Concordance of assessments of clients' mental and behavioral health with in vivo assessment of work performance.

The study included only 42 clients from two programs in one state, so it may not represent other regions or larger groups. About 50% of participants had schizophrenia-spectrum disorders, so findings may not apply to other diagnoses. Many participants were in simulated work rather than competitive employment, and the small sample could not fully compare these groups. Work performance ratings were relatively high, which may mean the scale missed differences at the top end. Providers were trained but most were not licensed clinical professionals, and their mental health ratings did not agree with clients' self-report. The study measured associations at one time point, so it cannot show that mental health status causes or predicts work performance. The paper states 42 clients were recruited but does not report how many participants were analysed or how dropouts were handled.

The authors reported no conflict of interest.

Do not read the failure of client self-report to track provider work ratings as proof that mental health does not affect work. The study only compared one short self-report measure with provider-observed work ratings in 42 clients, and it cannot show why the scores did not agree.