A comparison of cardiovascular measurements in the Gambia, Jamaica, and the United Republic of Tanzania*

Bulletin of the World Health Organization 57 (2): 281-289 (1979) A comparison of cardiovascular measurements in the Gambia, Jamaica, and the United Republic of Tanzania * J. P. VAUGHAN' & W. E. MIALL2 Epidemiological studies of the cardiovascular characteristics of three typically rural communities in the Gambia, Jamaica, and the United Republic of Tanzania were carried out by means of standardized methodology. This paper reports comparisons of arterial blood pressure distribution and electrocardiographic findings in relation to age, sex, and body build. Marked differences in blood pressure were found, with higher values in Jamaicans than in Tanzanians, who in turn had higher values than Gambians. These differences are not explicable in terms ofbody build. Heart rates and ECG amplitudes were also strikingly different, with higher values in Jamaicans than in Tanzanians and Gambians. The differences in ECG amplitudes cannot be explained by differences in body build, heart rate, or blood pressure. Thefindings agree with the hypothesis that some factor or factors associated with development contributes to the risk of cardiovascular disease in peoples of African origin. Environmental factors appear to play a major part in determining the differences in the prevalence of cardiovascular disease that have been found both between and within ethnic groups. Although there are societies in which blood pressure levels do not appear to rise with increasing age and where stroke, coronary heart disease, and hypertensive heart dis- ease are rare (1-6), the evidence from epidemiologi- cal surveys of migrant populations suggests that no ethnic group is immune to rises in blood pressure (7-9). Most of the published studies have shown arterial pressure to be higher, and to increase more rapidly with age, in populations that have been more exposed to the way of life in industrialized countries, but the nature of these environmental influences is poorly understood and the relative contributions of such physical factors as diet, salt intake, body build, and exercise on the one hand, and psychosocial and economic factors on the other, remain to be mea- sured. * This work was supported by grants from the Medical Research Councils of East Africa and the United Kingdom. I Senior Lecturer in Tropical Epidemiology, Department of Medical Statistics and Epidemiology and Ross Institute, London School of Hygiene and Tropical Medicine, London WCIE 7HT, England. 2 Member of Scientific Staff, MRC/DHSS Epidemiology and Medical Care Unit, Northwick Park Hospital, Harrow, Middlesex, England. This paper reports results from three standardized epidemiological studies designed to compare the cardiovascular characteristics of three rural popula- tions in the Gambia, Jamaica, and the United Republic of Tanzania. POPULATION SAMPLES The three populations were racially similar though not identical, all being derived from Negroid peoples in Africa. Tanzanians are Bantu and ethnically similar to West African Negroes, but a minority have features suggesting Hamitic ancestry. Gambians are West African Negroes and have a common ancestry with many of those taken to the Americas as slaves from the 16th to the end of the 18th centuries. Jamaicans are descendants of both West African Negroes and Bantu, with different degrees of mis- cegenation. Each of the communities studied was chosen non- randomly as being reasonably typical of rural ag- ricultural populations in the three countries. In the United Republic of Tanzania, the adult population of a community surrounding the village of Kwam- kono in Handeni district, 240 km north of Dar es Salaam, was selected. Kwamkono is the central village of an inland area of rolling bush and scrub- land where the scattered population depends on a subsistence farming economy based on maize, rice, 3799 -281- J. P. VAUGHAN & W. E. MIALL