Education and debate

Hypertension treatment and control in sub-Saharan Africa: the epidemiological basis for policy

Richard S Cooper, chairman, ^a Charles N Rotimi, assistant professor, ^a Jay S Kaufman, research associate, ^a Walinjom FT Muna, chief of cardiology, ^b George A Mensah, associate professor of medicine. ^c

^a Department of Preventive Medicine and Epidemiology, Loyola University Stritch School of Medicine, Maywood, IL 60153, USA, ^b University of Health Sciences, Yaounde, Cameroon, ^c Medical College of Georgia, Augusta, GA, USA

Correspondence to: Dr Cooper rcooper{at}orion.it.luc.edu

Although enormous challenges persist in the control of infection in sub-Saharan Africa, non-communicable diseases are also important threats to the health of adult Africans. ^{1 2} Controversy exists, however, over the priority these conditions deserve in the competition for scarce resources. It has recently been argued that hypertension treatment, for example, should not be attempted in sub-Saharan Africa given the high costs.³ Unfortunately, these discussions take place in an information vacuum, since it is impossible to define the burden of chronic conditions in societies where health statistics are unavailable.⁴ Cohort studies may serve as a proxy for vital statistics and give approximate answers to questions on the usefulness of treatment for chronic disease.⁵ Hypertension is particularly suited to this model because it is easily diagnosed, highly prevalent, and information on outcomes is plentiful.

View larger version (111K): [in this window] [in a new window]   Although the relative risk of a cardiovascular event in people with high and normal blood pressure is similar in Africa and the United States, the absolute risk is up to 13 times greater in Africans

Hypertension is the most common cardiovascular condition in the world and the problem of defining a strategy for control confronts all societies. Hypertension is fully treatable, but social conditions in Africa make the implementation of blood pressure control programmes difficult. Lack of a clear strategy based on evidence has undermined further these efforts. We outline here the epidemiological data on hypertension that are available to guide health policy in sub-Saharan Africa.

	Burden of hypertension in sub-Saharan Africa

Sub-Saharan Africa is a diverse region comprising 47 countries. It is home to approximately 480 million people.⁶ Among the elite in African society, the model for hypertension control currently in force in Europe and the United States would be entirely appropriate. However, most Africans (fully 75%) live in rural areas and are marginally integrated into the cash economy, while around 20% live in extreme poverty in cities. The challenge lies in developing effective strategies for these sections of society.

Distribution of hypertension
The prevalence of hypertension is low in rural Africa,⁷ and a graded increase is seen in the urban poor and working class.^8-12 Comparing prevalences in studies is difficult, however, as sampling and measurement vary. We recently completed surveys in three communities using a common sampling and measurement protocol (table).¹³ In southwest Nigeria, blood pressure in villagers rose modestly with age compared with values in residents of urban areas. Seven per cent of the rural sample had hypertension (defined as blood pressure greater than 160/95 mm Hg or antihypertensive treatment). High blood pressure was more common among the urban poor from Ibadan in Nigeria (17%), and substantially more prevalent in salaried workers in Harare, Zimbabwe (26%). Figure 1 shows the gradual upward shift in the distributions of blood pressures across these groups.

View larger version (21K): [in this window] [in a new window]   Fig 1.   Probability density distribution of age adjusted systolic blood pressure in three African communities.

Estimates of preventable deaths
These data represent the principal social strata of African society. They include the range of previous estimates and provide a reference point for considering the burden of hypertension. The estimated distribution of the African population between the three sectors was 75% rural, 20% urban poor, and 5% urban salaried workers and elite. Based on a sub-Saharan population of 500 million, half of whom are older than 25, a hypertension prevalence of 5-10% yields 10-20 million cases. If annual mortality is 2%, and 5% of deaths result from hypertension, then approximately 250 000 deaths each year are preventable.

Risk factors
In Africa, as elsewhere, obesity and sodium intake are risk factors for hypertension. ^{12 14-17} In industrialised societies such as the United States, obesity accounts for 25% of cases of hypertension. However, the relative leanness of Africans means that the contribution of obesity to high blood pressure is only around 10%.¹⁵ Psychosocial factors in hypertension have been studied little. Instruments for measuring these factors in African societies have not been developed. No trials of preventive measures that have reduced risk factors for hypertension have been reported from Africa. Drug treatment is therefore the only proved option at present.

