A Study of One State’s School District Physical Restraint and Seclusion Policies

Abstract

This descriptive analysis of policy content examined local school district policies on physical restraint and seclusion in one Midwestern state that did not have state legislation on these topics to determine whether districts had policies and, if so, whether their content included recommended principles from the U.S. Department of Education. No previous research has addressed district policies. A maximum variation sample of 90 districts was examined to determine whether policies were in place, whether recommended principles were included, and whether policies varied based on student enrollment. Although almost all districts had policies, many of the federally recommended principles were not addressed across the sample and less than 10% of the district policies indicated that these procedures should only be used in the case of imminent danger of serious injury to self or others. District enrollment size did not affect policy, but the substance of the policy was determined by the districts’ policy source from advising attorneys.

Keywords

physical restraint seclusion district policy

Children and adolescents have continued to be killed or seriously injured by the use of physical restraint procedures in schools and treatment settings (Eldeib, 2016; Kentucky Protection and Advocacy, 2016; U.S. Government Accountability Office [U.S. GAO], 2009). Although there is no official count of deaths or injuries as a result of physical restraint, a recent report that scanned media news accounts indicated there were 28 deaths of children and youth between 2003 and 2017 as a result of physical restraint procedures, with seven deaths between 2013 and 2017 (Holden & Nunno, 2019). Advocacy organizations have brought suit against public schools for lack of adequate training, lack of evidence-based prevention interventions, and lack of supervision leading to the increased use of physical restraint in special education programs (e.g., Kentucky Protection and Advocacy, 2016; Pillsbury & Disability Rights Legal Center, 2018; Samuels, 2017). The U.S. Department of Education (U.S. DOE) Office for Civil Rights (OCR) and some states have required districts to supply data on their use of physical restraint and seclusion procedures, but differing definitions, ambiguity in distinguishing high-risk situations and behavior, and inconsistency in gathering of these data has made interpretation of the federal data difficult. Moreover, data submitted to the OCR have substantially underreported the actual use of these procedures in schools (Abamu & Manning, 2019; Morrison, 2015; U.S. GAO, 2020).

Currently, there is no system to identify deaths or injuries to students or staff as a result of these procedures, nor has there been any reliable source of data collected on how often these procedures are used, or for what reasons. Most reports have indicated that restraint and seclusion have been employed most often with students with emotional or behavior disorders, and with students on the autism spectrum, although they have been used with students with other disabilities and students without disabilities (Gage et al., 2020; Peterson et al., 2013; Westling et al., 2010). Reported abuse, overuse, and discriminatory application of these procedures toward certain populations (e.g., students with disabilities, students of color) has continued to be problematic (OCR, 2016). This is despite the federal and state governments’ efforts over the past decade to promote policy development to improve the safe and appropriate use of physical restraint and seclusion in schools. Although there have been a few studies of state policies on these topics (e.g., Butler, 2019; Freeman & Sugai, 2013; Marx & Baker, 2017; Ryan et al., 2009), to date no studies have addressed local school district policies. The purpose of this descriptive analysis of policy content was to determine the extent to which local district policies are in place in one Midwestern state that has no state-level policy and, if so, whether the local policies include federally recommended policy principles. The current study also examined whether policy content varies by district enrollment.

Policy Guidance

Federal Policy

In 2009, former Secretary Arnie Duncan sent a letter to chief state school officers (Duncan, 2009), indicating that all states and school districts should develop and implement policies to improve safety and prevent the need for use of restraint and seclusion procedures. In 2012, the U.S. DOE issued a “guidance document” on physical restraint and seclusion, which identified and recommended 15 principles to be addressed when school districts created policies on this topic. This 40-page technical assistance document, which included a review of literature, was also intended to provide guidance to school districts on how to use function-based intervention and school-wide positive behavioral supports to prevent the need for restraint and seclusion. The 15 principles addressed within the document were developed in collaboration with the Substance Abuse and Mental Health Services Administration (SAMHSA) that had previously developed the evidence-based Six Core Strategies© model to prevent the use of restraint and seclusion in health care settings (U.S. DOE, 2012). In addition, the U.S. DOE guidance document included recommendations for good practice on this topic from several professional organizations, including the Council for Children With Behavior Disorders (2009). Both Secretary Duncan’s letter and the U.S. DOE guidance document indicated that “physical restraint or seclusion should not be used except in situations where the child’s behavior poses imminent danger of serious physical harm to self or others . . .” (U.S. DOE, 2012, p. 14). In December of 2016, the OCR wrote a Letter to Colleagues, which reported on how the use of restraint and seclusion in public schools may result in discrimination against students with disabilities. The OCR encouraged adherence to the U.S. DOE guidance document to ensure both equal access to education and that students with disabilities are not subjected to discriminatory practices.

Several federal bills have been proposed in the U.S. Congress since 2009, including one in the current Congress, to regulate the use of restraint and seclusion in schools. These bills have all included content similar to the principles outlined in the U.S. DOE guidance document. None of these bills have become law. However, Congress has also shown concern for this issue in other ways, albeit indirectly. The Every Student Succeeds Act (ESSA) in 2015 required state plans to address how to diminish the use of aversive procedures, with the Congressional Conference Committee Report indicating that aversive behavioral interventions included physical restraint and seclusion (H.R. Rep. No. 114–354, 2015). The U.S. House of Representatives’ budget amendments passed in 2017 requested that the U.S. GAO conduct an updated study of the use of restraint and seclusion. Finally, in January 2019, the U.S. DOE indicated that it would provide technical assistance to states and schools on this topic.

State Policies

Over the past decade, many states have passed legislation on this topic and others have implemented administrative policies on the use of restraint and seclusion in schools. Butler (2019) indicated that, as of July 2019, 30 states had laws regulating restraint and seclusion for all children, and 39 states had laws addressing students with disabilities. According to a recent analysis by Marx and Baker (2017), policy and legislation by states varied in their adherence to the 15 principles set forth by the U.S. DOE, which they indicated continue to be the most helpful source of federal guidance for states attempting to improve policy and reduce the use of restraint and seclusion in schools. Their analysis indicated that state policies typically aligned more closely with the 15 principles compared with state legislation.