and cassava; it has a few village shops, a small mission hospital, nearby primary schools, and daily communications by truck and bus, on dirt roads, with neighbouring villages and with the capital (18). In the Gambia, the population of an isolated village, Keneba, about 160 km from the capital, Banjul (formerly Bathurst), was selected. Keneba has a total population of about 800, 25 % of whom are Jolas of slave origin and the remainder are of free-born Mandinka ancestry. Both men and women are engaged in different aspects of the cultivation of groundnuts, millet, maize, and vegetables, with the women largely responsible for the rice crop; the economy is vulnerable to the effects of drought and rainy seasons (10, 11). Communication is by ox cart or bicycle on dirt roads; there is no sanitation, no market, and no school and the standard of living is lower than that of Kwamkono. The village of Keneba has been the centre of a number of field studies by the British Medical Research Council laboratories in Fajara, the Gambia. In Jamaica, a community living in a hilly area at Lawrence Tavern was chosen as representative of inland farming populations on the island (12). The population is largely composed of smallholders and their families, each growing a variety of fruit and vegetable crops on plots of land that are too small and too steep for mechanized farming. Lawrence Tavern is 24 km from Kingston, the capital, with which it has regular communication by bus and truck on asphalt roads; there are primary and secondary schools, a government clinic, mains electricity and piped water to the village itself, and some food shops. The standard of living appears, superficially, to be somewhat higher than that in Kwamkono, and considerably higher than that in Keneba. METHODS In each survey, people were asked to attend for cardiovascular investigations which were carried out at central clinics by means of the methodology recommended by the World Health Organization (13). In the United Republic of Tanzania the community was defined by private census and a random sample of those aged 25-64 years was selected; 75 % of the men but only 66% of the women listed in the census attended. However, the census was undertaken 6 months before the study and the survey included 89% of the men and 84% of the women still resident in the area. In the Gambia, 299 (92 %) of the 332 adults aged 25 years and over believed to be living in the village were seen, but 20 persons aged 65 years or more were excluded from the analysis. Men were less well represented than women. In Jamaica, random sam- ples of 200 men and 200 women in each of the three decades from 35 to 64 years of age were selected, and 89% of these 1200 took part in the study. The age and sex of the populations under age 65 are shown in Table 1. As few people knew exactly how old they were, age was determined by questioning and by comparison with a local history of events. Table 1. Representative rural population studies in the Gambia,a Jamaica, and the United Republic of Tanzania Population No. examined Response rateAge sample (%) group (years) Tan- Ja- Tan- e ia- Tan- Ja- zania maica zania Gambia maica zania maica Men 25-34 109 - 81 28 - 74 - 35-44 110 200 88 24 168 80 84 45-54 113 200 75 26 180 66 90 55-64 109 200 89 28 178 82 89 Total 441 600 333 106 526 75.5 87.7 Women 25-34 114 - 66 57 - 58 - 35-44 106 200 94 53 181 89 91 45-54 107 200 58 40 181 54 91 55-64 90 200 57 21 179 63 90 Total 417 600 275 171 275 65.9 90.2 'A total population sample was attempted and a complete demo- graphic census was not available. This comparison of cardiovascular characteristics is restricted to objective measurements. Blood pressure determinations were made after at least 5 min of rest, with the person seated, by means of a random zero sphygmomanometer (14) in Jamaica and the United Republic of Tanzania (18) and by means of a standard mercury manometer in the Gambia. Twelve-lead electrocardiograms (ECGs) were recorded on portable instruments, the same technician using the same instrument in Jamaica and the Gambia. All tracings were read in duplicate according to Minnesota Code criteria (13) and one of us (W.E.M.) coded the ECGs for all three studies. All ECGs were measured for certain amplitudes and 282 CARDIOVASCULAR MEASUREMENTS IN SUBJECTS OF AFRICAN ORIGIN 283 intervals. Height and weight were recorded by sex, for the three populations in Tables 2 and 3. In means of a portable stadiometer and a beam scale in men (Table 2), both systolic and diastolic pressures each study, and body build was expressed as were lowest in Gambians and highest in Jamaicans. Quetelet's index (weight/height2). No significant increase in either systolic or diastolic pressure with age was found in Tanzanian or Gam- RESULTS bian men, although the age trend among Jamaicans Arterial blood pressure distributions was highly significant for both pressures (P<O.O1). Mean values for systolic and diastolic (phase IV) Quetelet's index was similar in Tanzanians and pressure and Quetelet's index are shown, by age and Gambians, but significantly higher in Jamaicans. Table 2. Arterial blood pressure a and Quetelet's indexb in men, by age, in representative rural populations in the Gambia, Jamaica, and the United Republic of TanzaniaC Systolic pressure Diastolic pressure Quetelet's index Age group (years) Number Mean SD Mean SD Mean SD 25-34 Tanzania 81 120.6 (16.0) 11.4 74.6 ( 9.9) 11.8 2.84 0.25 Gambia 28 117.1 (15.6) 11.4 69.6 ( 9.2) 8.3 2.87 0.24 Jamaica - - - - - - - 35-44 Tanzania 88 124.5 (16.5) 19.3 77.2 (10.3) 11.9 2.89 0.26 Gambia 26 116.1 (15.4) 23.9 71.1 ( 9.4) 12.3 2.79 0.54 Jamaica 168 128.3 (17.1) 18.4 80.6 (10.7) 12.8 3.21 0.35 45-54 Tanzania 75 123.0 (16.3) 22.2 76.0 (10.1) 14.8 2.83 0.41 Gambia 24 117.9 (15.7) 25.1 71.1 ( 9.4) 13.3 2.72 0.55 Jamaica 180 134.5 (17.9) 22.5 84.4 (11.2) 13.5 3.20 0.33 55-64 Tanzania 89 124.9 (16.6) 19.1 77.5 (10.3) 13.9 2.71 0.41 Gambia 28 121.4 (16.1) 25.6 72.9 ( 9.7) 12.5 2.71 0.53 Jamaica 178 142.3 (18.9) 27.8 85.5 (11.4) 14.8 3.10 0.31 'Blood pressures are shown in mmHg; figures in parentheses are the equivalent pressures in kPa. bQuetelet's index = weight (lb)/ height2 (in) x 100. cThe figures quoted above for the United Republic of Tanzania differ slightly from those given previously (18), because in the above analysis only one pressure reading per subject is used whereas two were used previously. Table 3. Arterial blood pressure aand Quetelet's index' in women, by age, in representative rural populations in the Gambia, Jamaica, and the United Republic of Tanzania Systolic pressure Diastolic pressure Quetelet's index Age group (years) Number Mean SD Mean SD Mean SD 25-34 Tanzania 66 118.0 (15.7) 16.0 72.2 ( 9.6) 12.1 2.82 0.28 Gambia 57 110.1 (14.6) 15.8 66.0 ( 8.8) 9.8 2.86 0.28 Jamaica - - - - - - - 35-44 Tanzania 94 118.6 (15.8) 20.0 74.4 ( 9.9) 11.9 2.88 0.34 Gambia 53 105.5 (14.0) 15.7 66.1 ( 8.8) 9.9 2.79 0.51 Jamaica 181 127.9 (17.0) 20.1 82.7 (11.0) 12.4 3.48 0.62 45-54 Tanzania 58 124.9 (16.6) 19.2 75.9 (10.1) 12.6 2.72 0.39 Gambia 40 114.7 (15.2) 22.7 72.0 ( 9.6) 10.6 2.68 0.50 Jamaica 181 145.3 (19.3) 29.0 90.1 (12.0) 15.9 3.47 0.72 55-64 Tanzania 57 126.4 (16.8) 20.2 75.8 (10.1) 14.4 2.66 0.32 Gambia 21 129.4 (17.2) 38.7 77.3 (10.3) 16.6 2.56 0.57 Jamaica 179 156.5 (20.8) 28.7 91.1 (12.1) 15.8 3.27 0.55 ' Blood pressures are shown in mmHg; figures in parentheses are the equivalent pressures in kPa. b Quetelet's index = weight (lb)/height2 (in) x 100. J. P. VAUGHAN & W. E. MIALL Among women (Table 3), systolic and diastolic pressures were again lowest in Gambians and high- est in Jamaicans, except in the 55-64-year age group, where the mean values for the small group of Gambian women were higher than for Tanzanian women. All three female populations showed an increase in both pressures with age, and these were statistically significant with the exception of that for diastolic