Impact of hypertension on mortality
Although the sequelae of hypertension are predictable,¹⁸ the net impact of high blood pressure on all cause mortality is not. Given the high all cause mortality in Africa, and the small proportion of people who reach an age where sequelae are common, the contribution of hypertension is uncertain. We know of only one prospective study that has been published. In three years of follow up of 1200 adults from the rural district of Igboora-pa, Oyo State, Nigeria, the relative risk of death in people with hypertension was 1.6,¹⁹ a value observed in many other studies.²⁰ Mortality in this community was high (2.8% per year) and hypertension was associated with a large attributable risk of about 2% per year. It seems unlikely, however, that all of the attributable risk in people with hypertension resulted solely from cardiovascular diseases. Chronic diseases are a predisposing factor for fatal infection, and this could lead to short survival in patients with stroke, heart failure, or renal insufficiency. Under these circumstances prevalence surveys would underestimate disease burden, and prospective risk estimates would exaggerate the cause specific role of hypertension. If this analysis is correct, this interaction between chronic and acute conditions changes considerably the framework within which the value of treating chronic disease in Africa should be viewed.

	Potential for hypertension control with drugs

Who is a candidate for antihypertensive treatment? An answer to this critical question requires information on projected benefit, feasibility, and cost effectiveness of treatment.^24-28 While practical considerations will be paramount in the end, it remains essential to describe the medical consequences of the decision that is taken. The calculations presented here are preliminary ones, intended to place the value of hypertension treatment in context. In particular, they provide a counter argument to the view of some that treating hypertension is not cost effective in Africa, and that support should be removed.³

Analysis by "number needed to treat"
Data on the benefit of drug treatment for hypertension in industrialised countries probably understate its impact. Observational studies in westernised societies since the 1960s do not reflect the natural history of the disease, given its widespread treatment.²⁹ In trials, some patients in the placebo arm cross over to treatment. ^{30 31} Indeed, early placebo trials contradict the impression from later trials on "mild" hypertension, and should be considered as the background for policy decisions. In the first Veterans Administration cooperative trial, one of 73 patients taking treatment and 27 of the 70 control subjects had a cardiovascular event.^32-34 The excess in absolute event rates was 28% per year. Under these conditions, four patients would need to be treated each year to prevent one cardiovascular event. In the second Veterans Administration trial, in which patients with diastolic pressures of 90-114 mm Hg were enrolled, the annual cardiovascular event rates were 5.5% in the treated group and 16.7% in the control group, and death rates were 1% and 3% respectively. The corresponding number needed to treat values would be nine each year to prevent a cardiovascular event and 50 to prevent a death. In patients with milder hypertension, however, reductions in mortality were substantially smaller, and the number needed to treat rose to over 1000. ^{25 35}

View larger version (31K): [in this window] [in a new window]   Fig 2.   Individual attributable risk, number needed to treat, and population attributable risk in relation to thresholds for hypertension treatment. Smoothed plots were derived from analysis of adult mortality in Igboora-Pa, Nigeria19

	Proposed guidelines for treatment

Choosing an appropriate threshold for treating blood pressure is problematic in Africa. Because resources are scarce, an argument exists for raising the cut off point to reduce the cost. An objective decision, however, requires information that enables benefits to be quantified. The relative risk is the usual basis for determining the treatment threshold. In most studies, the relative risk in people whose blood pressure is greater than 140/90 mm Hg compared with those who have normal blood pressure is around 1.6, and increases 50% with each succeeding 10 mm Hg increase in systolic pressure.²¹ If the probability of an event is low, the absolute benefit of treatment is small, no matter what the relative risk. Annual death rates in cohorts in the United States are 0.15% and 0.30% for people with normal and high blood pressures respectively.²¹ These rates produce the same relative risk as cardiovascular event rates in people with normal and high blood pressure in Africa (2% and 4% respectively). In the United States, the reduction in attributable risk associated with treatment could be 0.15% at most, while in Africa it could be 2%some 13 fold higher. This order of magnitude difference in attributable risk is the central problem in assessing the potential benefit of treatment in Africa.

Absolute versus relative risk
Absolute levels of risk that warrant treatment in Africa are probably in the range of 1%-3%. In the preliminary findings of the prospective study in rural Nigeria, mortality was 2.8% per year in people with normal blood pressure, but 5.1% per year in those with hypertension.¹⁹ If treatment for hypertension eliminated all the excess, then the number needed to treat per year to prevent a death would be 43. Obviously, this estimate is based on the optimistic assumption that 100% control could be achieved, and is an upper bound of what is possible. However, as noted, the Veterans Administration trials gave similar estimates.^32-34

Making a compromise
Although benefit at the level of the individual rises as the treatment threshold increases, the benefit at the population level falls (fig 2). A compromise is required which makes treatment worthwhile for individuals, yet still has an impact on public health. The healthcare system must allocate funds to urgent priorities. All of these decisions ultimately require epidemiological data incorporated into the decision making model. An important attribute of the algorithm shown in fig 2 is its potential to determine a useful level of absolute risk (or the number needed to treat) and to calculate from this the blood pressure that should be used as the cut off point. With this approach it is not necessary to rely on relative risk thresholds adopted by external expert panels.

