On the Performance of a Class of Generalized Linear Mixed Model on Some Psychiatric Patients' Data

Abstract

A generalized linear mixed model (GLMM) is an extension to the generalized linear mixed (GLM) in which the linear predictor contains random effects in addition to the usual fixed effects. They also inherit from GLMs the idea of extending linear mixed models to non-normal data. There are several applications of various types of generalized linear mixed models (GLMMs) to various fields, especially in the areas of health and biological sciences. In this our study Poisson logistic mixed regression model (a class of GLMM) was adopted to investigate the performance of the above mentioned method on some psychiatric patients’ data. A clinical trial of ninety (90) mentally disordered patients was examined in this work. Patients suffering from some level of psychiatric disorder were randomized to receive either Amitryphylline or Benzhexol in addition to other therapy. This work is motivated by Thall and Vail, which investigated the performance of the Poisson logistic mixed model on some epileptics’ data. The two types of therapy have little effect on the patients, but the interaction (between treatments and visits) has a substantial impact on the patients. The number of seizures is reduced by visits, and a combination of visits and medicines decreases the number of seizures. The fact that the treatments are insignificant suggests that mental disorders are mostly treatable with currently available medications. These drugs only ‘manage' them for a short period of time.