Family and patient predictors of symptomatic status in schizophrenia

To test an interactive hypothesis that, in schizophrenia, a combination of patients' and relatives' characteristics at 1-month postdischarge from hospital (Time 1 [T1]) better predicts the level of psychotic symptoms at follow-up (Time 2 [T2]), than do the characteristics of patients or relatives alone.

Male patients (n = 38) with a diagnosis of schizophrenia, without substance abuse, and in contact with their families, were recruited at the time of hospital discharge. Patients' psychotic symptom levels were monitored every 2 weeks until follow-up, while family measures were administered at T1 and T2. The 4 predictor variables in the regression analysis were T1 symptom levels of the patient and 3 measures of family interaction (expressed emotion, family burden, and family functioning).

The model based on the family variable, family burden at T1, and the patient variable, patients' remitted levels of psychotic symptoms at T1, was found to significantly predict the level of psychotic symptoms at T2. These 2 T1 variables made independent and additive contributions to the level of psychotic symptoms at T2, predicting 19% of the variance. Neither expressed emotion nor family functioning at T1 added to the prediction.

This finding suggests a patient-family interactional component to symptomatic relapse in schizophrenia.