Meta-analysis of trials of interventions to improve medication adherence

摘要: The effect of tools and methods designed to enhance medication adherence that have been evaluated in randomized controlled trials was studied. A literature search performed with MEDLINE, International Pharmaceutical Abstracts, PsychLIT, ERIC, EMBASE for the period from 1966 December 2000. Only randomized, at least 10 subjects per intervention group were included. Of 484 articles evaluated, only 61 met criteria meta-analysis. Multiple interventions or study samples identified 23 articles. Each counted as a separate study, yielding 95 cohorts totaling 18,922 subjects. these subjects, 9,604 (51%) received 9,318 served controls. Cohorts reported between 1990 1999 accounted 53% sample; 56% all based physician offices 26% involved hypertensive patients. Behavioral 41 (8,885 subjects), educational 22 (6,392 combined 32 (3,645 subjects). Homogeneity groupings sizes (ESs) calculated each type intervention. Overall, data not homogeneous, so conclusions could be derived entire body data. nonhomogeneous (p < 0.001 p 0.01, respectively); however, behavioral cohort homogeneous (Q = 42.48, d.f. 40, 0.36). overall ES 0.07 (95% confidence interval [CI] 0.04-0.09). There no significant differences among interventions. Educational had an 0.11 CI 0.06-0.15); there 0.08 0.04-0.12). When stratifying by intervention, mail reminders largest impact (ES 0.38). Meta-analysis studies improve revealed increase 4-11%. No single strategy appeared best.