As part of a larger report discussing reported use of restraint and seclusion, Gagnon et al. (2017) analyzed trends in OCR data for school districts on the use of physical restraint and seclusion procedures during the 2009–2010 and 2011–2012 school years. However, OCR data on the use of restraint and seclusion has been criticized as significantly underrepresenting the actual use of these procedures (Morrison, 2015; U.S. GAO, 2020). Nevertheless, Gagnon et al. found that, although there were differences across states regarding the annually reported incidents of physical restraint and seclusion, there were far more significant differences in the frequency of use between districts within states. The authors concluded that “state policies on restraint and seclusion may only have a modest impact on how frequently the practices are conducted” (Gagnon et al., 2017, p. 73). Under this assumption, the use of restraint and seclusion in schools “is likely driven to a meaningful degree by local policy and school culture more generally” (Gagnon et al., 2017, p. 73). These findings were corroborated in a study conducted by Gage et al. (2020), which also analyzed OCR data trends across state, district, and school levels for the 2015–2016 school year. Thus, local school district policy may be a critical factor in determining how these procedures are utilized in schools (Gage et al., 2020; Gagnon et al., 2017).

Local District Policies

Even with state legislation or policy (which may vary in its guidance to districts), school districts are expected to have policies regarding physical restraint and seclusion procedures. However, there have been no empirical studies of school district policies on this topic. As a result, an analysis of the content of local district policies is not only warranted but is likely essential to understand the use of these procedures at the local level.

Purpose

The purpose of the current descriptive analysis of policy content in one state was to analyze whether local school district policies on restraint and seclusion are in place in a state that does not have state legislation on this topic and to what degree they address the safety principles contained in federal guidance. An additional purpose was to determine whether the principles contained in policies vary based on district enrollment size as districts with more students might be hypothesized to have greater need for such policies and a higher likelihood of employing physical restraint or seclusion procedures compared with smaller districts with fewer students.

Research Questions

This study specifically addressed three research questions:

Research Question 1 (RQ1): What percentage of the sample of the state’s school districts have policies governing the use of restraint and seclusion?

Research Question 2 (RQ2): To what extent do existing policies address the recommended principles in the U.S. DOE guidance document?

Research Question 3 (RQ3): Are there differences in the number of principles included in district policies based on district enrollment size?

Method

Procedures

In this section, we describe our refinement of the principles in the U.S. DOE document, our selection of a sample of districts from one state, and our methods for obtaining and analyzing the policies identified.

For the present study, we have chosen to maintain the numbering of principles that were included in the U.S. DOE guidance document for easy reference to that document. However, we did refine the list of 15 principles by excluding three of these and breaking four of the federal principles into two parts each (identified as A and B) based on the detailed discussion of the principles within the U.S. DOE guidance document.

Three of the federal principles were unlikely to be found in the “physical restraint and seclusion” sections of local district policies and, as a result, were intentionally not addressed in the current study. While they might be included elsewhere in district policy, it was beyond our ability to locate and identify them. These three DOE principles were, first, DOE No. 4, Policies . . . should apply to all children, not just children with disabilities. All school district policies normally are presumed to apply to all children in the district and an initial review of the sample district policies found no specific language was included within policies regarding application to all children or only to children with disabilities. Second, DOE No. 5, Any behavioral intervention must be consistent with the child’s rights to be treated with dignity and to be free from abuse, was similarly not included. Treating students with dignity and respect without abuse are concepts that might be included within a preamble or within other general policies of a school district that apply broadly to all aspects of education. Third, DOE No. 14, Policies . . . should be reviewed regularly and updated as appropriate. As most school districts engage in review and update virtually all of their policies on an annual or semiannual basis, this principle would not likely be documented within the more substantive part of their policy on physical restraint.

In addition, four of the DOE numbered principles, as described in the extended discussion in the U.S. DOE guidance document, addressed more than one concept. Thus, for the present study, these principles were each split into two parts to more precisely analyze content of district policies. The DOE number was maintained with an A or B, indicating the two parts within the original DOE principle. Based on extended discussion, DOE No. 1 in the guidance document, Every effort should be made to prevent the need for the use of restraint and for the use of seclusion, was divided into Principle 1A “A focus on prevention” and Principle 1B “Use of positive behavioral interventions for prevention.” Furthermore, DOE No. 2, Schools should never use mechanical restraints to restrict a child’s freedom of movement, and schools should never use a drug or medication to control behavior or restrict freedom of movement . . ., was divided into Principle 2A “No use of mechanical restraints,” and Principle 2B “No use of medication restraint.” Similarly, DOE Principle No. 3, Physical restraint or seclusion should not be used except in situations where the child’s behavior poses imminent danger of serious physical harm to self or others and other interventions are ineffective and this should be discontinued as soon as imminent danger of serious physical harm to self or others has dissipated, was split into Principle 3A “Situations for permitted or warranted use were identified in the policy” and Principle 3B “Emergency procedures only permitted for imminent danger of injury.” Finally, DOE No. 11, Every instance in which restraint or seclusion is used should be carefully and continuously and visually monitored to ensure the appropriateness of its use and safety of the child, other children, teachers and other personnel, was split into Principle 11A “Use must be monitored for safety of the student” and Principle 11B “Debriefing to prevent future occurrences.” Debriefing was specifically mentioned in the extended discussion of DOE No. 11 in the DOE guidance document.

In summary, three DOE principles were not addressed in this study, and four other principles were split into two principles each labeled as A and B. As a result, a total of 16 policy principles based on the original U.S. DOE list of 15 principles were included in the current analysis (see Table 1), with numbers referenced to the original DOE principles.

Table 1.

U.S. Department of Education (DOE) Policy Principles and Definitions as Modified for This Study.