pressure among Tanzanian women. Quetelet's index was again similar in Tanzanian and Gambian women, but significantly greater for Jamaican women. Fig. 1 and 2 show the frequency distributions of systolic and diastolic pressures in Jamaicans and Tanzanians. In each age/sex group the distributions of both systolic and diastolic pressures were shifted, and more skewed, towards higher values in Jamaica. Arterial pressure was, in general, positively corre- lated with body build, although in the Jamaican survey statistically significant coefficients were pre- sent only between Quetelet's index and systolic pressure in men aged 45-54 and 55-64 years and between Quetelet's index and diastolic pressure in men aged 45-54 (Table 4). Allowing for the differ- ences in Quetelet's index between the populations by linear regression analysis does not abolish the differences in blood pressure between Tanzanians and Jamaicans. In Table 5 the blood pressures of Tanzanians and Jamaicans are compared, assuming a similar body build in the two populations. 20 1 Men 10 - O-J 20] 1O a Women d.I1mmm.u Jamaica Tanzanz 20 - 10 - O J 20 - 1J 20 - 10 - 0J 20] 10] O- Jamaica Tanzania L ,-B1Thm 4f Tn I I I- I I I I i 80 120 160 200 240 I I I I I I I I__ l 80 120 160 200 240 Systolic pressure (mmHg) Fig. 1. Frequency distributions of systolic pressure in rural populations in Jamaica and the United Republic of Tanzania, by age and sex. 1 mmHg = 0.133 kPa. Id 35-44 years I 45-54 years 55-64 years - - - - - - -- -- ELMdOMM-M6M - 284 CARDIOVASCULAR MEASUREMENTS IN SUBJECTS OF AFRICAN ORIGIN Men Women L Jamaica 20 10] 20 0 3 r20 anzania a ma ica anzani 35-44 years 45-54 years LcaJmaica iT ~~TanzaniaS I I I II I I 40 80 120 160 40 80 120 160 Diastolic pressure (mmHg ) Fig. 2. Frequency distributions of diastolic (IV) pressure in rural populations in Jamaica and the United Republic of Tanzania, by age and sex. 1 mmHg = 0.133 kPa. Table 4. Regression equations of systolic and diastolic Electrocardiographic findings blood pressures (y) on Quetelet's index (x) in the Jamaican population Sex( A(eagrs)p Systolic Diastolic Men 35-44 y = 110.17 + 5.51 x y = 72.19 + 2.49 x 45-54 y = 93.11 + 12.92ax y = 52.45 + 9.97bX 55-64 y = 98.36 + 14.29 ax y = 65.30 + 6.57 x Women 35-44 y = 127.07 + 0.29 x y = 77.94 + 1.37 x 45-54 y = 137.67 + 2.09 x y = 82.76 + 2.09 x 55-64 y = 170.19 - 4.70 x y = 86.93 + 1.10 x a Significant (P< 0.05). b Significant (P < 0.001). Minnesota Code readings. In Table 6, the preva- lence of certain ECG abnormalities is shown for men and women aged 35-64 years in the three popula- tions, coded according to the criteria of the Min- nesota Code (13). The prevalence of abnormal Q/ QS patterns and left axis deviation was low in all three populations. High-amplitude R waves were much more common in Jamaicans, being five times more frequent in men and 2-3 times more frequent in women. Depression of the S-T segment and inversion of the T wave in leads other than the right 20 10 0 ] 20 10 - 20 103- 55-64 years 285 J. P. VAUGHAN & W. E. MIALL Table 5. Mean arterial blood pressures in mmHg of Jamaicans and Tanzanians after allowing for differ- ences in body build. Figures in parentheses are the equivalent pressures in kPa Men WomenAgeagroup Population(years) Population Systolic Diastolic Systolic Diastolic 35-44 Tanzanian 124.5 77.2 118.6 74.4 (16.6) (10.3) (15.8) (10.0) Jamaican 126.1 79.4 127.9 81.9 (16.8) (10.6) (17.0) (10.9) 45-54 Tanzanian 123.0 76.0 124.9 75.9 (16.3) (10.1) (16.6) (10.1) Jamaican 129.7 80.7 143.3 88.4 (17.2) (10.7) (19.0) (11.7) 55-64 Tanzanian 124.9 77.5 126.4 75.8(16.6) (10.3) (16.8) (10.1) Jamaican 137.1 83.1 157.7 89.9 (18.2) (1 1.0) (21.0) (11.9) chest leads were also more frequent in Jamaicans, whereas T wave inversion restricted to leads V2-V4 (the so-called African T waves) was much more common in Gambians and Tanzanians than in Jamaicans. ECG measurements. Means and standard devia- tions of heart rate and of the amplitudes (in