Factors associated with type 2 diabetes mellitus comorbidity among adult patients: a retrospective clinical audit of a medical centre in Abuja, Nigeria

Abstract

Background Diabetes mellitus (DM) is a disorder of metabolism in which an absolute or relative deficiency of insulin results in hyperglycemia leading to acute and chronic complications. Over the years, the issue of incidences of DM in different regions of the globe has become a huge public health problem, especially in cases of patients with comorbidity conditions. These conditions have made the treatment and management of DM extremely difficult, especially in sub-Saharan Africa, where there are weak health systems. Methods This study was a retrospective study based on the record review of 162 patients receiving care at Federal Medical Centre, Jabi Abuja. Using the electronic medical records, key variables of interest were extracted, and the effects of comorbid conditions on Type 2 Diabetes Mellitus (T2DM) were determined. Results The summary results revealed that the incidence of T2DM was approximately 107(66.1%) had T2DM, while about 159 (98.2%) out of the 162 patients with one or more other diseases had T2DM. There were more female patients, 97(59.9%) with cases of DM than male patients. Patients aged 40-69 years have more prevalence cases of DM (73.5%), with those engaging in businesses or trading common in this category (29.63%). Most of these patients had Hypertension as a comorbid condition 110(67.9). The results based on the logistic model showed that patients with concordant comorbid conditions are 2.11 times more likely to develop T2DM. The Bayesian logistic regression model revealed a positive association between gender and comorbidity; the male patients were 1.65 times more likely to develop comorbidity, and patients with a smoking history were 1.56 times more likely to develop comorbidity. T2DM was not associated with systolic blood pressure. Conclusions This present study provides elaborate details on incidences of DM and associated comorbidity conditions in the study area and its environs. A Bayesian model seems to be more accurate than conventional logistic regression based on the present findings. This, in turn, could be used to formulate health policies geared towards effective management, control and prevention of DM of any kind and community sensitization on DM in Nigeria.