| Summary: | 碩士 === 國立臺北大學 === 統計學系 === 103 === Psychiatric disorders receive more and more attention nowadays. They are chronic and can result in disability. In turn, it will not only increase the burden of society, but also have the negative impact of patient and family life. Also, high frequent hospitalization can induce more medical expenditure. The reason behind this can be de-institutionalized, bad recovery, or disease recurrent. As a result, it is important to investigate the influential factors associated with the frequency of hospitalization. Based on appropriate program, in turn, it might help them to improve the symptoms.
The main purpose of this thesis is to evaluate the factors associated with frequency of psychiatric hospitalization after onset. We used data collected for “The Examination on the Unity Model of Recovery-The Development of the Stage of Recovery Instrument and the Application of Strengths Perspective” project extracted from Survey Research Data Archive (SRDA), Sinica. There were 420 eligible cases of schizophrenia and affective disorders. The log liner model was used to assess the association between the frequency of hospitalization and potential risk factors. Owing to possible overdispersion, two distributions, Poisson and Negative Binomial, are considered. To adjust for the different duration, the offset adjustment is used. Demographic variables, interaction terms, and recovery stage factors are included in the model sequentially to conduct three sets of model, and from which selects the best fitted model.
The model assuming a negative binomial distribution provides a better fit. The results show that the higher the age of onset, the higher of average frequency of hospitalization (β=0.016，P<0.05), and for the members back to the community, who have job averagely show lower frequency of hospitalization than the members who have no job (β=-0.862，P<0.05). The results of this study can provide reference for the relevant personnel as early medical intervention or follow-up care plan.
Key words: frequency of Psychiatric hospitalization, Negative Binomial Regression, offset, overdispersion, Poisson Regression.
|
|---|