Costs in relation to numbers needed to treat
Given the data on reductions in mortality associated with diuretic drugs and  blockers, a strong argument exists for using these as standard treatment ^{30 31 37} Costs in Nigeria are 10-15 cents (US$) per tablet, yielding annual expenditure of $36 (£22) for treatment with one drug only. Assuming a number needed to treat of 50, the cost of drugs alone to prevent one death would be $1800 (£1104) in Africa. In the United States, however, the number needed to treat per year to prevent a death is 1354 for people of similar age, and costs for drugs alone to prevent one death range from $14 000 to $1 million (£8589 to £613 496), depending on the drug used.²⁵

	Conclusions

Complex problems confront any attempt to design a public health strategy to control chronic disease in developing countries, particularly Africa. The obligation for the health professions is to assess potential benefit to patients in the local context, without imposing external standards. For example, the rationale for dismissing the value of hypertension treatment was based on the costs estimated in the United States, which are unlikely to apply in Africa.³ Only after the health benefits have been defined can useful discussions take place on the social and political possibility that such treatment will be made available. While acknowledging the obstacles to implementing chronic care for asymptomatic conditions in Africa today, the spectre of low cost effectiveness should not foreclose the scientific debate. Otherwise concern over chronic disease in developing countries is little more than hand wringing. Empirical evidence challenges the assertion that hypertension treatment should not be a health priority, suggesting instead that investment in an organised care system would reap large gains in adult health.

	Acknowledgments

Funding: This work was supported by grants from the NHLBI (HL 45508 and HL 47910).

Conflict of interest: None.

