Family burden and expressed emotion in the parents of adults with schizophrenia, a critique and empirical test of an integrated model

grantor: University of TorontoThis study critiqued the dominant theoretical models of families and schizophrenia and empirically tested relationships between subjective family burden and expressed emotion (EE) of parents of adults with schizophrenia. Integration of the models is theoretically plausible, consistent with empirical evidence, congruent with dominant social work practice models, and has the potential to advance clinical practice. An integrated EE/family burden model was proposed and formed the basis of research questions. Exploratory secondary data analyses were conducted. The research tested whether the EE components, EOI and criticism, mediate relationships between parent subjective burden and each of patient social functioning, patient symptoms, parent knowledge about schizophrenia, and parent experience with schizophrenia. The results were indefinite. The findings that neither EOI nor criticism act as statistical mediators were inconclusive because of low statistical power for the tests. There were trends for associations between the EE components, subjective burden, and patient social functioning. Relationships between EE change and subjective burden change were tested. The results indicate that EE may explain, at most, a small part of subjective burden change. There was a trend for a relationship between EE change and subjective burden change. Among parents who were initially high EE, lower subjective burden was associated with subsequent change to low EE. EE was more likely to be stable than to change. Change from high to low EE was more common than change from low to high EE. The major limitations of the research design were the: (1) non probability sample; (2) small sample size and the associated low statistical power for many of the tests; and (3) the large number of univariate tests which increased the chance of Type I error. Thus, the findings must be interpreted with caution. Design features which compensate for weaknesses in the published research include: (1) use of a theoretical framework; (2) avoiding confounding in measurement; (3) preservation of sample size through the use of multiple imputation for ignorable missing data; (4) analysis of potential sample bias; and (5) use of multivariate statistical techniques.Ph.D