Dropout may be a biased indicator of provider performance...

Halley Brown and her colleagues recently received notice that their paper, Problems with the evaluation of mental health providers: Differences in type of patient "drop-out," will be published in Psychiatric Services. This project examined clinical perceptions of drop-out as related to patient satisfaction. Please see below for the abstract. Congratulations Halley!

Objective: Retention or ‘dropout’ is a common metric for evaluating psychotherapy treatment quality, but can be problematic when used to assess provider performance. Provider differences were evaluated in three types of patient dropout: general dropout, ‘bad’ dropout - failure to return and report of negative treatment experiences and outcome, and ‘good’ dropout – failure to return but report of positive treatment experiences and outcome. Methods: Patient utilization records were paired with satisfaction surveys (n=3,054). Binomial mixed effects models were used to examine provider differences in each dropout type. Results: Thirty four percent of treatment episodes resulted in dropout (n=1,032, 34%). Of these, 149 (14%) were classified as bad dropout and 277 (27%) met criteria for good dropout. Providers accounted for approximately 17% of the variance in general dropout, 10% in bad dropout, and 10% in good dropout. Provider rank fluctuated based on type of dropout. Conclusions: Retention-based provider assessment should isolate dropout type, as non-specific metrics may lead to biased estimates of performance.