	References

Top References

1. Murray CJL, Lopez AD. Mortality by cause for eight regions of the world: global burden of disease study. Lancet 1997; 349: 1269-1276[Medline].
2. Feachem RG, Jamison DT, Bos ER. Changing patterns of disease and mortality in sub-Saharan Africa. In: Feachem RG, Jamison DT, eds. Disease and mortality in sub-Saharan Africa. , Washington, DC: World Bank, 1991:3-27.
3. In: Feachem RGA, Kjellstron T, Murray CJL, Over M, Phillips MA, eds. The health of adults in the developing world. , Washington DC: World Bank, 1992:273.
4. Kaufman JS, Asuzu MC, Rotimi CN, Johnson OO, Owoaje EE, Cooper RS. The absence of adult mortality data for sub-Saharan Africa: a practical solution. Bull WHO 1997; 75: 389-395[Medline].
5. Asuzu MC, Johnson OO, Owoaje EE, Rotimi CN, Kaufman JS, Cooper RS. Questions on adult mortality. World Health Forum 1996; 17: 373-376[Medline].
6. World Bank. Social indicators of development, 1990. Baltimore, MD: Johns Hopkins University Press, 1991. 
7. Donnison CP. Blood pressure in the African native. Lancet 1929; i: 6-11.
8. Kaufman JS, Barkey N. Hypertension in Africa: an overview of prevalence rates and causal risk factors. Ethnic Dis 1993; 3(suppl): S83-101.
9. Poulter N, Khaw KT, Hopwood BEC, Mugambi M, Peart WS, Rose G, et al. Blood pressure and associated factors in a rural Kenyan community. Hypertension 1984; 6: 801-813.
10. Bunker CH, Ukoli FA, Nwankwo MU, Omene JA, Currier GW, Holifield-Kennedy L. Factors associated with hypertension in Nigerian civil servants. Prev Med 1992; 21: 710-722[Medline].
11. Cooper R, Rotimi C, Ataman S, McGee D, Osotimehin B, Kadiri S, et al. The prevalence of hypertension in seven populations of West African origin. Am J Public Health 1997; 87: 160-168[Abstract/Free Full Text].
12. In: Akinkugbe OO, ed. Non-communicable diseases in Nigeria: final report of a national survey. Lagos: Federal Ministry of Health and Social Services , 1997.
13. Cooper RS, Puras A, Tracy JA, Kaufman JS, Ordunez PO, Mufunda J, et al. Evaluation of an electronic blood pressure device for epidemiologic studies. Blood Pressure Monitoring 1997; 2: 35-40.[Medline]
14. Kaufman JS, Owaje EE, James SA, Rotimi C, Cooper RS. The determinants of hypertension in West Africa: Contribution of anthropometric and dietary factors to urban-rural and socio-economic gradients. Am J Epidemiol 1996; 143: 1203-1218[Abstract/Free Full Text].
15. Kaufman JS, Durazo-Avizu RA, Rotimi CN, McGee DL, Cooper RS, for the International Collaborative Study on Hypertension in Blacks. Obesity and hypertension prevalence in populations of African origin: results from the international collaborative study on hypertension in blacks. Epidemiology 1996;7:398-405.
16. Bennett S, Wilkins HA. Within-person variation in urinary sodium, potassium and creatinine concentrations, and their relationship to changes in the blood pressure of adult male Gambians. J Trop Med Hygiene 1993; 96: 267-273.[Medline]
17. M'Buyamba-Kabangu JR, Fagard R, Lijnen P, Mbuy wa Mbuy R, Staessen J, Amery A. Blood pressure and urinary cations in the urban Bantu of Zaire. Am J Epidemiol 1986;124:957-68.
18. Collins R, Peto R, MacMahon S, Hebert P, Fiebach NH, Ederlein KA, et al. Blood pressure, stroke, and coronary heart disease. Part 2. Short-term reductions in blood pressure: overview of randomized drug trials in their epidemiological context. Lancet 1990; 335: 827-838[Medline].
19. Kaufman JS, Rotimi CN, Brieger WR, Oladukun MA, Kadiri S, Osotimehin BO, et al. The mortality risk associated with hypertension: preliminary results of a prospective study in rural Nigeria. J Hum Hypertens 1996; 10: 461-464[Medline].
20. Stamler R. The primary prevention of hypertension and the population blood pressure problem. In: M Marmot, P Elliott, eds. Coronary heart disease epidemiology: from etiololgy to public health. New York: Oxford University Press, 1992:415-34.
21. Cooper RS, Osotimehin B, Kaufman J, Forrester T. Adult mortality in Africa: what should we conclude in the absence of data? Lancet 1998; 351: 208-210[Medline].
22. Adetunji JA, Murray CJL, Evans TG. Causes of death in Africa: a review. Population Association of America annual meeting, New Orleans, LA, 9-11 May 1996. (Abstract.)
23. Project team, policy implications of adult morbidity and mortality: a preliminary report of the adult morbidity and mortality project, Tanzania, December, 1993. In: Feachem RG, Jamison DT, eds. Disease and mortality in sub-Saharan Africa. Washington, DC: The World Bank, 1991. 
24. Murray CJL, Lopez AD. Evidence-based health policylessons from the global burden of disease study. Science 1996; 274: 740-743[Free Full Text].
25. Pearce KA, Furber CD, Psaty BM, Kirk J. Comparative cost-effectiveness of first-line drugs for uncomplicated hypertension. Am J Hypertens 1998 (in press).
26. In: Jamison DT, Mosley WH, eds. Disease control priorities in developing countries. New York: Oxford University Press , 1993.
27. Whelton PK, Brancati FL, Appel LJ, Klag MJ. The challenge of hypertension and atherosclerotic cardiovascular disease in economically developing countries. High Blood Pressure 1995; 4: 36-45.
28. World Hypertension League. Economics of hypertension control. Bull WHO 1995; 73: 417-424[Medline].
29. Sytkowski PA, D'Agostino RB, Belanger AJ, Kannel WB. Secular trends in long-term sustained hypertension, long-term treatment and cardiovascular mortality. The Framingham heart study 1950 to 1990. Circulation 1996; 93: 697-703[Abstract/Free Full Text].
30. SHEP Cooperative Research Group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension: final results in the systolic hypertension in the elderly program (SHEP). JAMA 1991; 265: 3255-3264[Abstract].
31. Hypertension Detection and Follow-up Program Group. Five-year findings of the hypertension detection and follow-up program. I. Reduction in mortality of persons with high blood pressure, including mild hypertension. JAMA 1979; 242: 2562-2571[Abstract].
32. Freis ED. The Veterans Administration cooperative study. In: Paul O, ed. Epidemiology and control of hypertension. , New York: Symposia Specialists, 1975:449-484.
33. Veterans Administration Cooperative Study Group on Antihypertensive Agents. Effects of treatment on morbidity in hypertension: results in patients with diastolic blood pressure averaging 115 through 129 mm Hg. JAMA 1970; 202: 1028-1034.
34. Veterans Administration Cooperative Study Group on Antihypertensive Agents. Effects of treatment on morbidity in hypertension: results in patients with diastolic blood pressure averaging 90 through 114 mm Hg. JAMA 1970; 213: 1143-1152[Medline].
35. MRC Working Party. Medical Research Council trial of treatment of hypertension in older adults: principal results. BMJ 1982;304:405-12.
36. Chalmers J. Treatment guidelines in hypertension: current limitations and future solutions. J Hypertens 1996; 14(suppl 4): S3-S8.
37. Joint National Committee on Detection, Evalution and Treatment of High Blood Pressure. The fifth report of the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure (JNC V). Arch Intern Med 1993;153:154-83.