Guiding principle	Definition for policy review
1A. Focus on prevention	Policy explicitly states every effort should be made to prevent the need for the use of restraint and seclusion.
1B. Positive behavioral strategies identified	Policy explicitly states that positive behavioral strategies should be employed to prevent and/or address dangerous behavior.
2A. Mechanical restraint	Policy states if/when the use of mechanical restraints is permitted.
2B. Medical restraint	Policy states if/when the use of medical restraints is permitted.
3A. Situations for permitted or warranted use identified	Policy explicitly states when the use of restraint and seclusion is permitted or warranted, such as when danger is present, in the event of property damage, written into student’s IEP, and so on.
3B. Emergency procedures only permitted for imminent danger of injury	Policy explicitly states that restraint and seclusion are emergency procedures only to be used when a student’s behavior presents imminent risk of serious injury to self or others and does not permit use for other situations.
4. Not addressed in study	Policies apply to all children, not just to children with disabilities.
5. Not addressed in study	Any behavioral intervention must be consistent with the child’s rights to be treated with dignity and to be free from abuse
6. Restriction of use for purposes of discipline and coercion	Policy explicitly states when the use of physical restraint and seclusion is not allowed; or prohibited for purposes of discipline, punishment, coercion, or as a means of convenience.
7. Restriction of use so as to not harm child	Policy explicitly states specific holds or procedures that are prohibited to ensure the safety of students.
8. Supervision, oversight, and review	Policy explicitly states that accurate records will be kept for each incident of restraint and seclusion.
9. Cause or function of behavior identified	Policy explicitly states that behavioral strategies employed should address the underlying cause or purpose of the behavior.
10. Staff training	Policy explicitly states which staff are permitted to use restraint and seclusion. Policy may state recertification requirements.
11A. Use monitored for safety	Policy explicitly states that each restraint or seclusion application should be continuously visually monitored to ensure the appropriateness of its use and safety of the child and others.
11B. Debriefing	Policy explicitly states that debriefing will occur after every incident of restraint and seclusion.
12. Guardians informed of policy	Policy explicitly states that parents and guardians will be informed of school or district policies on restraint and seclusion.
13. Reporting incidents of use to guardians	Policy explicitly states when and how parents/guardians will be notified following use of restraint or seclusion with their child.
14. Not addressed in study	Policies . . . should be reviewed and updated regularly.
15. Documentation of incidents	Policy explicitly states that each incident involving the use of restraint or seclusion will be documented in writing.

Note. Definitions provided are shortened versions of those used for rating policies. The complete rating rubric and definitions of principles may be obtained by contacting the first author. IEP = Individualized Education Program.

Sample of district policies

As no previous research has addressed local school district policies, a sample of school districts within one Midwestern state was utilized as the source of the policies analyzed in this exploratory study. Although this may limit generalizability, it permitted testing of policy analysis procedures and it avoided potentially complex differences in policy process across states that would make interpretation of data more complex. The state chosen for this analysis is one of approximately 25% of states with no state laws on this topic. The only state regulation on this topic in the current state is a state school accreditation requirement that districts should have restraint and seclusion policies in place. This regulation does not provide any specific requirement regarding policy content. This allowed analysis of district policies in the absence of state statute or regulation, which had been suggested as important by advocacy organizations (Butler, 2019).

A sample of 90 school districts was selected out of the state’s 244 public school districts represented in enrollment figures for the 2016–2017 academic year. To represent the spectrum of district enrollment size, a “maximum variation sampling” procedure was employed (Patton, 1987). To do this, the 30 school districts with the largest student enrollment (range of 1,465–52,344 students) and the 30 districts with the smallest student enrollment (range of 51–175 students) were included in the sample. In addition, 30 school districts were identified and included by calculating the mean population across all school districts in the state (M = 453 students), and then selecting the 15 districts falling immediately above and immediately below the state mean enrollment (medium group student enrollment of 411–529). These 90 school district policies represented the sample of districts for this study.

Data Gathering and Analysis

Obtaining district policies

The authors searched for each of the 90 school districts’ restraint and seclusion policies through their district websites or through direct contact with the district special education director or the intermediate unit to which that district belonged. Fifty-six district policies were found and obtained directly from district websites, and 29 were collected through direct contact. Two district respondents indicated that their districts did not have policies at the present time. Three districts did not respond, and their policies, if they had them, could not be obtained. As a result, 85 districts representing 94% of the 90 selected districts were included in the current analysis.

Coding procedures

Operational definitions were developed for each of the 16 principles in the current study to provide a more objective measure of the presence and depth of the principles in each district policy. Two of the coauthors thereafter examined 14 policies to adjust definitions in an iterative process resulting in a rating rubric.¹ An abbreviated version of this definition rubric is included as Table 1. Policies were scored as “0” if a principle was not present in the policy. If the principle was present within the policy language, the raters then judged completeness of that principle scoring a “1” if it was partially articulated or if key aspects were not included, and a “2” was scored if the principle was judged to be fully articulated in the policy. The third coauthor was then trained on the scoring rubric by the two coauthors who had created the rubric. All three coauthors were advanced doctoral students in school psychology or special education.

Some identical policies were found (as discussed in the following). Each distinct policy that was not identical to another was then rated on all 16 principles separately by at least two of the coauthors, utilizing the established rubric. In total, 25 of the 85 district policies (29%) were independently coded across three of the study coauthors. Principles for which all three raters assigned the same rating were identified as agreements. Interobserver agreement (IOA) percentage was calculated for each principle by dividing the number of agreements by the total number possible, then multiplied by 100. The final mean of IOA across all principles was 95% (range, per principle, was 89%–100%).

Results

As soon as the analysis of the district policies began, it became apparent that many districts had similar if not identical policies. These identical policies were sorted into five “groups.” Seventy-eight districts (92%) of the total sample were represented across the five groups, with a range of three to 25 districts within each policy group. We surmised (and later confirmed) that school districts obtain policy content guidance from legal firms specializing in district policy support and/or from organizations such as the state administrator’s organizations. This unanticipated finding suggests that many districts do not create their policies independently, but that instead they rely on policy advisers to supply core sets of policies to their client school districts.

Many of the districts that adopted these adviser-created core policies made minor alterations so that the policy reflected the programs and procedures employed by the district, such as indicating specific positive behavioral strategies implemented (e.g., Positive Behavioral Interventions and Supports [PBIS]) or by indicating the specific vendor provided crisis intervention training program used for staff training and certification (e.g., The MANDt System^® or Crisis Prevention Institute [CPI]). Alterations such as these did not change the principle ratings of the policy, so they were not identified as unique policies. Two districts altered policy content to the extent that principle ratings were substantially different from the adviser-created core policy (e.g., adding or removing types of situations in which restraint or seclusion would be warranted). The authors decided that district policies in which codes changed on four or more of the principles from the core policy would be considered a substantial difference. District policies with substantial differences, as well as district policies that were independently developed by the school district were identified as “unique” policies. Seven district policies were identified as “unique” and were grouped together for analysis purposes (labeled as Group 6: Unique). Table 2 displays the number of districts in each group as well as how these groups were spread across districts with large, medium, and small enrollment sizes.

