Predicting outcomes of mood, anxiety and somatoform disorders: the Leiden routine outcome monitoring study.

BACKGROUND Mood, anxiety and somatoform (MAS) disorders are highly prevalent disorders with substantial mutual comorbidity and a large disease burden. Early identification of patients at risk for poor outcome in routine clinical practice is of clinical importance. The purpose of this study was to predict outcomes in outpatients with MAS disorders using routine outcome monitoring (ROM) data. METHODS We conducted a cohort study of 892 adult MAS patients in a naturalistic outpatient psychiatric specialty care setting and validated our results in a replication cohort of 1392 patients. Poor outcome was defined as a <50% reduction (compared to baseline) on the self-report brief symptom inventory (BSI) or a score of ≥3 on the observer-rated clinical global impression severity scale (CGI-S). During a follow-up of up to 2 years, Cox regression models were used to analyze the independent baseline predictors for poor outcome. RESULTS In multivariable Cox regression models, independent and replicated predictors for poor outcome were higher age (overall p<0.001 for combined cohorts in multivariable Cox regression model), having comorbid MAS disorders or a somatoform disorder (<0.001), dysfunctional personality traits (i.e., tendency to self-harm [p<0.001], intimacy problems [p<0.001] and affective lability [p<0.001]), and a low reported general health status (p<0.001). LIMITATIONS Detailed treatment information was not available. CONCLUSIONS MAS patients meeting the profile of being elderly, suffering from comorbid MAS disorders or a somatoform disorder, with cluster B personality traits, and a poor reported general health may need special preventive measures to minimise the risk of poor outcome.