of these efforts. The report has been updated accordingly. Dr. Hasson’s commitment to ensuring that Chapter 33 reporting requirements are met and that schools are supported in reducing the use of emergency interventions was clear. And this commitment is clearly necessary. 12 required for federal reporting. Additional categories, including gender and grade level, should also be included. These are also already required for Chapter 33 incident reports and for federal reporting. 7) MDOE should convene a stakeholders group to examine the high rate of use of seclusion and restraint with students with disabilities and to develop a plan to reduce reliance on these emergency interventions. 13