Table 2.

District Study Sample Divided Into Policy Groups by District Size.

Policy group	Group as % of sample		District size within group
	Group as % of sample		Large		Medium		Small
	n	%	n	%	n	%	n	%
Group 1	5	6	3	60	1	20	1	20
Group 2	25	29	6	24	10	40	9	36
Group 3	22	26	7	32	4	18	11	50
Group 4	23	27	3	13	14	61	6	26
Group 5	3	4	3	100	0	0	0	0
Group 6: Unique	7	8	6	86	0	0	1	14

Note. District size based on student enrollment.

RQ1: What Percentage of the Sampled Districts Have Policies?

As noted earlier, only two districts out of 90 (2.2%) indicated that they did not have policies. Within three additional districts, the existence of a policy could not be determined (3.3% of the sample). Based on this sample, 94.5% or more school districts had policies.

RQ2: To What Extent Do District Policies Address Federal Principles?

Presence of principles

As displayed in the second and third columns of Table 3, the number of policy principles addressed by each district policy ranged from five principles present (31% of principles) in two district policies, to all 16 principles (100%) present in five district policies. On average, 10 principles (61%) were addressed by the district policies across the sample. The percentage of all districts that addressed each individual principle ranged from one principle (No. 3A Situations for Warranted Use Identified) addressed by all 100% of school districts (n = 85) to another principle (No. 3B Emergency Procedures Only Permitted for Imminent Danger of Injury) addressed by only 9% of districts (n = 8). A total of nine principles were addressed by over half (69%–100%) of the district policies. The remaining seven policy principles were present in 40% or fewer of the district policies, with three principles (No. 1B Positive Behavioral Strategies Identified, No. 11B Debriefing, and No. 3B Emergency Procedures Only Permitted for Imminent Danger of Injury) addressed in less than 15% of the policies.

Table 3.

Percentage of Policies That Addressed Each Principle and Whether Fully or Partially Articulated.

Principles addressed in policies from greatest to least	n	% Present	% Fully articulated	% Partially articulated
3A. Situations for warranted use identified	85	100	73	27
10. Staff training	84	99	68	32
8. Supervision, oversight, and review	83	98	43	57
11A. Use monitored for safety	81	95	32	68
6. Restriction of use for discipline and coercion	80	94	100	0
7. Restriction of use so as to not harm child	80	94	40	60
15. Documentation of incidents	61	72	95	5
1A. Focus on prevention	60	71	60	40
13. Reporting incidents of use to guardians	59	69	95	5
9. Cause or function of behavior identified	34	40	97	3
2A. Mechanical restraints	33	39	100	0
2B. Medical restraints	32	39	100	0
12. Guardians informed of policy	30	35	87	13
1B. Positive behavioral strategies identified	13	15	100	0
11B. Debriefing	11	13	91	9
1B. Only permitted for imminent danger^a	8	9	100

Note. N = 85. Principles 4, 5, and 14 were not examined.

Component was not rated for level of completeness as it could only be fully articulated or not present based on the operational definition.

Comprehensiveness of principles

Policy principles represented within district policies were further rated to identify the level of completeness (fully present or partially present or not present) for each principle (see Table 3). Principle No. 1B, Positive Behavioral Strategies identified, and No. 3B, Emergency Procedures Only Permitted for Imminent Danger of Injury, were not judged for completeness as these could only be fully present or not present according to the definitions. Guidance on the use of Principles No. 6, Restriction of Use for Purposes of Discipline and Coercion (n = 80), No. 2A, Mechanical Restraint (n = 33), and No. 2B, Medical Restraint (n = 32), were fully present across all the district policies that addressed these principles. The principles that were only partially addressed across the majority of district policies were No. 11A, Use Monitored for Safety (68%, n = 55), No. 7, Restriction of Use so as to not harm the child (60%, n = 48), and No. 8, Supervision, Oversight, and Review (57%, n = 47).

Presence of principles by group

Table 4 displays how these group policies addressed the study’s 16 restraint and seclusion policy principles. Groups 1 and 5 included all or almost all of the 16 principles (94%–100%) within their policies, whereas Groups 2, 3, and 4 included fewer of the principles (50%–69%) and varied in which principles they included. Group 6: Unique varied the most in the inclusion of study policy principles, ranging from two policies, addressing only five principles (31%), to one district policy, addressing 14 principles (88%). The distribution of the number of school districts using these policy groups across district size was considerably balanced.

Table 4.

Percentage of District Policies by Groups That Addressed Each Recommended Principle.

Policy principle	Group 1 (n = 5) %	Group 2 (n = 25) %	Group 3 (n = 22) %	Group 4 (n = 23) %	Group 5 (n = 3) %	Group 6: Unique (n = 7) %
1A. Focus on prevention	100	96	100	0	100	86
1B. Positive behavioral strategies	100	0	0	0	100	71
2A. Mechanical restraint	100	0	100	4	100	29
2B. Medical restraint	100	0	0	100	100	14
3A. Warranted use	100	100	100	100	100	100
3B. Only permitted for imminent danger of injury	40	0	0	0	100	43
6. Restriction discipline/coercion	100	100	100	100	100	29
7. Restriction of use so as to not harm child	100	100	100	100	100	29
8. Supervision, oversight, and review	100	100	100	100	100	71
9. Cause/function of behavior	100	0	0	100	100	43
10. Staff training	100	100	100	100	100	86
11A. Use monitored for safety	100	100	100	100	100	43
11B. Debriefing	100	0	0	0	100	43
12. Guardians informed of policy	100	0	100	0	100	0
13. Reporting use to guardians	100	100	100	0	100	57
15. Documentation of incidents	100	100	100	0	100	86

RQ3: Does District Enrollment Size Affect Number of Principles Addressed?

