A systematic review assessing the effectiveness of interventions to improve persistence with anti-resorptive therapy in women at high risk of clinical fracture
White HJ., Bettiol SS., Perera R., Roberts NW., Javaid MK., Farmer AJ.
Background: Despite availability of effective treatments for osteoporosis, impact on fracture rates may be suboptimal because of failure to adhere to recommended anti-resorptive therapy. Objective: To identify randomized controlled trials (RCTs) evaluating interventions intended to improve persistence with anti-resorptive therapy for treating women with osteoporosis or osteopenia. The design of the study is a systematic review and meta-analysis of RCTs. Methods: Included trials were those reporting interventions to improve persistence with or adherence to anti-resorptive treatment compared to a control medication or usual care. A search of MEDLINE, EMBASE, CINAHL and the Cochrane Library was supplemented by review of cited literature. Reports were reviewed and data pooled where appropriate. The primary outcome was duration of persistence with medication. Results: Six trials met inclusion criteria, including four reporting persistence as an outcome measure indicating a relative reduction in non-persistence of 22% (pooled relative risk: 0.78, 95% confidence interval 0.65-0.95) for active compared to control interventions. Heterogeneity between the trial effects was present but not significant (I2 = 47%, P = 0.11). Interventions were varied in design, and some measurements of adherence were subject to self-report bias. Two trials included the majority of participants (3386/3497), accounting for >90% of the weight in the pooled estimate. Conclusions: Trials to date suggest potential for improving persistence with medication taking thus improving treatment outcomes and reducing fracture risk. More precise measurement of medication taking and promoting fidelity to a precisely defined intervention protocol may lead to better assessment of impact on clinically important outcomes. © The Author 2010. Published by Oxford University Press. All rights reserved.