mm) of R in aVL, S in V1, and R in V5 and V6 are shown in Table 7, as measured from the tracings. There were striking differences in heart rate, with the lowest rates in the Gambia and the highest in Jamaica in most age/sex groups. For men, R wave amplitudes in aVL, V5, and V6 were consistently highest in Jamaicans and lowest in Tanzanians; the pattern was similar, but with smaller differences and less consis- tent, among the women. Adjusting the R wave amplitudes of Jamaicans to those that would be expected were they of the same age, body build, systolic and diastolic pressure, and heart rate as Tanzanians, using multiple regression equations derived from the Jamaican data (15), does not abolish the marked differences between the ECGs of Tanzanians and Jamaicans, as illustrated in Table 8 for the values of R in V5. DISCUSSION Marked differences in cardiovascular characteris- tics were found among three populations from rural areas of the Gambia, Jamaica, and the United Republic of Tanzania. The study communities were not randomly selected and were therefore not rep- resentative of the national populations and compari- sons should be strictly limited to these samples. Blood pressure distributions were shifted towards Table 6. Prevalence of certain ECG abnormalities in representative rural populations (35-64 years) in the Gambia, Jamaica, and the United Republic of Tanzania Men Women ECG Minnesota Tanzania Gambia Jamaica Tanzania Gambia Jamaica abnormality Code (n = 252) (n = 78) (n = 526) (n = 209) (n = 104) (n =541) No % No % No % No % No % No % Q/QS patterns 11.2 1 0.4 - - 9 1.7 2 1.0 1 1.0 6 1.1 13 2 0.8 - - 8 1.5 4 1.9 - - 7 1.3 Left axis 2, 7 2.8 1 1.3 20 3.8 6 2.9 2 1.9 7 1.8 deviation High-amplitude 31 15 6.0 4 5.1 154 29.3 11 5.3 5 4.8 67 12.4 R waves S-T depression 41.3 4 1.6 1 1.3 19 3.6 2 1.0 3 2.9 29 5.4 T wave inversion: 51-2 in lateral leads 3 1.2 1 1.3 19 3.6 4 1.9 2 1.9 12 2.2 in right chest leads 8 3.2 2 2.6 - - 10 4.8 12 11.5 6 1.1 in other leads - - - - - - 3 1.4 1 1.0 11 2.0 Sinus tachycardia 87 5 2.0 - - 22 4.2 16 7.7 2 1.9 76 14.0 Sinus bradycardia 89 2 0.8 14 17.9 12 2.3 - - 3 2.9 - - Low-amplitude 91 17 6.7 1 1.3 19 3.6 14 6.7 1 1.0 21 3.9 R waves 286 CARDIOVASCULAR MEASUREMENTS IN SUBJECTS OF AFRICAN ORIGIN Table 7. Certain ECG characteristics, by age and sex, in representative rural populations in the Gambia, Jamaica, and the United Republic of Tanzania. Heart rate is in beats/min; R, S, and T wave amplitudes are in millimetres Age group (years) Heart rate R in aVL S in Vl R in V5 R in V6 T in V6 Number Mean SD Mean S D Mean S D Mean S D Mean SD Mean SD Men 35-44 Tanzania Gambia Jamaica 45-54 Tanzania Gambia Jamaica 55-64 Tanzania Gambia Jamaica Women 35-44 Tanzania Gambia Jamaica 45-54 Tanzania Gambia Jamaica 55-64 Tanzania Gambia Jamaica 88 69.1 10.7 2.4 1.8 26 56.9 12.6 2.6 1.3 168 72.6 12.3 3.1 2.7 75 72.7 13.5 1.8 1.6 24 55.8 10.2 2.9 2.4 180 71.4 14.4 4.0 3.5 89 71.8 15.2 2.2 2.2 28 59.9 11.2 3.0 2.6 178 72.0 13.7 4.1 3.5 94 80.0 12.3 2.5 2.2 53 67.3 13.3 3.2 2.2 181 83.3 15.0 4.3 3.2 58 76.9 13.3 2.8 2.7 40 66.7 12.6 3.6 2.3 181 84.3 15.5 4.7 3.1 57 78.6 19.9 2.9 2.4 21 65.3 13.9 3.7 2.2 179 84.1 16.4 5.8 3.8 11.4 5.5 16.7 6.1 10.1 4.0 10.1 5.3 17.9 7.4 10.6 4.3 12.5 5.3 21.2 8.5 17.1 6.8 9.8 4.9 15.2 5.6 8.7 3.2 10.6 5.9 17.4 6.2 10.0 3.8 11.4 5.6 21.1 8.1 16.7 6.5 10.0 5.9 15.8 7.7 8.9 4.3 9.6 4.6 16.5 7.7 10.1 5.2 12.2 7.7 21.3 8.9 16.5 6.8 10.0 5.2 14.1 4.4 8.6 2.8 11.7 4.4 16.4 5.7 10.3 3.4 10.6 4.6 15.3 5.6 13.0 4.2 10.2 5.1 14.6 5.1 8.6 3.2 11.4 4.5 18.1 5.4 11.4 3.9 11.3 5.5 17.4 5.9 14.3 4.9 11.3 5.4 17.2 7.6 9.6 4.7 10.9 3.7 16.3 7.0 9.0 3.9 12.5 5.3 20.1 8.0 16.3 6.5 Table 8. Mean amplitudes of R in V5 (mm) in Tanza- nians compared with those in Jamaicans of similar age, body build, systolic and diastolic pressure, and heart rate Age group Men Women (years) Tanzanians Jamaicans Tanzanians Jamaicans 35-44 16.7 21.1 14.1 15.5 45-54 15.2 20.2 14.6 17.8 55-64 15.8 19.8 17.2 