Prevalence and Complications Associated with Diabetes Mellitus at the Nairobi Hospital, Nairobi City County, Kenya

Abstract

Background and Purpose: Diabetes mellitus is a lifestyle disease and there is an increase in mortality and morbidity related to diabetes in Kenya. This study purposed to identify sub-types of diabetes in the EHR dataset through clustering.

Methods: A retrospective study design was employed in the analysis of 652 records of cases of diabetes mellitus collected and stored in the EHR during routine care between January 2012 and December 2016 at the Nairobi Hospital located in Nairobi City. The dataset was subjected to pre-processing; cleaning; handling cases of missing values, identification and handling of outliers, and resolving cases of inconsistencies. Thereafter, data were subjected to the analysis of measures of central tendencies and measure of dispersion. Data were then normalized using the z-score standardization method before being subjected to a density-based clustering algorithm; DBSCAN for cluster analysis.

Results: The prevalence of T2DM is at 92% and there is an increasing trend of the presence of T1DM and hypertension among cases of T2DM. On the other hand, the most common complications of diabetes include; retinopathy (12%), neuropathy (11%), and cardiovascular (11%). As a result, there is need for public awareness of the dangers of diabetes mellitus.

Conclusions: There is need for sensitization for the members of the public on the usefulness of physical exercise and dietary requirements for healthy living to slow the development and progression of diabetes mellitus.