The concordance of self-report with other measures of medication adherence: a summary of the literature

Abstract

Objective: The objective of this study was to evaluate the concordance of self-report measures of medication adherence (interview, diary, or questionnaire) with nonself-report measures of adherence (administrative claims, pill count or canister weight, plasma drug concentration, electronic monitors, or clinical opinion).

Methods: A literature search was conducted to identify published reports in which self-report and nonself-report measures of adherence were used within the same study. The concordance of measures within each study was categorized as high, moderate, or low based on a comparison of the adherence estimates.

Results: Eight-six comparisons of self-report to nonself-report measures of adherence were identified. Thirty-seven of the 86 comparisons (43%) were categorized as highly concordant. However, concordance varied substantially by type of self-report measure and nonself-report measure. Self-report measures, in general, were highly concordant with electronic measures in only 17% of comparisons, whereas they were highly concordant with other types of nonself-report measures in 58% of comparisons (chi-square = 14.30, P <0.01). When comparing self-report measures, interviews had significantly lower concordance with nonself-report measures as compared with questionnaires or diaries (chi-square = 8.47, P = 0.01). In 15 comparisons of interviews with electronic measures, none of the comparisons were highly concordant, whereas questionnaires and diaries had moderate-to-high concordance with electronic measures in 12 of 16 comparisons (75%).

Conclusions: The concordance of self-report and other measures of medication adherence varies widely based on the type of measures used. Questionnaires and diaries tend to have moderate-to-high concordance with other measures of medication adherence. However, interview-based self-reports are not concordant with electronic measures. Questionnaire and diary methods could be preferable to interviews for self-reported medication adherence.