A scoping review on two-stage randomized preference trial in mental health

Abstract

Abstract Background: Randomized Controlled Trial is the most rigorous study design to test the efficacy and effectiveness of an intervention. Patient preference may negatively affect patient performance and decrease the generalizability of a trial to clinical population. Patient preference trial have particular implications in mental health since mental health interventions are generally complex, blinding of intervention is often difficult or impossible, patients may have strong preference, and outcome measures are often subjective patient self-report which may be greatly influenced if patient’s preference did not match with the intervention received. Methods: In this review, we have surveyed the application of two-stage randomized preference trial with focus on studies in the field of mental health. The study selection followed guideline provided by Preferred Reporting Items for Systematic Reviews and Meta-Analysis. Results: Six two-stage randomized preference trials (ten publications) have been identified in the mental health field. In these trials, the pooled dropout rates were 18.3% for the preference arm, and 28.7% for the random arm, with a pooled RR of 0.70 (95% CI, 0.56-0.88; P=0.010) indicating significantly lower risk of dropout in the preference arm. The standardized preference effects varied widely from 0.07 to 0.57, and could be as large as the treatment effect in some of the trials. Conclusion: This scoping review has shown that two-stage randomized preference trials are not as popular as expected in mental health research. The results indicated that two-stage randomized preference trials in mental health would be beneficial in retaining patients to expand the generalizability of the trial.