Using analog baselines to assess the effects of naltrexone on self-injurious behavior.

The team gave naltrexone to two adults who hurt themselves. First they ran a short functional analysis to see why the self-injury happened.

They then set up small test sessions that copied each function. Drug and placebo were given in an ABAB pattern.

Naltrexone only helped a little. One person slapped his head less during work tasks, but other self-injury stayed the same.

The second person showed almost no change. The drug worked only for one shape of behavior and only when demands were present.

Green et al. (1987) also used an ABAB design and saw no benefit at all. The newer study looks better because it tested the drug only after it knew the behavior’s function.

Tantam et al. (1993) and Garcia et al. (1999) show that some people stay safe for years after they stop naltrexone. These cases seem to clash with the weak acute effects, but they tracked long-term outcomes, not short drug tests.

Schmidt et al. (2021) used the same idea—brief functional tests to pick treatment—but chose behavioral plans instead of pills. Their strong results remind us that function-based behavior plans often beat medication.

Run a functional analysis before you think about naltrexone. If the assessment shows demand escape, you might trial the drug for that one situation, but keep strong behavioral treatment in place. The pill is at best a helper, not a fix.