On average, across each district size (large, medium, and small), large districts addressed 67% of all principles, medium districts covered 58% of principles, and small districts covered 60% of principles in their policies (see Table 5). All districts, regardless of size, provided a purpose for the use of restraint and seclusion (Principle 3A). Within the district size category of small, this was the only principle that was addressed by all responding districts (n = 28). The remaining 15 principles were addressed in 4% (No. 11B, Debriefing; n = 1) to 96% (No. 8, Supervision, Oversight, and Review; No. 10, Staff Training; and No. 11A, Use Monitored for Safety; n = 27) of the small district policies. Within the medium enrollment size category, six principles were addressed by all districts, whereas the remaining principles ranged from 3% (No. 11B, Debriefing, n = 1) to 52% (for No. 2B, Medical Restraint; No. 9, Function of Behavior Identified; No. 13, Reporting incidents of use to guardians; and No. 15, Documentation of Incidents; n = 15). In districts identified as large, there were two principles that were present in all district policies: No. 3A, Situations for Permitted or Warranted Use, and No. 10, Staff Training (n = 28). The percent of districts addressing each of the remaining principles ranged from 32% (No. 11B, Debriefing; n = 9) to 96% (No. 8, Supervision, Oversight, and Review; n = 27).

Table 5.

Percentage of Recommended Principles Addressed and Ranked by District Size Grouping.

	Small			Medium			Large
Principle addressed	n	%	Rank	n	%	Rank	n	%	Rank
3A. Warranted use	28	100	1	29	100	1	28	100	1
10. Staff training	27	96	2	29	100	2	28	100	2
8. Supervision, oversight, and review	27	96	3	29	100	3	27	96	3
11A. Monitored for safety	27	96	4	29	100	4	25	89	4
6. Restriction discipline/coercion	27	96	5	29	100	5	24	86	6
7. Restriction of use so as to not harm child	27	96	6	29	100	6	24	86	7
1A. Focus on prevention	22	79	7	14	48	11	24	86	8
15. Documentation	21	75	8	15	52	7	25	89	5
13. Reporting to guardians	21	75	9	15	52	8	23	82	9
2A. Mechanical restraint	12	43	10	6	21	12	15	54	10
12. Guardians informed of policy	12	43	11	5	17	13	13	46	11
9. Cause/function of behavior	8	29	12	15	52	9	11	39	12
2B. Medical restraint	7	25	13	15	52	10	10	36	13
1B. Positive behavior strategies	2	7	14	1	3	14	10	36	14
3B. Only permitted for imminent danger	2	7	15	0	0	16	6	21	16
11B. Debriefing	1	4	16	1	3	15	9	32	15

Note. Total sample of small districts, n = 28; medium districts, n = 29; and large districts, n = 28.

Discussion

Specific Policy Principles

All district policies in the sample obtained for this study provided guidance on the conditions under which the use of restraint and seclusion were warranted or permitted (Principle 3A). This is not surprising as almost all policy guidance indicates that such procedures be used only in limited situations and describes situations where they should not be employed (e.g., as discipline or punishment).

However, many of the policies in the sample allowed for situations for use that are much broader than suggested by the federal guidelines, such as property destruction, interventions written in student Individualized Education Programs (IEPs), or even in the absence of imminent danger of serious physical injury (U.S. DOE, 2012). As a result, 91% of district policies (n = 77) did not contain specific language that restraint and seclusion were only warranted when imminent risk of serious injury was present (Principle 3B) and had no contraindicated situations in which restraint and seclusion were permitted (e.g., property damage, class disruption). This is probably the most alarming finding of this study. Specifically, the requirement for “clear and imminent danger of someone being injured” is critical to limit the wider use of these procedures and to limit the risk of exposing students to physical and emotional harm unnecessarily. This principle has been stressed by the U.S. DOE guidance document (U.S. DOE, 2012), experts (Peterson et al., 2013; Scheuermann et al., 2015), and a wide variety of other professional organizations.

Sixty district policies in the sample (71%) indicated that school culture and strategies employed within the school district should focus on preventing dangerous behaviors and the need to use restraint and seclusion procedures (No. 1A). However, as seen in Table 3, only 60% of these districts (n = 36) fully addressed this principle. Most districts that addressed this principle provided a general statement that preventive measures would be employed to maintain safety and promote a positive learning environment for students and staff. Notably, less than half of the districts in the sample identified specific procedures to prevent the use of restraint and seclusion, such as identifying the function of dangerous behaviors, employing positive behavioral strategies, or the use of de-escalation techniques. Identification of school-wide positive behavioral systems and methods for evaluating the environmental conditions and factors that contribute to dangerous student behaviors is essential for reducing problem behaviors, teaching appropriate replacement behaviors, and limiting the need to employ restraint and seclusion (Dunlap et al., 2011; U.S. DOE, 2012).

Thirty-four districts (40% of our sample) included guidance on identifying the cause or function underlying student behaviors (Principle 9) that led to the use of restraint or seclusion. Many of the district policies that addressed this principle included guidance on implementing functional behavioral assessments and/or developing behavior intervention plans (BIPs). Some districts only required an evaluation of the function of the dangerous behavior for students receiving special education services. Furthermore, most of the districts that addressed this principle only required staff to identify the function of dangerous student behavior following recurring incidents of restraint or seclusion. Only 15% (n = 13) of the district policies addressed the use of positive behavioral strategies (Principle 1B). This is a drastic difference compared with the 71% of districts in the sample that indicated a focus on prevention (Principle 1A). It is possible that districts felt positive behavioral strategies were implicit in the statement of prevention or identifying the function of dangerous behavior. However, it is important for district policies to provide explicit guidance on the use of positive behavioral strategies to address the underlying function of dangerous behavior, reduce the future need for restraint and seclusion, and maintain student safety. Positive behavioral strategies should be employed on a school-wide, universal level (prior to and regardless of the occurrence of problem behavior), as well as specific, evidence-based intervention strategies included in student BIPs to prevent future occurrences of dangerous behaviors (Dunlap et al., 2011; Miller et al., 2005; Rozalski et al., 2013).

