Type 2 diabetes in rural Uganda: Prevalence, risk factors, perceptions and implications for the health system.

Abstract

Background: Between 2010 and 2030, a 69% increase in type-2 diabetes is expected in low-income countries compared to a 20% increase in high income countries. Yet health system responsiveness to non-communicable diseases has been slow in sub-Saharan Africa. Data on the prevalence of type 2 diabetes and its associated factors in mainly rural settings is lacking, yet such data can guide planning for diabetes control. Objective: The aim of these studies was to assess the prevalence of type 2 diabetes, its risk factors, risk perceptions, and possible screening tools among people aged 35-60 years so as to inform primary care level intervention in rural low-income settings. Methods: Four studies (I-IV) were conducted among people aged 35-60 years in a mainly rural demographic surveillance site in eastern Uganda (2011-2013). Study I, a cross-sectional survey assessed the prevalence of diabetes-related risk factors (including overweight, hypertension and seven socio-behavioural risk factors) in 1,656 people. Study II then estimated the prevalence of abnormal glucose regulation (AGR) using fasting plasma glucose (FPG), and assessed its socio-behavioural correlates in 1,497 people. To compare the utility of FPG and glycated haemoglobin (HbA1C) rapid tests in risk stratification, a comparative survey of 795 people was nested into study II (Study III). To assess perceptions about diabetes, 12 Focus Group Discussions were conducted among people afflicted or at higher risk of type 2 diabetes (Study IV). The assessments used standard measurements and cut-offs. Results: About 18% of people aged 35-60 years were overweight, while 21% had hypertension. Women (OR 3.7; 95% CI 2.7-5.1), peri-urban dwellers (OR 2.5; 95% CI 1.5-3.0), and wealthier people (OR 4.1; 95% CI 2.4-7.0) were more likely to be overweight. Only 34% had adequate knowledge about lifestyle diseases (I). Prevalence of diabetes was 7.4% while pre-diabetes was 8.6% and 20.2% with WHO and American Diabetes Association criteria respectively (II). Prevalence of AGR was twice higher in obese people compared to those with a normal BMI (APRR 1.9, 95% CI 1.3-2.8) (II). Sufficient physical activity and diverse diet were associated with lower likelihood of AGR (II). The direct medical cost of screening one person was US$0.53, translating to US$2 per person detected with AGR. Agreement between FPG (the reference) and HbA1C in classifying diabetes was moderate (Kappa=22.9; AUC =75%), while that for AGR was low (Kappa=11.0; AUC=59%) (III). However, agreement was high (over 90%) among negative tests and in participants with risk factors for type 2 diabetes. FPG was more practical than HbA1C (III). Participants’ strong perceptions of diabetes as a very severe disease were incongruent with their perceived urgency for lifestyle change (IV).While people with diabetes perceive obesity as ‘sickness’, those without diabetes say it signifies ‘success’ (IV). Poverty, access to food and large families were cited as barriers to healthy diets. Domestic work was the preferred platform for physical activity increments. Conclusions: Obesity, insufficient physical activity and unhealthy diets are possible aids to identifying people at higher risk of type 2 diabetes in primary care. Mass screening for abnormal glucose regulation may not be affordable to struggling health systems. However, rapid tests have utility in further evaluation of people with risk factors. Health education about lifestyle diseases is a priority and should target to change the community’s notion of health and healthy lifestyles. Several challenges of adding NCD services to overstretched health systems come to light.