Logistic Regression Analysis to Determine Risk Factors for the Clinical Diagnosis of Diabetes; Case Study: Komfo Anokye Teaching Hospital (2008-2009)

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
The study was conducted to find out if factors like age, gender, occupation status, marital status, smoking, alcohol consumption, hypertensive, contribute to the clinical diagnosis of diabetic. The data used for the study was a secondary data from the physiotherapy department of the Komfo Anokye Teaching Hospital in Kumasi in the Ashanti region of Ghana. Two years data was sampled out, that was, 2008 and 2009. The software package used for the analysis of the data was SAS, version 9.1. The preliminary results which involved an independent test of our outcome variable, diabetes, with all the predictor variables (alcohol, smoking, hypertension, cholesterol level, gender, family history, occupation, and marital status) showed that alcohol, hypertension, cholesterol level, and occupation were statistically significant at 5% to the outcome of diabetes. Marital status, smoking, family history and gender were not statistically significant to diabetes. A test of association between diabetes and alcohol with gender as a confounder showed that one can get diabetes when he/she consumes alcohol and this can be influenced by the gender of the person. The results from the statistical package showed that occupation, alcohol, cholesterol level and age were statistically significant to our outcome, diabetes. The odds ratio of occupation (0.497), alcohol (1.958), cholesterol level (2.259) and age (1.021) showed that the unemployed, those who take alcohol, and people with high cholesterol levels have a higher risk in getting diabetes. Age increase also increases one‘s risk to having diabetes.
A thesis submitted to the School of Graduate Studies, Kwame Nkrumah University of Science and Technology, Kumasi, in partial fulfilment of the requirements for the award of Master of Philosophy in Mathematics