Current and projected prevalence of arterial hypertension in sub-Saharan Africa by sex, age and habitat: an estimate from population studies

As of the year 2000, raised blood pressure (BP) was reported as the biggest single contributor to death rates in both the developed and the developing world 1. In 2005, on the basis of data accrued between 1980 and 2002, Kearney et al.2 estimated that the absolute numbers affected by hypertension in eight regions of the world (as defined by the World Bank) would increase between 2000 and 2025 such that an extra 560 million people would be affected in the intervening 25 years. The same authors reported that in sub-Saharan Africa (SSA), which consists of 47 countries and comprises about 12% of the world's population, the numbers affected by hypertension would be expected to almost double in both men and women during the same 25-year period. Most of the predicted effect is driven by anticipated population increases. However, Kearney et al. also estimated an increase in prevalence of hypertension (which more closely reflects changes in the age distribution of the population) in seven of the eight World Bank regions. The exception was SSA, in which limited changes in the population distribution are expected and hence no change in prevalence was predicted. Nevertheless, the predicted 9 and 13% increase in global prevalence of hypertension among men and women, respectively, translating into 560 million more adults affected by 2025 are daunting statistics concerning the risk factor, which was already the biggest single killer in 2000. The object of the study by Twagirumukiza et al.3 was to produce a more accurate updated estimate of the current prevalence and numbers affected by hypertension (as of 2008) in SSA and to estimate projected rates and numbers affected as of 2025. From 28 possible studies only 15 studies from 11 countries satisfied eligibility criteria, after undergoing quality appraisal. Hence, less than one-quarter of eligible countries in SSA were represented. The study samples were not necessarily representative of their population let alone nationally, but they were conducted between 1998 and 2008, had a sample size of at least 300 and several were conducted according to the WHO STEPwise approach (‘or similar’). Only three studies had sufficient stratified data on the three criteria – age, sex and urban/rural status – changes in which formed the basis on which 2025 estimates of hypertension were made. However, a logistic regression model imputation was used to generate a full set of age/sex/habitat-specific prevalences of hypertension in each country, extrapolated to the whole of SSA in 2008. Using United States Census bureau data, expected population estimates for 2025 and changes in urbanization rates were then used to predict prevalence rates and numbers affected in 2025 (assuming age/sex/habitat-specific hypertension prevalences remain unchanged). The prevalence of hypertension in 2008 in SSA is estimated at 16.2% (slightly higher in men than women and higher in urban than rural populations). In 2025, the equivalent figure rises to 17.4% (95% confidence interval 15.4–22.6%). The number of hypertensive patients in 2008 and 2025 is estimated to be 74.7 and 125.5 million, respectively – a 68% increase. The estimated prevalence rates of hypertension are low compared with the USA and most European countries 4, but this reflects the low mean ages of the SSA population included (average 40 years) and after age standardization the rates come into line with western populations (23.3%). Furthermore in these analyses, data from those aged 15 and above were included – younger than in many other surveys (including Kearney's analyses 2). This age difference, along with including more studies and more contemporary studies (up to 2008 versus 2002) and considering urban/rural status, may explain the differences between the current results and those of Kearney 2. Compared with Kearney's results (which related to 2000) this latest study (relating to 2008) reported marginally lower age-adjusted prevalence rates (27.6 vs. 23.3%) and slightly fewer people affected (79.8 vs. 75 million), but significantly less people were estimated to be affected in 2025 (150.7 vs. 125.5 million). However, estimates of ‘current’ numbers affected in both studies are more than three times higher than recently reported by the WHO Regional Office for Africa 5. Assuming the figure of 75 million affected in 2008 is reasonably accurate, the burden of disease associated with these numbers is enormous because the vast majority of these cases are currently untreated and are likely to remain so 6. In the context of SSA, untreated hypertension translates mainly into ischaemic and haemorrhagic stroke – the latter usually being more fatal and/or more devastating that ischaemic stroke. A very recent systematic analysis of survey data from 199 countries including 5.4 million people shows that global BP levels and prevalence of hypertension have fallen on average between 1980 and 2008, among men and women 7. However, region-specific inspection of these data shows that in the last decade among men and women from central, east, south and west Africa mean SBP has risen and the highest mean BPs recorded worldwide among men and women were in some east and west African counties. These data add to the concerns raised by Twagirumukiza et al.3. The prediction of prevalence and numbers affected by hypertension published in the accompanying paper 3 and those of Kearney 2 are likely to be conservative in that they do not allow for any age/sex stratified increases in the prevalence of hypertension due to increased exposure to the main determinants of raised BP. In reality, with ‘development’ increased exposure to several adverse risk factors for hypertension usually occurs. These changes include, increasing age (as infectious diseases are brought under control), increased body weight, increased intake of salt, fats and alcohol, reduced intake of fruit and vegetables and reduced exercise. This is clearly a perfect recipe for generating increased BP levels and rates of cardiovascular disease. Hence, if the impact of these hitherto, inevitable consequences of development were factored into the predictions for 2025, the estimates of hypertension prevalence and numbers affected would be significantly worse. In summary, these latest data from SSA (despite the inevitable difficulties associated with assessing hypertension in 47 countries) flag up the very clear message that urgent efforts and resources are required to evaluate the current size of the hypertension problem in SSA and to instigate preventive strategies. Where necessary, effective treatments also need to be made available and accessible. All the major evidence-based drug classes can now be acquired very cheaply and it is vital that they are made available cheaply to the markets of SSA, via an infrastructure that can deliver them effectively. The alternative is to stand back and watch the dreadful toll of human suffering caused by hypertension, knowing it had been largely preventable.