Evidence for compliance with long-term medication: a systematic review of randomised controlled trials

Background Pharmacists play a pivotal role in optimising medication use which often includes actions to maximise compliance with long-term medication. The best evidence to support medication use is derived from randomised controlled trials (RCTs). It is often assumed that 100 % compliance is required to obtain the outcomes identified in the trial. This assumption needs to be examined. Objective To systematically review the reporting of compliance in RCTs of long-term medications. Method RCTs published in the New England Journal of Medicine, Journal of the American Medical Association, Lancet and BMJ in 2012, were reviewed to identify trials of medications for long-term use in adults. These trials were examined to evaluate the reporting of compliance. Main outcome measures The proportion of trials reporting compliance data, the methods used, and the proportion of trials using more than one method to determine compliance. Results Of the 289 RCTs published in 2012, 25 assessed long-term medications in adults. Compliance was reported in 12 (48 %) studies and only 2 (8 %) studies used more than one method to measure compliance. Pill count was the most commonly reported method for measuring compliance, with patient reports and blood levels also being used. Conclusion The reporting of compliance in RCTs is poor and the methodology inconsistent. The methods used overestimate compliance. If compliance in a clinical trial is low, the evidence for the effectiveness and most importantly safety of the medication(s) is questionable. Two or more methods, one of which is standardised, should be used to measure compliance in clinical trials. The requirement to report compliance should be included in publication guidelines.