Dropout Rates From Psychotherapy Trials for Borderline Personality Disorder: A Meta-Analysis

Borderline personality disorder (BPD) is a serious mental illness associated with heightened disability, risk for suicide, and costs to society. This study aims to meta-analytically quantify dropout rates from psychotherapies of BPD, identify moderators, and assess reasons for dropout and time taken to dropout. PubMed, PsycINFO, and MEDLINE were screened from database inception to March 2020 for trials that investigated psychotherapies for individuals with BPD reporting dropout rates. The primary outcomes were pooled dropout rates and differential treatment retention across all studies, all randomized controlled trials (RCT), all outpatient studies, and all outpatient RCTs. Random effects meta-analysis, metaregression analyses, and publication bias tests were conducted. Information on reasons for dropout and time to dropout was synthesized qualitatively. Dropout rates were 22.3% considering all studies, and 28.2% when only considering outpatient randomized controlled trials. Odds of dropout were not significantly higher in the control condition than in the intervention condition. Longer duration, randomization, phone coaching, and outpatient setting were associated with higher dropout rates, but only when considering all studies. Publication bias-adjusted dropout rates were as high as 29.9%. Reasons for dropout included dissatisfaction with treatment, expulsion from treatment, and lack of motivation. Most dropouts occurred in the first half of treatment. Dropout is an important and prevalent issue in BPD psychotherapies. Reported rates are minimized by publication bias, and moderators of dropout rates are inconsistent. Subsequent research should identify obstacles to completing treatment and investigate ways to organize treatment allocation to enhance treatment retention.