Adherence to interferon β-1a therapy using an electronic self-injector in multiple sclerosis: a multicentre, single-arm, observational, phase IV study
BACKGROUND Prior research has shown a decrease in medication adherence as dosing frequency increases; however, meta-analyses have not been able to demonstrate a significant inverse relationship between dosing frequency and adherence when comparing twice-daily versus once-daily dosing. OBJECTIVE To determine the effect of scheduled dosing frequency on medication adherence in patients with chronic diseases. METHODS A systematic literature search of Medline and Embase from January 1986 to December 2011 and a hand search of references were performed to identify eligible studies. Randomized and observational studies were included if they utilized a prospective design, assessed adult patients with chronic diseases, evaluated scheduled oral medications taken 1 to 4 times daily, and measured medication adherence for at least 1 month using an electronic monitoring device. Manual searches of reference sections of identified studies and systematic reviews were also performed to find other potentially relevant articles. Standard definitions for medication taking, regimen, and timing adherence were used and evaluated. Studies were pooled using a multivariate linear mixed-model method to conduct meta-regression accounting for both random and fixed effects, weighted by the inverse of the variance of medication adherence. RESULTS Fifty-one studies, comprising 65, 76, and 47 dosing frequency arms for the taking, regimen, and timing adherence endpoints were included. Unadjusted adherence estimates were highest when the least stringent definition, taking adherence, was used (range for dosing frequencies: 80.1%-93.0%) and lowest when the most stringent definition, timing adherence, was used (range for dosing frequencies: 18.8%-76.9%). In multivariate meta-regression analyses, the adjusted weighted mean percentage adherence rates for all regimens dosed more frequently than once per day were significantly lower compared with once-daily regimens (for 2-times, 3-times, and 4-times daily regimens, respectively: differences for taking adherence: -6.7%, -13.5%, and -19.2%; regimen adherence: -13.1%, -24.9%, and -23.1%; and timing adherence: -26.7%, -39.0%, and -54.2%). CONCLUSION Patients with chronic diseases appear to be more adherent with once-daily compared with more frequently scheduled medication regimens. The use of more stringent definitions of adherence magnified these findings.