*Georgina Murphy  did a PhD in Public Health and Primary Healthcare at the University of Cambridge. This work was done as part of a collaboration between the University of Cambridge, Wellcome Trust Sanger Institute, and Medical Research Council/Uganda Virus Research Institute, Uganda. Georgina has now moved to the University of Oxford, Nuffield Department of Medicine (firstname.lastname@example.org). Picture credit: zirconicusso and http://www.freedigitalphotos.net.
Obesity is rapidly becoming a growing problem across low- and middle-income countries, including those in sub-Saharan Africa. In many African countries, the prevalence of overweight is even estimated to be higher than the prevalence of underweight. With on-going changes in diet, urbanisation, and an increasingly sedentary lifestyle, the prevalence of obesity is predicted to continue to grow.
But what does it mean to be obese? You may have heard from your doctor or from public health campaigns that if your BMI is over 25 then you are overweight, if it's over 30 then you are obese, and if your waist circumference is more than 94 cm as a man or 80 cm as a women then you have central obesity.
These numbers were chosen because, above these thresholds, you are considered to be at substantially increased risk of diseases like diabetes and cardiovascular diseases. Globally, 44% of the burden of diabetes and 23% of ischaemic heart disease are attributable to overweight and obesity. However, this risk of developing disease at certain levels of body fat and size has been shown to differ between ethnic groups and population.
Although such ethnic variation exists, most of what we know about obesity is based on studies conducted in populations of European descent. By contrast, the relationship between obesity and disease in sub-Saharan African populations is poorly understood.
In collaboration with colleagues at the Medical Research Council/Uganda Virus Research Institute, Uganda, we conducted a study of approximately 6,000 rural Ugandan adults in 2011. In this population, we found that waist circumference and BMI may be useful tools for identifying people who had diabetes, hypertension or dyslipidaemia.
However, the currently recommended cut-offs or waist circumference and BMI may not be appropriate for African populations.
Instead, we suggest that a substantially lower cut-off for waist circumference (≥78 cm) should be considered for men. A slightly higher cut-off (≥82 cm) should be considered for women. We also recommend that, for screening purposes, a BMI cut-off lower than the current recommendation of 25 kg/m2 should be considered.
This was the first study of BMI cut-offs and the largest study of waist circumference cut-offs in sub-Saharan Africa. Much more work still needs to be done to decide on what the best cut-offs should be for African populations. This work is now, therefore, being expanded by the African Partnership for Chronic Disease Research (APCDR) into a large-scale collaborative project across sub-Saharan Africa. We have, so far, collated data on 56,000 individuals from 11 countries across the region.
Public health importance
A clearer understanding of the relationship between anthropometric measures (body size and shape) and risk of disease may be particularly important for sub-Saharan African populations for a number of reasons.
Firstly, the prevalence of obesity and cardiometabolic disease (such as diabetes and heart disease) is increasing across sub-Saharan Africa. Thus, the need for clinical guidelines and prevention programmes related to adiposity (body fat) is growing.
Secondly, resource-limited settings may benefit greatly from low-cost, easy-to-implement indicators of cardiometabolic risk, such as anthropometric measures.
Lastly, sub-Saharan Africa is struggling under a double burden of under- and over-nutrition. It is, therefore, important that clinical guidelines and public health messages are designed to communicate an optimal healthy, rather than simply reduced, body size.
Overall, a better understanding of obesity within sub-Saharan populations will be important for the design and implementation of public healthcare policy and population prevention programmes in an effort to address the growing burden of cardiometabolic diseases.