A longitudinal study of the predictors of quality of life in patients with major depressive disorder utilizing a linear mixed effect model

Abstract

We set out in this study to examine a longitudinal dataset using a linear mixed effects model. Our ultimate aim is to identify predictors of the quality of life (QOL) domains and items amongst patients suffering from major depressive disorders. Four categories of variables are included in our analysis, composed of 'personal predisposition', 'psychosocial', 'illness-related' and 'time', while the outcome variables for this study are the 'physical', 'psychological', 'social' and 'environmental' domains of QOL, in conjunction with all of the items within the scale. A total of 104 subjects from an outpatient clinic of a university-affiliated hospital participated in this longitudinal study, with a one-time follow up being carried out on 70 of these subjects (67.3%) who agreed to participate in the follow-up study. The 'severity of depression', 'sense of competence' and 'sense of mastery', 'use of anti-depressant medication' and 'environmental resources' are found to be significant predictors of the detailed aspects of QOL. Of these, 'symptom severity', 'sense of competence' and 'sense of mastery' were found to occur most often. Finally, the results of the present study demonstrate that 'illness-related' and 'psychosocial' categories are capable of predicting the various QOL domains for patients suffering from depression.