OBJECTIVE: The aim of this study was to identify predictors of non-response in inpatient depression treatment by means of routinely collected data from 10 psychiatric clinics. METHODS: Evaluation was carried out through univariate analyses first; subsequently a model was developed by means of multivariate analyses. RESULTS: About one third (31.5 %) of the 511 patients included in the study was classified as non-responder. Concerning patient level, psychic comorbidity, severity of depression symptoms and serious impairment of lifestyle as indication for hospitalisation have an impact on treatment outcome. In contrast, socio-demographic characteristics are not suitable as predictors. On the level of process characteristics, psychotherapy and patient compliance to medication enhanced the chance of profiting from inpatient treatment. CONCLUSIONS: The results can be integrated into a well interpretable model of the current inpatient depression treatment. Non-response is not yet sufficiently predictable through routine data.
OBJECTIVE: The aim of this study was to identify predictors of non-response in inpatient depression treatment by means of routinely collected data from 10 psychiatric clinics. METHODS: Evaluation was carried out through univariate analyses first; subsequently a model was developed by means of multivariate analyses. RESULTS: About one third (31.5 %) of the 511 patients included in the study was classified as non-responder. Concerning patient level, psychic comorbidity, severity of depression symptoms and serious impairment of lifestyle as indication for hospitalisation have an impact on treatment outcome. In contrast, socio-demographic characteristics are not suitable as predictors. On the level of process characteristics, psychotherapy and patient compliance to medication enhanced the chance of profiting from inpatient treatment. CONCLUSIONS: The results can be integrated into a well interpretable model of the current inpatient depression treatment. Non-response is not yet sufficiently predictable through routine data.