School Restraint and Seclusion Report: Definition Disputes and Data Accuracy Challenges

To obtain more accurate data on school restraint and seclusion practices, the U.S. Department of Education is proposing to revise the definitions for reporting in the Civil Rights Data Collection (CRDC). Previously, education organizations, disability and civil rights advocacy groups, and the Government Accountability Office (GAO) raised concerns about misreporting and problematic data in past CRDC data collections, with some very large school districts reporting extremely low rates of restraint and seclusion.

However, the revised definitions proposed by the Department of Education's Office for Civil Rights (OCR) may further increase the difficulty of reporting for school and district administrators and could lead to continued misreporting in the future, according to the leader of the Council of Administrators of Special Education (CASE).

Phyllis Wolfram, Executive Director of CASE, pointed out that a major obstacle to accurate data reporting is the differences between state and federal definitions of physical restraint and seclusion. She said, "Based on experience over the past many years and the current proposal, we still believe data reporting will be inaccurate and will not achieve the desired effect—unless they spend more time effectively working with states to ensure districts understand what to report, how to report, and establish corresponding accountability mechanisms."

Ron Hager, Managing Attorney for Education and Employment at the National Disability Rights Network (NDRN), believes that while the revised definitions may not eliminate all ambiguity, they are a good step forward in addressing the challenges of data collection and reporting on restraint and seclusion—practices that are disproportionately applied to students with disabilities. Hager said, "OCR's continued attempts to clarify these definitions and make information easier to obtain is a very good trend."

OCR is also adding and revising other data elements—from preschool discipline rates to adding a non-binary option to the sex category—applicable to the 2021-22 school year data, which will be reported starting the next school year. The public comment period is open until February 11, with 631 comments submitted as of Monday.

This data collection, which began in 1968, is typically biennial. But OCR announced last August that it would conduct the collection for two consecutive years for the first time (2020-21 and 2021-22 school years) to understand the impact of the pandemic on students and school systems.

Focus on inappropriate practices

CASE's position is that schools should use restraint and seclusion as little as possible and only in emergency situations. The organization says certain restraint practices, such as placing a student in a prone position, should never be used. CASE also supports sound documentation and reporting of such incidents.

But some terms in the proposed revised definitions are unclear and could lead to different interpretations and potential underreporting, said Kevin Rubenstein, Assistant Director of Student Services for Elmhurst Community Unit School District 205 near Chicago. Rubenstein, who also serves as CASE's Policy and Legislation Chair, said, "We want to focus on inappropriate practices. We know these practices exist. We don't want inappropriate practices in schools, so if these definitions leave more loopholes, we cannot allow that to happen."

Specifically, the revised definition of physical restraint reads: "Physical escort includes a temporary touching or holding of the hand, wrist, arm, shoulder, or back of a student for the purpose of guiding the student to a safe location, with contact ceasing once the safe location is reached." The next sentence reads: "A physical escort involving methods used to maintain control of a student should be considered a physical restraint." In Rubenstein's view, the last sentence contradicts the previous one.

The revised seclusion definition begins with "the involuntary confinement of a student alone in a room or area, with or without adult supervision, from which the student is prevented from leaving." But what concerns special education administrators is the next sentence: "A student who believes they have been told by school personnel that they cannot leave the room or area should be considered in seclusion."

Although the definition also states that seclusion does not include situations where a student feels free to leave a location, the definition can be interpreted in multiple ways, said CASE President Erin Maguire at the Council for Exceptional Children Convention and Expo held January 18 in Orlando, Florida. Maguire asked, for example: Could a student told to wait in the principal's office for a conversation be said to be in seclusion because they feel they cannot freely leave? Maguire said, "Everyone here works in a school environment. Considering the power structure between adults and students in schools, the concept that a student can be kept in any space, I think, may have been overlooked in the discussions when developing this definition." Maguire also serves as Director of Student Support Services for the Essex Westford School District in Essex Junction, Vermont.

