The Relationship Between Air Pollution And Diabetes

1096 Words5 Pages
2) Methods and Materials. List of the countries of the WHO African Region was acquired from the regional WHO website (Appendix 1). Various databases were searched to obtain the most recent country-specific data on all the variables. Links to data on all indicators is available in Appendix 2. 2.1 Diabetes Prevalence The data on Diabetes Prevalence (% of the population aged 20 to 79) from 2014 was obtained from the World Bank World Development Indicators (WDI) Database. The prevalence represents the percentage of people with T1 and T2 diabetes. Since 90% of individuals affected by diabetes are suffering from T2 diabetes [12], this indicator was used as the measure of T2 diabetes in Africa for this study. 2.2 Indicators of Air…show more content…
The density of physicians (per 1000 population) and General Government Expenditure on Health (% of total expenditure), retrieved from the same database as variables above, were used as an estimate of accessibility and quality of healthcare in the state. Per Capita GDP, PPP (2011 constant international dollar, $) and Urban Population (%) data was obtained from the World Bank WDI Database and these indicators were used as a proxy for the welfare of the state and urbanisation respectively. 2.5 Statistical Analysis Appropriate conversions of data were made to present data in SI units. Linear regression analysis was then used to evaluate the association between independent (indicators of air pollution, food supply quantities, socio-demographic and lifestyle factors) and dependent (diabetes prevalence) variables. This was done to describe the variables against diabetes prevalence. Stepwise multiple linear regression for indicators of air pollution and diabetes prevalence was then performed to identify whether any of the dietary, socio-demographic or lifestyle variables modify, improve or predict a model. Independent variables added to the model were: all food supply quantities, socio-demographic and lifestyle factors. 13 countries were excluded from the final models due to lack of
Open Document