Compilation of Quantitative Overviews of Studies of Adherence

Abstract

Purpose of Study. The purposes of this article are to: (1) Compile and critique quantitative reviews of studies of adherence with prescribed medications, and (2) Summarize current knowledge of adherence from the perspective of this literature.Methods. MEDLINE and other bibliographic databases were searched for quantitative reviews in which adherence with prescribed drugs was an endpoint. The reviews were classified according to type and whether they addressed either the factors associated with adherence or interventions designed to improve adherence.Findings. Three types of quantitative reviews are defined: consensus reviews, meta-analyses, and systematic reviews. There is a general agreement across review types and disease categories about the factors associated with adherence. To a first approximation, socio-demographic and disease-related factors are not associated with adherence, whereas factors related to the regimen and patient socio-behavioral factors are associated, though probably weakly. Meta-analyses and systematic reviews of interventions to improve adherence have taken disparate approaches to the pooling of primary study data. Some interventions, for example, reminder packaging, mailed reminders, and rewards, improve adherence, whereas patient information by itself typically does not. There is conflicting evidence for patient counseling, perhaps reflecting the variety of approaches or the relatively weak associations between socio-behavioral factors and adherence.Conclusions. Consensus reviews have generally been consistent in determining that features of the drug regimen and socio-behavioral factors are associated with adherence. The disparate methodological approaches of quantitative overviews of interventions to improve adherence reflect opposing ideas about combining primary data. There is, however, evidence favoring the effectiveness of some types of intervention, for example, reminders, and a consensus that patient information by itself typically is ineffective in improving adherence.