Patient-Reported Outcomes are Frequently Incomplete in Randomized Controlled Trials Focused on Alcohol Use Disorder: a Meta-Epidemiological Analysis

Introduction: Alcohol Use Disorder (AUD) can significantly affect patients’ lives through diseases processes, quality of life (QoL), mental health, social skills, and physical functioning. Patient-reported outcomes (PROs) are reports coming directly from the patient about these various health perspectives that are not interpreted by a clinician or anyone on the healthcare delivery team. These PROs need monitoring owing to the disease burden of AUD. As the importance of evaluating outcomes shifts towards QoL measures and PROs, it is necessary to appraise the quality in which these PROs are reported. Given there is limited research assessing how well PROs are reported among AUD trials, our primary objective is to evaluate the completeness of reporting of PROs in clinical trials involving AUD.

Methods: We used MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials (CENTRAL) to search for published RCTs focused on AUD. Following these searches, title and abstract screening, and full-text screening were performed by two investigators. Trials meeting inclusion criteria were evaluated for completeness of reporting using the Consolidated Standards of Reporting Trials - Patient-Reported Outcomes (CONSORT-PRO) extension adaptation. These trials were also evaluated for risk of bias (RoB) using the Cochrane RoB 2.0 tool. Screening and data collection were all performed in masked, duplicate fashion. Additionally, we extracted the PRO measures used in each RCT and identified the appropriate therapeutic area reported by the Mapi Research Trust’s ePROVIDE™ platform — a database for Clinical Outcome Assessments.

Results: Nineteen RCTs were evaluated in our analysis. Our primary outcome, the mean completion score for CONSORT-PRO, was 40.8%. Our secondary outcome — the identification of factors associated with completeness of reporting — found that trials published after 2014 (i.e after the publication of CONSORT-PRO extension) were 15.0% more complete than trials published before 2014. No other statistically significant results were found. Twenty-six unique PRO measures were found in our study encompassing the following therapeutic areas: two alcohol, five behavioral, five general, eight psychological, and six quality of life.

Conclusion: We found that the completeness of PRO reporting in RCTs involving AUD was deficient. Complete reporting of PROs is instrumental in understanding the effects of interventions, encourages patient participation in their treatment, and may increase clinician confidence in the value of PROs. Substantial heterogeneity exists among the types of PRO measures used to assess the same PRO domain. We recommend a more strict adherence to the CONSORT-PRO checklist and the creation of a Core Outcome Set, a minimum group of standardized outcome recommendations that should be measured and reported for a specific condition, for PRO measures in AUD trials. Complete PRO reporting in AUD trials will better reflect patient perspective as well as increase clinician confidence in the value of PROs. High quality treatment strategies for AUD require properly reported PROs.