A Systematic Review of Direct Assessments to Evaluate Psychotropic Medication Effects for Children With Disabilities.

Hoyle et al. (2022) hunted for every direct behavior test used to see if non-stimulant psych meds help kids with disabilities.

They screened hundreds of papers and kept only the ones that watched actual behavior before and after dose changes.

The team then graded each tool for quality: clear definitions, repeated measures, and blind raters.

Almost no studies passed the quality check. Most relied on parent checklists, not timed behavior counts.

The few direct tools seen—like progressive-ratio button pressing—were buried in small pilot work.

Bottom line: we rarely collect solid behavior data before tweaking meds.

de Korte et al. (2021) is one of those rare pilots. Their progressive-ratio test showed kids worked harder for toys after a med change, proving the tool can work. Hoyle et al. (2022) folds that exact study into its map, so the papers fit like puzzle pieces.

Hudson et al. (2012) looked only at autism and asked “do the drugs work?” Hoyle et al. (2022) widens the lens to all disabilities and asks “how do we measure if they work?”—an update, not a clash.

Cox et al. (2022) found behavior plans, not med changes, drove big drops in severe problem behavior. That outcome supports N et al.’s call: without sharp behavior data, we may credit the pill when the teaching did the job.

You can’t taper or defend a med dose without hard numbers. Ask for repeated, timed behavior samples—rate of SIB per hour, not “he seems calmer.” Push for tools like progressive-ratio or brief functional analysis before the next script refill. One clean line graph beats a stack of rating scales in the team meeting.