A novel SCL-90-R six-item factor identifies subjects at risk of early adverse outcomes in public mental health settings

To increase access to treatment, Italy made assessment at community mental health centers (CMHCs) independent of medical referral, resulting in increased numbers of patients to be triaged efficiently. To support this process, we evaluated SCL-90-R item-ratings to identify factors that best predicted adverse early outcomes among persons seeking first-time CMHC care in a 24-month period in Rome. A psychiatric nurse screened subjects with a brief interview and self-administered SCL-90-R and psychiatrists provided CGI ratings and ICD-9 diagnosis. Of 832 screened subjects, 32 (3.85%) were hospitalized or attempted suicide within 90 days. Six SCL-90 items (15,41,55,57,78,88) scored much higher with than without such adverse outcomes; their sum is proposed as a predictive measure (“SCL-6″). In binary multivariable logistic modeling, this factor, but not age, sex, diagnosis, or other SCL-90-derived subscales strongly predicted adverse outcomes. A ROC curve for SCL-6 reflected a strong separation between subjects with versus without adverse outcomes (AUC = 0.76). This simple screening tool may support timely identification of patients at risk of early adverse clinical outcome who require especially close follow-up.