Correlates of restraint use in an institutional population.
Self-injury is the clearest warning that a restraint will last too long—so treat the behavior and check for pain fast.
01Research in Context
What this study did
Sturmey (1999) watched adults with developmental delay in one large institution. Staff logged every restraint and the reason. The team then asked: which behaviors predict short restraints? Which predict long ones?
They coded four resident traits: hurts others, hurts self, low body weight, and withdrawn behavior. Stats told which traits linked to restraint minutes.
What they found
Hurting others led to quick holds that ended fast. Hurting self, low weight, and staying withdrawn led to holds that lasted hours.
Self-injury was the strongest single predictor of long restraints. If a resident hit himself, he was likely to stay tied or pinned for a long time.
How this fits with other research
Storch et al. (2012) later showed that self-injury can spike when the person is in pain. Pain signs are easy to miss, so staff may keep restraining instead of treating the hurt. Together the papers hint: check for pain first, then decide on restraint.
Lambrechts et al. (2009) and Griffith et al. (2012) found that staff feel more anger and use more control when residents show challenging behavior. Those emotions can stretch restraint time, matching P’s data that resident traits alone do not decide duration—staff reactions matter too.
Oliver et al. (2002) review adds a lens: severe disability plus stress plus poor attachment raises behavior. Their stress-attachment model helps explain why the same self-injury that predicts long restraint also keeps the cycle going.
Why it matters
If you work in a group home or school, track which residents hurt themselves. That single flag tells you who is at risk for long, harmful restraints. Pair that data with a quick pain check and staff emotion regulation training. You can cut restraint minutes and keep everyone safer.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Start a 2-minute pain screen before every restraint for clients who hit themselves.
02At a glance
03Original abstract
Demographic and behavioral correlates of the use of restraint were analyzed in an institutional population of 300 persons with developmental disabilities. Examination of the frequency distributions of restraint frequency and duration indicated that there were 33 consumers who experiences relatively few, short-duration restraint and 11 consumers who had daily restraint for many hours. Separate analyses of these two kinds of restraint were performed. Short-duration restraint was predicted by behavioral variables indicating extra-personal maladaptive behaviors. Demographic variables did not predict short-term restraint. Multiple regression analysis indicated that only independent predictor of short-term restraint was the severity rating of Hurts Others. Similar results were found for predictors of the duration of short-term restraint. The only correlates of long-duration restraint was low weight, Hurts Self, and Withdrawn Behavior. Multiple regression analysis indicated that the only variable that independently predicted long-duration restraint was the severity rating of Hurts Self. The implications of these data for the management of restraint are discussed.
Research in developmental disabilities, 1999 · doi:10.1016/s0891-4222(99)00015-3