No one knows how many Colorado students are shut inside rooms, alone, at school.

But Riley Ellis-Beach knows that it happened to her son. He was a small kindergartener, barely 30 pounds, with big emotions and a diagnosis of autism. When asked to do a task he didn’t want to do, he’d spit, growl or cuss at school staff, according to a school behavior log. Sometimes he’d bite or hit. The staff began taking him to the “opportunity room,” where he’d stay for as little as five minutes or as long as an hour and a half, the log shows.

“Not until I had to go in and pick him up did we realize it was an empty room with nothing in it,” Ellis-Beach said. One time, she said she showed up to school to find her son in the room with the door shut. His special education teacher was peeking at him through a window, she said.

“He looked like a wild animal that had been trapped in a cage,” Ellis-Beach said.

State lawmakers have tried in recent years to better monitor and oversee seclusion, which means shutting a student inside a room, and restraint, which means physically restraining a student. But a loophole that legislators and advocates say was unintentional has meant that there is almost no public data about seclusion, while restraint data is more readily available.

Only 10 of Colorado’s 179 school districts included a separate tally of how many of their students were secluded last school year in the annual reviews that districts are required to file with the Colorado Department of Education, according to a Chalkbeat analysis of the reports. That’s despite a 2022 state law that was supposed to make data about seclusion available and “easily accessible.”

The lack of information makes it impossible to understand the scope of a controversial practice that some educators say is necessary to control dangerous student behavior but that many families say leaves their children with lasting trauma.

Emily Harvey, co-legal director of Disability Law Colorado, called the use of seclusion “an unknown problem.”

“We have no idea how often it’s happening and who it’s happening to,” she said.

The data blackout also makes it hard for state lawmakers to know exactly what to do about seclusion. Two separate bills introduced last month in the state legislature aim to address the information gap. One of the bills goes even further and seeks to ban seclusion altogether.

Change to solitary confinement law

In 2022, Colorado lawmakers passed a bill requiring the Colorado Department of Education to collect from school districts an annual review of the districts’ use of restraint and seclusion.

However, when the state education department went to put the bill into practice, it ran into what department spokesman Jeremy Meyer described in an email as “the complexity in our statutes regarding restraint and seclusion.”

The complexity has to do with the definitions of restraint and seclusion in Colorado law. The 2022 bill required school districts to produce annual reviews of their use of restraint alone. That’s not because lawmakers didn’t want data on seclusion; they thought it would be included in the reports because Colorado law for years defined seclusion as a form of restraint.

What lawmakers didn’t realize at the time was that in 2016, previous state lawmakers who were worried about restraint and seclusion in a different setting — youth jails — had created separate definitions of the two practices in Colorado law. Because the state definition of restraint only refers to physical restraints, the disclosure rules developed by the Colorado Department of Education say nothing about seclusion. Under the rules, school districts don’t have to report any data on it.

“It has created such a problem,” said Harvey of Disability Law Colorado. “A lot of people are asking, ‘How widespread is this problem?’ And I’m like, ‘Who knows?’ ”

The Colorado Department of Education began collecting the annual reviews in the summer of 2024.

Chalkbeat obtained all 179 reports through an open records request. While all the reports include data about the number of physical restraints performed on students, only 10 districts included separate information about seclusion: Cherry Creek, Aurora, Adams 12, Academy 20, Harrison, Cheyenne Mountain, Weld RE-1, Weld RE-4, Weld RE-5J, and Pikes Peak BOCES.

That means just 5% of Colorado districts reported seclusion data in their annual reviews.

Two of the 10 districts, Cheyenne Mountain and Weld RE-5J, reported zero seclusions in the 2023-24 school year. The data in the Weld RE-1 and Harrison 2 reports was redacted. The state allows the data to be suppressed if there are fewer than 16 incidents.

The other six districts reported a total of 907 instances of seclusion last school year. The 2022 bill also required the state education department to “develop easily accessible, user-friendly profile reports for each school district” that would be posted on its website.

The reports were meant to include a host of data, including the number of students restrained or secluded.

Meyer, the department spokesman, said the department collected restraint and seclusion data for the reports “but ultimately did not post it because we have concerns about its accuracy this year.” He pointed to “significant misunderstandings from districts in this first year of implementation.”