19.3 160 r 140 L 0-4 Uganda *-- Jamaica o-_o Nigeria Tanzania la-- Gambia MALES 0 0- 120 l.. - 20 30 40 50 60 Age in years higher values in both sexes and all age groups in 1X - Jamaicans compared with Tanzanians, who in turn had higher values than Gambians. Heart rates, the / prevalence of Q/QS abnormalities and left axis ,. FEMALES deviation, and the prevalence of tall R waves in the v lateral leads of the electrocardiogram appeared to i follow a similar pattern. - Fig. 3 shows the mean systolic blood pressure 120 - levels from the Gambian, Jamaican, and Tanzanian surveys, together with those of two large-scale Afri- can studies based on more developed and accultu- rated groups. A Nigerian survey was in a "village Age in yers conglomeration" with a population of 22 000 (163 Fig. 3. The mean systolic blood pressure for five and some Ugandan data are from a study in Kasan- different community studies in Negro peoples of Afri- gati, about 20 km from the capital city, Kampala ca. 1 mmHg = 0.133 kPa. 2.5 2.4 3.5 2.3 2.5 3.5 2.3 2.3 2.9 1.9 1.8 2.2 2.0 2.0 2.0 2.0 1.7 2.2 1.3 1.1 2.2 1.3 1.3 2.0 1.2 1.4 2.2 1.0 0.9 1.0 1.0 1.1 1.5 1.1 0.7 1.7 287 J. P. VAUGHAN & W. E. MIALL (17). In both males and females, the Nigerian and Ugandan levels were very similar to the rise with increasing age shown in Jamaicans, whereas those for the Gambia and the United Republic of Tan- zania showed much lower levels with little or no rise with increasing age. The surveyed populations in Jamaica, Nigeria, and Uganda were designated as "rural" by their investigators, but they cannot be said to have been very rural when Laurence Tavern in Jamaica was only 24 km from the capital King- ston, Ilora in Nigeria was a village of 22 000 people, and Kasangati in Uganda was only 20 km from Kampala. These findings agree with the hypothesis that some factor or factors associated with development contributes to cardiovascular risk in populations of African origin, and the differences in body build that parallel these differences in cardiovascular measure- ments tend to support this concept within these three communities. Adjusting blood pressure levels for body build, however, had only a marginal effect on the differences between Jamaican and Tanzanian levels, in either sex. Similarly, adjusting ECG am- plitudes for differences in body build, blood pressure, and heart rate did not abolish the marked differences between Jamaicans and Tanzanians in left ventricular hypertrophy, as judged by the am- plitudes of R waves in left lateral chest leads. In an analysis of the quantitative aspects of ECGs in Jamaicans, Ruiz et al. (15) showed, by multiple regression methods, that only 4% of the variance of the amplitude of R in V5 in men, and 22% in women, was explained by a consideration of the five variables age, Quetelet's index, systolic pressure, diastolic pressure, and heart rate. Clearly, there are major influences within and between similar populations that affect cardiovascu- lar variables and which are not being taken into consideration in such analyses. It appears likely that these influences are largely environmental in nature rather than genetic. This particularly applies to our understanding of blood pressure in developing coun- tries. ACKNOWLEDGEMENTS We gratefully acknowledge the help of the people of Kwamkono, Keneba, and Lawrence Tavern who participated in these surveys and also the help given by the field teams in the Gambia, Jamaica, and the United Republic of Tanzania. In particular we would like to thank Dr L. Ruiz and Miss S. Standring for ECG measurements, Dr T. W. Meade and Mrs V. Jutsom for Minnesota coding, and Mr P. Brennan for statistical assistance. RItSUM1t COMPARAISON DES CARACTtRISTIQUES CARDIO-VASCULAIRES EN GAMBIE, EN JAMAIQUE ET EN RtPUBLIQUE-UNIE DE TANZANIE Des 6tudes epidemiologiques des caracteristiques cardio- vasculaires de trois communautes typiquement rurales en Gambie, en Jamaique et en Republique-Unie de Tanzanie ont