Identifying outliers

Hager said the seclusion definition may still have room for interpretation, but the revised definitions at least help better collect data on incidents that clearly should be recorded as restraint and seclusion. He said, "The most important thing is to get the best possible data and correct it when there are obvious errors."

A 2020 GAO report found that data quality control processes for restraint and seclusion were largely ineffective or nonexistent. For example, GAO researchers found that in the 2015-16 CRDC, 70% of school districts reported zero restraint and seclusion incidents. However, CRDC business rules only flagged potential data errors when a district had at least 100,000 enrolled students. At that time, only 30 of the more than 17,000 districts nationwide had at least 100,000 students.

Los Angeles Unified School District—the second-largest school system in the nation—reported in the 2015-16 collection that 10% of its schools had restraint incidents and reported no seclusion incidents.

Hager also emphasized that data collection is not a judgment on whether practices are appropriate. He said data can help educators and advocates identify trends and outliers, and understand where more support is needed in practice or reporting.

Wolfram said that for special education administrators, restraint and seclusion data help determine interventions for students or training needs for staff. OCR also relies on CRDC data to investigate complaints alleging discrimination or violations of federal civil rights laws and takes proactive measures, such as providing policy guidance and technical assistance to schools, parents, students, and others.

Additionally, OCR collects data on the use of mechanical restraint—defined as devices or equipment that restrict student activity. A proposed revised definition adds that mechanical restraint "includes the use of handcuffs or similar devices by sworn law enforcement officers or other school security personnel to prevent a student from moving their arms." Some disability rights advocacy organizations, such as the Disability Rights Education & Defense Fund, have called on OCR to also collect information on schools' use of medication to control behavior or restrict activity.

Disproportionate use

Although restraint and seclusion are used on only a small number of students—101,990 of more than 50.9 million students in the 2017-18 school year—their use is disproportionately high for students with disabilities. In 2017-18, 80% of students restrained and 77% of students secluded were eligible for services under the Individuals with Disabilities Education Act, even though they made up only 13% of the total student population.

Frederick County Public Schools in Maryland recently reached a settlement agreement with the U.S. Department of Justice after an investigation found that over two and a half school years from 2017 to 2020, the school system implemented 7,253 restraints or seclusions on 125 students, averaging 58 per student. Every student secluded was a student with a disability, and 99% of restrained students (all but one) were students with disabilities.

There is currently no federal regulation limiting the use of restraint and seclusion in schools. A bill in Congress would ban seclusion, mechanical and chemical restraints, and physical restraints that restrict a student's breathing. However, policy observers doubt the legislation will make progress this year due to other pressing matters in Congress and the midterm elections.

According to an undated tally by the National Association of State Boards of Education, all states except Oklahoma and South Carolina have policies on restraint and seclusion. These policies limit the use of restraint and seclusion and provide certain protections. According to the Education Commission of the States (ECS), state laws also include provisions for parent notification and staff training.

The CRDC currently collects data on mechanical restraint, including the use of devices that restrict student activity.

Improving data quality control

The CRDC collection on restraint and seclusion began in 2009 and is disaggregated by race and ethnicity, sex, disability status, and English learner status.

In a letter to Education Secretary Miguel Cardona last year, GAO said improving data quality on restraint and seclusion incidents should be a key priority area for the Department of Education. Specifically, GAO recommended that the Department set business rules targeting schools and districts with extremely low and extremely high numbers of reported incidents, and set data-driven thresholds to detect such incidents. OCR has been working to address the inappropriate use of restraint and seclusion and to improve data quality procedures, providing technical assistance to districts and schools.

Wolfram said OCR's efforts are commendable, but school systems may need more support for high-quality recordkeeping and professional development to address staffing shortages and high turnover exacerbated by the pandemic. Wolfram said the instructional disruptions caused by the pandemic have also increased the need for support for students with social, emotional, and behavioral challenges and have reignited concerns that restraint and seclusion may be used inappropriately. She said administrators need clear guidance on reporting expectations at both the state and federal levels. Wolfram said, "We need this data so we can identify where intervention and improvement are needed."