The U.S. DOE guidance document recommends that parents and guardians of all students are informed of the district restraint and seclusion policies. Roughly, one third of the districts (n = 30) in the sample identified how or when guardians would be informed of these policies (Principle 13). The majority of districts that included this principle either stated that the district superintendent would inform parents and guardians of the policies at the beginning of the school year, or they provided a statement about the policy that would be included in the district parent handbook and indicated how guardians could obtain a full copy of the policy. Many districts may have procedures for informing parents and guardians of all school policies or provide guidance for informing guardians of policies in a separate section of the school policies. If this is the case, it may explain why many districts did not specifically address Principle 13 in the restraint and seclusion policy.

Principle 8, Supervision, Oversight, and Review was covered by 98% of the sampled districts (n = 83). However, during the process of operationalizing this principle, it was broadened to include many activities that may be included under supervision, oversight, and review (e.g., collecting data, appointing an administrative staff member responsible for reviewing documentation, and analyzing data patterns). Due to the scope of this category and the limitations in the rating scale, it is not clear whether the level and type of supervision, oversight, and review recommended by experts is reflected in these policies. Table 3 indicates that only 43% of districts (n = 36) fully cover this principle, which indicates that minimal amounts of supervision, oversight, and review may be happening in many districts. All district policies that fully covered this principle indicated that a district- or school-level staff member was appointed to obtain and maintain copies of restraint and seclusion incident reports. However, few policies provided guidance for actual review and oversight of these incidents, such as review of data indicating recurring uses of restraint and seclusion, to inform staff training and district procedures or reporting data on the frequency of incidents of use to district-level personnel or state agencies. Thorough supervision, oversight, and review is critical to ensuring the safety of restraint and seclusion procedures, as well as improving the safety of implementation and reducing their use over time (Miller et al., 2005; Peterson et al., 2013). Ideally, there should be an individual specifically assigned to review the use of restraint and seclusion, including analyzing usage data for patterns of use, misuse, and overuse (U.S. DOE, 2012).

Furthermore, the fact that only 11 districts (13% of the sample) addressed debriefing (No. 11B) also indicates that formal debriefing procedures are likely not occurring regularly in schools. Interestingly, Positive Behavioral Strategies, Principle 1B, was also one of the lowest principles in the sample (n = 13). As mentioned above, perhaps districts do not include this principle as they feel it was covered under content addressing a focus on prevention. Generally, this principle aims to cover districts’ efforts to systematically identify and address the underlying reasons (i.e., function) for behavior that resulted in an instance of employing restraint and seclusion. Ideally, schools should have procedures in place that require a systematic review of the circumstances surrounding each instance of physical restraint and seclusion (Peterson et al., 2013; U.S. DOE, 2012). Repeated use of these procedures should lead practitioners to conduct a functional assessment of the target problem behavior and develop or modify a BIP that includes implementation of positive behavior strategies that address the student’s specific needs (Dunlap et al., 2011; U.S. DOE, 2012).

Comprehensiveness of Policies

Overall, when reviewing the breakdown of principles by partial- and full-level completeness, the principles that were more likely to be only partially present in sample policies were those that necessitated specific terms or language (e.g., continuous monitoring, function of behavior, restricted procedures to reduce harm or for safety of student). For instance, if a policy did not explicitly indicate that continuous monitoring should occur for both restraint and seclusion, it would receive a score of “1” for being partially present. The authors deemed specific terms or language to be necessary for these categories as specific language is required to align with expert recommendations. In contrast, whereas medical and mechanical restraint guidelines were covered less often than most principles, when they were included in district policies, they were fully covered 100% of the time. This reflects the more clear-cut nature of this issue, as well as a stronger consensus that these procedures are not intended to be used by school personnel as emergency behavior interventions (except for resource officers’ use of mechanical restraints).

Policies by District Enrollment Size

Although there was variability in the number of principles addressed based on district enrollment, there was no clear distinction of differences in policy content across these three enrollment size categories based on the data. There was consistency among the principles most likely to be addressed across all size categories. Debriefing (No. 11B) was present in the smallest amount of policies, whereas Situations for Warranted Use (No. 3A) was evident in each district policy. In addition, five principles were addressed by a minimum of 85% of the districts within each size category of the sample: Principle No. 6, Restriction of Use for Purposes of Discipline and Coercion; No. 7, Restriction of Use so as to Not Harm; No. 8, Supervision, Oversight, and Review; No. 10, Staff Training; and No. 11A, Use Monitored for Safety. However, within each size category there was variability in the number of districts addressing each principle. Large districts in total addressed 67% of all principles across all policies. This was the largest percent of principles addressed across the three size categories (small = 60%, medium = 58%). Interestingly, the large district category had the largest variance in policy source and unique policies. Whereas both the small and medium groupings had a higher degree of consistency across policy content, the group policies that were adopted by the majority of these districts addressed the fewest principles. It is possible that small- and medium-sized districts may employ restraint and seclusion procedures less often, resulting in a reduced focus on policy content guiding these procedures. One study found a positive correlation between school size and the frequency of documented restraints in day treatment and residential school settings (Fogt et al., 2008). There may be a similar correlation between district size and the frequency of the use of restraint and seclusion. On the contrary, medium and small districts may simply prefer somewhat shorter and less explicit policies.

Groups as Sources of Policies

It is not clear what level of knowledge or expertise the individuals who developed group core policies (presumably attorneys) had regarding the use of physical restraint and seclusion in schools. Nor is it evident how they developed and distributed these policies to their clients. If an attorney does not have a good grasp on recommended guidelines for restraint and seclusion policies, it may lead many districts to have inadequate or incomplete policies. For example, many of the policies indicated that restraint and seclusion should only be used when a student’s behavior presented imminent risk of serious harm to self or others. However, later in the policy, contradicting guidance was provided, permitting these procedures in the event of property destruction, assistance in calming a child, or written in a student’s IEP or BIP without the stipulation of imminent danger. While group policies may have increased districts’ likelihood to adopt a restraint and seclusion policy in general, the limitations to this model are also evident. The adoption of such policies may have led some districts to implement procedures contrary to best practice when there were contradictions and omissions regarding important practices to keep students and staff safe from undue harm. As a result, improvement in policy development for local districts should be directed to the law firms or attorneys and organizations that provide advice and guidance to schools on these topics. Being able to teach or provide guidance to law firms supplying district policies might be the most efficient way to make these policies more complete. Interestingly, when districts created their own policies independently (not a part of any of the group-provided policies), they generally were less likely to be as complete in identifying and including the U.S. DOE principles.