ete faites en appliquant des methodes uniformes. Le present article a pour objet la comparaison de la distribu- tion des niveaux de pression arterielle et des resultats de l'electrocardiogramme, consignes par Age, sexe et corpu- lence. De fortes differences de pression arterielle ont ete constatees, avec des valeurs plus elevees chez les Jamai- cains que chez les Tanzaniens, qui a leur tour accusent des pressions plus hautes que les Gambiens. Ces 6carts ne s'expliquent pas par les differences de corpulence. Le rythme cardiaque et l'amplitude de l'onde R presentent aussi de tres fortes differences, des valeurs plus elev6es etant notees chez les Jamaicains que chez les Tanzaniens et les Gambiens. Les ajustements operes dans l'amplitude de l'onde R en fonction de la corpulence, de la pression arterielle et du rythme cardiaque n'ont pas fait disparaitre les importants ecarts constates dans cette amplitude (tableau 8). Les resultats de l'etude vont dans le sens de l'hypothese selon laquelle un ou plusieurs facteurs associ6s au developpement contribueraient au risque de maladies cardio-vasculaires dans les populations africaines ou d'ori- gine africaine. 288 CARDIOVASCULAR MEASUREMENTS IN SUBJECTS OF AFRICAN ORIGIN 289 REFERENCES 1. LOWENSTEIN, F. W. Blood pressure in relation to age and sex in the tropics and sub-tropics. Lancet, 1: 389- 392 (1961). 2. EPSTEnN, F. H. & ECKOFF, R. D. The epidemiology of high blood pressure-geographic distributions and etiological factors. In: Stamler, J. et al., ed. The epidemiology of hypertension. New York, Greene & Stratton, 1967, p. 155. 3. LOVELL, R. R. H. Race and blood pressure. In: Stamler, J. et al., ed. The epidemiology of hyperten- sion. New York, Greene & Stratton, 1967, p. 122. 4. HUIZINGA, J. Casual blood pressure in populations. In: Vorster, D. J. M., ed. Human biology and environmen- tal change. London, International Biological Program- me, 1971, p. 164. 5. SHAPER, A. G. Communities without hypertension. In: Shaper, A. G. et al., ed. Cardiovascular disease in the tropics. London, British Medical Association, 1974, p. 77. 6. VAUGHAN, J. P. A brief review of cardiovascular disease in Africa. Transactions of the Royal Society of Tropical Medicine and Hygiene, 71: 226-231 (1977). 7. PAGE, L. B. ET AL. Antecedents of cardiovascular disease in six Solomon Island societies. Circulation, 49: 1132-1146 (1974). 8. CASSEL, J. Studies of hypertension in migrants. In: Paul, O., ed. Epidemiology and control of hyperten- sion, New York, Stratton Intercontinental Medical Book Corporation, 1975, p. 41. 9. VAUGHAN, J. P. A review of cardiovascular disease in developing countries. Annals of tropical medicine and parasitology, 72: 101-109 (1978). 10. MCGREGOR, D. A. ET AL. Growth and mortality in children in an African village. British medical journal, 2: 1661-1666 (1961). 11. REE, G. H. Arterial pressures in a West African (Gambian) rural population. Journal of tropical medicine and hygiene, 76: 65-70 (1973). 12. MIALL, W. E. ET AL. Longitudinal study of heart disease in a Jamaican rural population. Bulletin of the World Health Organization, 46: 429-442, 685-694, 695-708 (1972). 13. ROSE, G. & BLACKBURN, H. Cardiovascular survey methods. Geneva, World Health Organization, 1968 (Monograph Series, No. 56). 14. WRIGHT, B. M. & DoRE, C. F. A random zero sphygmomanometer. Lancet, 1: 337-338 (1970). 15. RuIz, L. ET AL. Quantitative aspects of electrocardio- grams of adults in a Jamaican rural population. British heart journal, 35: 829-839 (1973). 16. AKINKUGBE, 0. 0. & Ojo, 0. A. Arterial pressures in rural and urban populations in Nigeria. British medical journal, 2: 222-224 (1969). 17. SHAPER, A. G. & SAXTON, G. A. Blood pressure and body build in a rural community in Uganda. East African medical journal, 46: 228-245 (1969). 18. VAUGHAN, J. P. Blood pressure and heart murmurs in a rural population in the United Republic of Tanzania. Bulletin of the World Health Organization, 57: 89-97 (1979).