What Belongs in an ABA Agency's Restrictive Procedure Policy

Most agencies have trained their clinical staff in a crisis-prevention curriculum, and those staff can carry out the techniques competently. Ask the same agency for the policy that says when a restrictive procedure may be used, who can authorize one, and what happens afterward, and the answer sometimes takes a while to find.

Training answers how. Policy answers whether, who, and under what conditions.

That gap matters because a payer, a licensing surveyor, or regulator reviewing an event later doesn't evaluate the staff member's technique in isolation. They ask what the organization authorized, and where it wrote that down.

Define the category before writing the rules

The policy needs its own definitions section, and the category runs wider than physical restraint. Response blocking, protective equipment, and contingent restriction of access can all land inside it.

Accreditor language helps calibrate. The Joint Commission's revised behavioral health care requirements, effective January 1, 2025, define restraint as any chemical or physical method of restricting freedom of movement to manage behavior, and they treat physical holding as a type of restraint rather than a separate category. An agency needn't adopt that wording, but it needs wording of its own, since staff can't apply a standard to a category nobody defined.

Planned use and emergency use follow different paths

The policy has to separate two uses that carry different authorization requirements:

• Planned use. Written into the treatment plan, selected after assessment, consented to in advance, and reviewed on a schedule.
• Emergency use. Unplanned, limited to imminent danger, and never a treatment strategy.

Blurring the two invites drift in either direction: an emergency standard applied to a planned intervention, skipping assessment and consent, or repeated emergency use functioning as an unwritten plan nobody ever reviews.

What an authorization pathway has to accomplish

Rather than prescribe a structure, define the work the structure has to do. Any pathway, at any agency size, needs to deliver four things:

1. A clinical rationale on the record. Assessment support, plus a written explanation of why less restrictive alternatives may not suffice. The BACB Ethics Code directs behavior analysts to continually evaluate and document effectiveness, and to modify or discontinue an intervention that isn't working.
2. A set of eyes independent of the person who wrote the plan. The Ethics Code also directs compliance with any required review process, naming a human rights review committee as one example.
3. A named person accountable for the decision. Accountability that lands on a role, not on a room.
4. An expiration date. Time limits and fading criteria, so authorization ends unless someone renews it deliberately.

Whether a committee fits depends on the agency, and on what state rule or payer contract requires.

Three ways agencies size the same pathway

Small agencies with one or two behavior analysts. Independence likely has to come from outside: a contracted senior analyst or a peer-review arrangement with another agency, with the clinical director signing and the expiration date calendared at approval.

Mid-sized agencies with several analysts across one or two sites. Independence can come from inside. A clinical director who didn't write the plan reviews and signs, and the compliance lead confirms consent and current training before implementation.

Larger agencies delivering services across several sites. A standing review group meeting on a defined interval, with a named chair holding the decision and a documented path for cases that can't wait for the next meeting.

Whichever shape fits, write the pathway the agency can actually staff. A nine-member committee that never convenes reads worse to a reviewer than a two-signature process followed every single time.

Name what the agency won't do

A prohibition list carries as much weight as an authorization pathway. Candidates include prone and supine restraint where state rule or the agency's own standard bars them, mechanical restraint, seclusion, and any use for discipline, convenience, or a staffing shortfall.

State law offers model language. California's education code bars seclusion and behavioral restraint used for coercion, discipline, convenience, or retaliation, and Max Benson's Law prohibited prone restraint in California educational settings as of January 1, 2025. Rules like these attach to specific settings and provider types, so confirm what governs the settings where the agency actually delivers services.

The emergency standard

Emergency use needs a threshold stated plainly: imminent danger of serious physical harm that a less restrictive response can't prevent. Trained staff only, with release criteria tied to the danger resolving rather than to the client's compliance. The Association for Behavior Analysis International's position statement on restraint and seclusion, adopted as official ABAI policy, points the same direction: predetermined criteria, de-escalation designed to avoid physical intervention, the minimum restrictiveness needed to keep people safe, and precise, mandatory release criteria.

One more line earns its place: repeated emergency use with the same client should trigger plan review, not habit.

Training, oversight, and review

Name the curriculum, require demonstrated competency rather than attendance, set the recertification interval, and build a way to verify a staff member's training status on the date of a given event, since that's what an event review asks.

Then say who reviews these events, on what schedule, against what threshold, and what a pattern triggers: plan revision, retraining, or removal of a procedure. Compliance programs work as systems rather than binders, and the review cycle turns this policy into one. Our post on why quality assurance isn't compliance covers why clinical review alone doesn't close the loop.

State licensing rules, school-setting rules, and payer contract terms sit above all of it, and any can impose stricter requirements. Point to those sources rather than restating them, since restated language locks in wording that changes.

Frequently asked questions

What counts as a restrictive procedure in ABA? The category typically covers physical restraint, response blocking, protective equipment, seclusion where permitted, and contingent restriction of access. Definitions vary by accreditor and by state, so each agency should define the term in its own policy.

Does a small ABA agency need a formal review committee? No. It needs a pathway delivering a documented clinical rationale, independent review, a named accountable decision-maker, and a time limit. A contracted reviewer or a peer arrangement can supply the independence a committee would.

Does an agency need a restrictive procedure policy if it never uses restraint? Yes. A policy naming the prohibited procedures and describing what staff do instead during a crisis gives staff a standard and a reviewer an answer.

How often should a restrictive procedure policy come up for review? Annually works as a floor, with off-cycle review whenever state rule changes, a payer contract adds requirements, or an event review surfaces a gap.

Where this leaves you

Nobody writes a restrictive procedure policy on the day it gets used. It earns its value in advance, giving a technician a standard to apply, a clinical director a decision to make, and leadership a pattern to watch.

A companion post is coming on what the record has to establish after a restrictive procedure: documentation, family notification, and the debrief that feeds the review cycle above.

If you're building this policy set from scratch, or sizing an authorization pathway to the staff you actually have, the Compliance Collective offers policy tools and a place to ask the questions that come up. For an agency wanting its existing policy reviewed against state rule and payer terms, Project-Based Support handles that as a defined engagement.