Multivariate prediction of relatives' stress outcome one year after first hospitalization of schizophrenic and depressed patients.

OBJECTIVE: The present study presents the 1-year follow-up findings of the Munich 5-year follow-up study on relatives of first hospitalized patients with schizophrenia or depression. The aim of the study was to determine which factors moderate the impact of the patient's illness on the stress experienced by his key relative in different dimensions.

METHOD: Of the relatives who had participated in the baseline assessment, 90% could be reassessed (n = 69) with respect to their objective and subjective burden, well-being, self-rated symptoms and subjective quality of life as well as different personal dispositions and resources.

RESULTS: Data demonstrate a considerable level of burden in most of the relatives at 1-year follow-up. Compared to baseline, a significant reduction in objective and subjective burden as well as significant improvements in well-being, self-rated symptoms and subjective quality of life occurred in two-thirds of the relatives, while well-being and self-rated symptoms remained worse when compared to norm values. Although stress reduction was significantly associated with patients' residual symptoms on the bivariate level, no main effects on relatives' stress outcome at 1-year follow-up could be observed under multivariate conditions. In multivariate linear regression models for each stress indicator, different combinations of predictors resulted, explaining up to 75% of the total variance of the stress indicators. The most relevant predictors were relatives'expressed emotion and neuroticism, their generalized negative stress response and life stressors, having significant direct and indirect effects on relatives' stress outcome. Stress reduction was mainly caused by an interaction of relatives' generalized positive stress response and patients' residual symptoms.

CONCLUSION: Findings support the transactional character of the stress process in caring for a patient with a severe mental disease. A multidimensional approach is necessary to identify the most important predictors of burden in order to improve family intervention strategies which aim to reduce burden.