Prospective State and Trait Predictors of Daily Medication Adherence Behavior in HIV.

Only 53 of 87 recruited participants had usable behavioral adherence data, because 34 did not return MEMS bottles. The study was preliminary and tested each predictor separately, so the findings need replication. MEMS bottles track openings, not actual swallowing of pills, and monitoring can cause reactivity, though the authors controlled for study week. Daily surveys were not timed to medication dosing, so same-day effects may differ from next-day effects. The sample included more White men who have sex with men than the U.S. HIV population, although it also included women and minority participants. State and trait measures used different wording, which could contribute to differences, although a sensitivity analysis with identical items still showed state-trait differences. Most participants had undetectable viral load, so the findings may not apply to patients with less well-controlled HIV or poorer adherence.

The metadata lists NIH extramural research support. The supplied text does not include an author conflict-of-interest statement.

Do not conclude that nurses should boost motivation or change mood to improve adherence. This was a preliminary observational study of naturally occurring daily states, not a test of an intervention, and the mood measure was everyday mood rather than clinical depression; more severe depression might still disrupt adherence.