Adherence to maintenance-phase antidepressant medication as a function of patient beliefs about medication

Abstract

Purpose: This study aimed to identify the demographic, psychiatric, and attitudinal predictors of treatment adherence during the maintenance phase of antidepressant treatment, ie, after symptoms and regimen are stabilized.

Methods: We surveyed 81 primary care patients given maintenance antidepressant medications regarding general adherence, recent missed doses, depression and treatment features, medication beliefs (necessity, concerns, harmfulness, and overprescription), and other variables. Additional data were collected from medical and payer records.

Results: Median treatment duration was 75 weeks. Adherence and beliefs were broadly dispersed and unrelated to treatment duration and type, physical functioning, and demographics. Multivariate analysis adjusting for social desirability, depression severity, and treatment duration indicated that an antidepressant-specific "necessity-minus-concerns" composite was strongly associated with both adherence outcomes. Specifically, adherence was highest when necessity exceeded concerns and lowest when concerns exceeded necessity. We crossed these 2 dimensions to characterize 4 patient attitudes toward antidepressants: skepticism, indifference, ambivalence, and acceptance.

Conclusions: Patients given maintenance antidepressants vary widely in adherence. This variation is primarily explained by the balance between their perceptions of need and harmfulness of antidepressant medication, in that adherence is lowest when perceived harm exceeds perceived need, and highest when perceived need exceeds perceived harm. We speculate on ways to tailor adherence strategies to patient beliefs. Subsequent research should determine whether patients' perceptions about medication predict depression outcomes, can be used to improve clinical management, and respond to behavioral intervention.

Figure 1.

Antidepressant belief groups (recent percentage of adherence in parentheses).

Figure 2.

Morisky adherence scores by beliefs group. Note: Higher scores indicate lesser adherence. ns = nonsignificant. * P = .057 † P = <.05 ‡ P = <.01 § P = <.001