Limitations

This was an exploratory policy evaluation of a maximum variation sample (Patton, 1987) of 85 school districts in one state without state statutory guidance. Results may not be representative of all 244 districts in the state nor of the district policies in the other 22 states that do not have state policies. Future studies could also examine whether districts in states that have statewide policies differ from those identified in the current study. Although the authors attempted to be as objective as possible, the definitions and operationalization of codes as well as the scoring of these policy principles are inherently subjective. In addition, this study did not analyze data regarding the frequency of the use of procedures, injuries as a result of these procedures, or whether districts provided training to staff on these policies or procedures. All of these concerns suggest follow-up research to this first exploratory analysis of district policies.

Conclusion

One important finding is that almost all of the public school districts in this state had policies in place and most of those have a substantial number of the important principles identified in their policies. This might argue against the need for state legislation on this topic. Eighty-four percent (n = 71) of the policies examined addressed over half of the policy principles recommended based on the U.S. DOE guidance document. Districts can follow federal guidelines even without state policies in place. On the contrary, a substantial number of districts had fewer than half of the DOE principles included. This is generally similar to the analysis of Marx and Baker (2017) regarding state legislation on this topic.

Content of the restraint and seclusion policies still varied considerably. Interestingly, variation in district policies did not appear to be affected by district size although it might be expected that larger districts would be more likely to have more complete policies. The nature of district policies appeared to be affected more by the source of the policy guidance that the district received from attorneys and other organizations than district size. Nevertheless, districts that have policies may vary substantially in the completeness and content of those policies.

While not surprising that the wording of the policies varied, it is concerning that some of the substantive content related to crucial principles recommended by the U.S. DOE guidance document and other professional organizations (e.g., Council for Children With Behavior Disorders, 2009) were not included in many policies. The most disconcerting discovery is the infrequent inclusion of the preeminent safety standard that these procedures are not to be used except in situations where the child’s behavior poses imminent danger of serious physical harm to self or others. Only a small number of districts (n = 8; 9%) included this principle in their policies, and many districts allow these procedures to be used in circumstances that were not indicative of imminent danger (e.g., property damage, calming a student, or classroom disruption). Marx and Baker (2017) used a more general statement of DOE No. 3, which did not split the two components and, as a result, we do not know whether state statutes included this statement in their state legislation.

One key contribution of having prescriptive state-level policies for district policies is that such guidance could make for more consistency across districts and could potentially ensure that the most important principles are included. A significant finding here was that most districts sought policy guidance from law firms and organizations. If this is consistent across other states that lack legislation (as we suspect may be the case), an alternative strategy to make district policies more consistent and complete may be to provide input or training to the attorneys and organizations that are the sources for many district policies. As school district policies are typically updated on an annual or semiannual basis, this strategy could result in more complete policies across many districts in a relatively short time frame—probably shorter than the legislative process. This is a very important finding that could result in a shift of focus from attempting to create national and state policy to instead directly improving local district policies.

Gagnon et al. (2017) indicated that the use of restraint and seclusion is “likely driven to a meaningful degree by local policy and school culture more generally” (p.74). The current analysis would suggest that local policy is unlikely to be “driving” the use of these procedures although it is possible that school culture shapes these policies and the use of these procedures. It is not clear whether having local policies, whether comprehensive or minimal, has any impact on better practices. School culture is shaped by the individuals who are in that school, and it is unclear whether key individuals in the schools, who might be in a position to implement these procedures, actually are informed of their local policies sufficiently to be able to implement them. This study did not address whether key staff in the district were trained regarding policies that districts had in place. Also, as outside vendors are most often contracted to provide district training on crisis intervention procedures, it is unclear whether those training sessions addressed local policies.

Suggestions for Future Research

The present study suggests that local policies with substantial content can be in place without state legislation. A comparison of district policies between states with more detailed state legislation and policy would be beneficial in understanding whether having explicit state policies affects the existence or comprehensiveness of local district policies. Similarly, future research should address whether having comprehensive policies affects the use of crisis intervention procedures, including restraint and seclusion, and if so, in what ways.

Studies might also determine whether key educators in districts that have policies on these topics understand them sufficiently or have had training to permit appropriate implementation and adequate supervision of staff. Such a study could determine whether the training on crisis intervention provided to educators by training vendors teaches compliance with state or local district policies on these topics. Outsourced vendor training is provided to a large number of educators; a recent estimate is that more than 750,000 educators, mostly individuals in special education, receive this training each year in the United States (M. Mason, personal communication, November 1, 2016). Nevertheless, very little is known about the content of the training being provided, nor whether such training addresses federal guidance principles or state or local district policies where they exist.

Future studies could also compare the completeness of policies with the frequency of use of restraint and seclusion within school districts. Although sufficient accurate data on the use of these procedures is not currently available (at present OCR data appears unreliable; Abamu & Manning, 2019; U.S. GAO, 2020), such a study would help to indicate whether policy significantly influences the use of these procedures in schools.

Overall, the results of this study represent a “glass half full” in that most districts have policies on physical restraint and seclusion even in a state that has no statewide legislation. Those policies also have most of the policy principles suggested in federal guidelines. At the same time, the “glass is half empty” in that some of the most significant safety principles are typically missing in these district policies, leading to questions of whether district restraint and seclusion policies actually make the use of these procedures safer when employed.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded in part by Nebraska Department of Education (USDE Grant H027A160079). The article’s contents do not necessarily represent the policy of Nebraska Department of Education or the U.S. Department of Education and no endorsement should be assumed.

ORCID iD

Eryn Y. Van Acker

Notes

References

Abamu

Manning

(2019, June 5). Desperation and broken trust when schools restrain students or lock them in rooms [Radio program episode] (Morning ed.). NPR. https://www.npr.org/2019/06/05/726519409/desperation-and-broken-trust-when-schools-restrain-students-or-lock-them-in-room?utm_medium=social&utm_source=facebook.com&utm_term=nprnews&utm_campaign=npr

Butler

(2019, July). How safe is the schoolhouse? An analysis of state seclusion and restraint laws and policies. Autism National Committee.

