Attrition Rates in Treatment Trials: A Systematic Review and Meta-analysis of Clinical High-Risk for Psychosis Interventions

Poster B97, Tuesday, October 9, 11:30 am - 1:00 pm, Essex Ballroom

Megan Farris1, Dan Devoe1, Jean Addington1; 1Hotchkiss Brain Institute, Department of Psychiatry, University of Calgary, Calgary, Alberta, Canada

Attrition is a complex issue affecting study validity. Our objective was to synthesize the relative attrition rates in trials conducted in clinical high-risk (CHR) for psychosis populations. We searched the following electronic databases: MEDLINE, Embase, PsychINFO, CINAHL and EBM with no date, geographic or language restrictions. Inclusion criteria was any treatment-based randomized controlled trial (RCT) conducted in CHR samples that reported attrition. Relative attrition rates were calculated using random-effects meta-analysis, stratified by time, and reported as odds ratios (ORs), proportions, and 95% confidence intervals (CIs). Twenty-one RCTs met our inclusion criteria. The mean age was 19.4 years and 56% were male. Attrition rates between cognitive behavioural therapy and control groups were not statistically different at 6-months (OR=1.13, 95% CI=0.80-1.60, n=729, trials=6), 12-months (OR=1.08, 95% CI=0.81-1.44, n=842, trials=7) or 24-months (OR=0.75, 95% CI=0.52-1.07, n=540, trials=3). All other treatment interventions: integrated psychological therapy (24-months; OR=0.59; 95% CI=0.28-1.24, n=162, trials=2), omega-3 (6-months: OR=1.05, 95% CI=0.69-1.60, n=512, trials=3; 12-months: OR=0.97, 95% CI=0.64-1.46, n=512, trials=3), cognitive remediation therapy (2-months: OR= 1.13, 95% CI= 0.37-3.41, n=157, trials=3), and glutamate modulators (4-months: OR=1.52, 95% CI=0.40-5.82, n=52, trials=2) were not statistically different from controls at any time-point. When accessing overall study attrition, the pooled attrition rate was 25% (95% CI=16%-34%) with significant heterogeneity between types of treatment (p<0.001). Attrition rates did not statistically differ between treatment and control groups, however an overall attrition of 25% in CHR trials is high relative to other populations. Future research is warranted to prevent potential biases associated with high attrition rates.

Topic Area: Ultra High Risk / Prodromal Research

Back to Poster Schedule