Search | Search by Center | Search by Source | Keywords in Title
Voils CI, King HA, Neelon B, Hoyle RH, Maciejewski ML, Reeve BB, Yancy WS. Within-person variability in the extent of and reasons for antihypertensive medication nonadherence. Poster session presented at: Society of Behavioral Medicine Annual Meeting and Scientific Sessions; 2013 Mar 27; San Francisco, CA.
Objective: Medication nonadherence is typically analyzed in a cross-sectional fashion, which assumes that it is stable over time. However, nonadherence may fluctuate circumstantially and over short time periods, which would have implications for dosing and tailoring of interventions aimed at reducing nonadherence. The goal of this study was to characterize within-person variability in medication nonadherence. Method: 269 veterans (mean age 64 years, 93% male) with a diagnosis of hypertension and prescribed at least one antihypertensive medication completed four telephone assessments, each approximately 2 weeks apart (range 11-17 days), for a total of 1076 occasions. Extent of nonadherence was measured with a valid, reliable three-item scale, and 22 possible reasons for missing medications were measured. The response distributions were generally bimodal, so we dichotomized the extent scale as adherent vs. nonadherent and each reason as endorsed vs. not endorsed. Logistic multilevel models were then fitted to provide estimates of within- and between-person variability across the four occasions. Results: Nonadherence was reported on 18% of occasions. The proportion of total variance attributable to within-person variation was 49%, indicating substantial within-subject variability in the extent of nonadherence across the two months in which four occasions were assessed. No single reason was endorsed on more than 10% of occasions. Reasons with the lowest within-person variability were "not having any symptoms," "taking medication between meals," and "possible interactions with other medications." Reasons with the highest within-person variability were "forgetting," "cost," and "being busy." Conclusion: There is both within- and between-patient variability in the extent of nonadherence and reasons for nonadherence. To improve medication adherence, interventions may need to be tailored to time-varying reasons for nonadherence and address forgetting doses, medication costs, and busy schedules.