Council for Children With Behavior Disorders. (2009, May). CCBD position on the use of physical restraint procedures in school settings.

Duncan

(2009, July 31). Letter to colleagues from the U.S. secretary of education. U.S. Department of Education. https://www2.ed.gov/policy/elsec/guid/secletter/090731.html

Dunlap

Ostryn

Fox

(2011). Preventing the use of restraint and seclusion with young children: The role of effective, positive practices. Technical Assistance Center on Social Emotional Intervention for Young Children. https://eric.ed.gov/?id=ED526387

Eldeib

(2016, May 27). Teen’s death at residential center emblematic of “system problem”: DCFS. Chicago Tribune. http://www.chicagotribune.com/news/watchdog/ct-allendale-shaquan-allen-death-met-20160526-story.html

Every Student Succeeds Act, 20 U.S.C. § 6301. (2015). https://www.congress.gov/114/plaws/publ95/PLAW-114publ95.pdf

Fogt

J. B.

George

M. P.

Kern

White

G. P.

George

N. L.

(2008). Physical restraint of students with behavior disorders in day treatment and residential settings. Behavioral Disorders, 34(1), 4–13. https://doi.org/10.1177%2F019874290803400101

Freeman

Sugai

(2013). Recent changes in state policies and legislation regarding restraint or seclusion. Exceptional Children, 79(4), 427–438. https://doi.org/10.1177%2F001440291307900403

10.

Gage

N. A.

Pico

D. L.

Evanovich

(2020). National trends and school-level predictors of restraint and seclusion for students with disabilities. Exceptionality. Advance online publication. https://doi.org/10.1080/09362835.2020.1727327

11.

Gagnon

D. J.

Mattingly

M. J.

Connelly

V. J.

(2017). The restraint and seclusion of students with a disability: Examining trends in U.S. school districts and their policy implications. Journal of Disability Policy Studies, 28(2), 66–76. https://doi.org/10.1177%2F1044207317710697

12.

Holden

Nunno

(2019, October 29). Learning from tragedy: A twenty five-year (1993-2017) survey of the fatal restraints of children and young people in the United States [Conference session]. PowerPoint presentation to F.I.C.E. Conference, Tel Aviv, Israel.

13.

H.R. Rep. No. 114-354. (2015). https://www.congress.gov/114/crpt/hrpt354/CRPT-114hrpt354.pdf

14.

Kentucky Protection and Advocacy. (2016). The near-death of Brennan Long: Autopsy of an abusive restraint. http://www.kypa.net/current-events.html

15.

Marx

Baker

(2017). Analysis of restraint and seclusion legislation and policy across states: Adherence to recommended principles. Journal of Disability Policy Studies, 28(1), 23–31. https://doi.org/10.1177/1044207317702069

16.

Miller

George

M. P.

Fogt

J. B.

(2005). Establishing and sustaining research-based practices at Centennial School: A descriptive case study of systematic change. Psychology in the Schools, 42(5), 553–567. https://doi.org/10.1002/pits.20091

17.

Morrison

(2015, June). Restraint and seclusion in schools: Recommendations for California (Publication No. CM61.01). Disability Rights California. https://www.disabilityrightsca.org/system/files/file-attachments/CM6101.pdf

18.

Patton

M. Q.

(1987). How to use qualitative methods in evaluation. SAGE.

19.

Peterson

Ryan

Rozalski

(Eds.). (2013). Physical restraint and seclusion in schools. The Council for Exceptional Children.

20.

Pillsbury & Disability Rights Legal Center. (2018, October 2). Pillsbury and disability rights legal center salute settlement in disability rights cases [Press release]. Pillsbury. https://www.pillsburylaw.com/en/news-and-insights/pillsbury-and-disability-rights-legal-center-salute-settlement.html

21.

Rozalski

Peterson

R. L.

Ryan

J. B.

Losinski

(2013). Proactive techniques to prevent and reduce conflict. In Peterson

R. L.

Ryan

J. B.

Rozalski

(Eds.), Physical restraint and seclusion in schools (pp. 51–67). The Council for Exceptional Children.

22.

Ryan

J. B.

Robbins

Peterson

Rozalski

(2009). Review of state policies concerning the use of physical restraint procedures in schools. Education and Treatment of Children, 32(3), 487–504. https://doi.org/10.1353/etc.0.0062

23.

Samuels

(2017, June 15). Parents awarded $900K after hidden cameras capture student restraint. Education Week. https://www.blogs.edweek.org/speced/2017/06/parents_awarded_900k_in_restra.html

24.

Scheuermann

Peterson

Ryan

J. B.

(2015). Professional practice and ethical issues related to physical restraint and seclusion in schools. Journal of Disability Policy Studies, 27(2), 86–95. https://doi.org/10.1177/1044207315604366

25.

U.S. Department of Education. (2012). Restraint and seclusion: Resource document. https://www.ed.gov/policy/seclusion/restraints-and-seclusion-resources.pdf

26.

U.S. Government Accountability Office. (2009, May 19). Seclusions and restraints: Selected cases of death and abuse at public and private schools and treatment centers (Publication No. GAO-09-719T). https://www.gao.gov/products/GAO-09-719T

27.

U.S. Government Accountability Office. (2020, April). K-12 education: Education needs to address significant quality issues with its restraint and seclusion data (Publication No. GAO-20-345). https://www.gao.gov/products/GAO-20-345

28.

U.S. Office for Civil Rights. (2016, December). Dear colleague letter on restraint and seclusion of students with disabilities. U.S. Department of Education. https://www2.ed.gov/about/offices/list/ocr/letters/colleague-201612-504-restraint-seclusion-ps.pdf

29.

Westling

D. L.

Trader

B. R.

Smith

C. A.

Marshall

D. S.

(2010). Use of restraints, seclusion, and aversive procedures on students with disabilities. Research and Practice for Persons with Severe Disabilities, 35(3–4), 116–127. https://doi.org/10.2511%2Frpsd.35.3-4.116