(Hypertension. 1995;26:616-623.)
© 1995 American Heart Association, Inc.
Articles |
Weight Threshold and Blood Pressure in a Lean Black Population
From the Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh (Pa) (C.H.B., A.M.K., S.L.H., L.H.K.); Department of Community Health, University of Benin Teaching Hospital, Benin City, Nigeria, West Africa (F.A.U.); and Department of Psychiatry, University of Pittsburgh (Pa) School of Medicine (K.A.M.).
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract Hypertension is virtually absent in very lean rural African populations but is becoming more common in higher-weight urban African populations and is very common in predominantly obese Westernized black populations. This implies that there is a threshold above which weight is related to blood pressure. We studied urban Nigerian civil servants, a lean population in transition toward a more Westernized lifestyle. Blood pressure, fat-related measurements, fasting insulin, physical activity, alcohol intake, macronutrient intake, and electrolyte excretion were measured in 500 male and 299 female civil servants in Benin City, Nigeria, in 1992. Median body mass index (BMI) was 21.5 kg/m2 in men and 24.0 kg/m2 in women. Examination of age-adjusted mean blood pressure across quantiles of BMI in men and women suggested a threshold of 21.5 kg/m2 below which blood pressure was not correlated with BMI. Above this threshold blood pressure was correlated with BMI. Comparison of groups above and below the lower BMI threshold found that differences in blood pressure–BMI covariation were not explained by differences in alcohol intake, caloric or macronutrient intake, or electrolyte excretion. Physical activity was higher in men below the threshold. Fasting insulin and waist-hip ratio were strongly correlated with BMI even in this very lean population but neither was independently related to blood pressure. We conclude that there is a threshold below which little relationship between blood pressure and weight is observed. Above this threshold even at levels considered lean in US blacks, weight is a major determinant of blood pressure in this population of African blacks, which shares ancestry with US blacks.
Key Words: Africa, Western • blacks • blood pressure • insulin • exercise • male • obesity
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Obesity is an important correlate of blood pressure (BP) in Westernized black populations.1 2 3 In these populations both body mass index (BMI) and BP increase across age groups. Since most studies have involved predominantly overweight Westernized blacks, few data exist regarding the consistency of the relationship of weight and BP across the range of weight from thin to obese. However, early cross-sectional studies of very lean African black populations found that neither weight nor BP increased across age groups.4 5 More recent studies of somewhat heavier African populations have found that both weight and BP increase across age groups.6 7 8 9 10 This suggests that there is a threshold weight above which BP increases with weight, although the influence of very different lifestyles and environments may also make important contributions.
It has been hypothesized that insulin insensitivity is an important mediator of the relationship between obesity and BP. Numerous studies in whites, though not entirely consistent,11 12 tend to support this hypothesis, but the few studies available in blacks are less consistent.13 14 15 16
In the present study we have examined the relationship of BP to BMI and weight-related factors to determine whether there are thresholds in these relationships in a population of Nigerian civil servants. We have also examined a number of factors that might explain or confound these relationships, eg, dietary intake, alcohol intake, physical activity, and electrolyte excretion. This population is of interest because it appears to be in transition toward a more Westernized lifestyle and toward higher levels of BP.10 However, it is still on average a lean population and provides the opportunity for studying these relationships across a range of BMI values not easily observed in Westernized black populations.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Study Protocol
The protocol was approved by the Biomedical Institutional Review Boards for the University of Pittsburgh and the University of Benin Teaching Hospital. Both verbal and written informed consents were obtained from all subjects. The study was described as a health survey.
Recruitment of Subjects
The study was carried out in 1992 in three large Edo state
government ministries located in Benin City, Nigeria. A
study-conducted census was used to update personnel lists. The goal
was total recruitment of all censused employees of each ministry.
Subjects were recruited by work area.
Data Collection
Each subject participated in a series of three visits scheduled
over one week. At each visit BP was measured three times based on first
and fifth Korotkoff sounds with the use of a conventional mercury
sphygmomanometer by certified observers according to a standardized
protocol.17 The average of the second and third readings
at all visits was used in data analysis. Height was measured
with a metal tape, which was fixed to the wall, and a right angle
device. Weight was measured to the nearest 0.1 kg with a digital
bathroom scale that was checked periodically by weighing carefully
measured volumes of water. The waist was measured at the narrowest
point or at the umbilicus in the absence of a narrowest point. Hips
were measured at the widest part of the buttocks. All measurements were
made with subjects in light clothing without shoes.
Questionnaires were used to collect data on demographic factors. Average leisure and occupational physical activities over the past year were assessed with a detailed questionnaire developed for use in population studies.18 Activities were weighted by an estimate of the relative intensity of the activity, termed the metabolic equivalent (MET; ratio of the metabolic rate during the activity to the resting metabolic rate). Time spent on each activity was weighted by its MET value, summed across activities, and reported here as MET hours per week averaged over the past year.18 Included in the occupational section of the physical activity questionnaire was an estimate of the minutes spent walking or biking to work each day. A 24-hour dietary recall was collected for a weekday and a Sunday. Standard local measures and crude food models were used to assist with quantification of intake. Data were entered and analyzed with the NUTRITIONIST IV computer program (version 3.0, 1993, N-Squared Computing). The program was customized by addition of local foods. Macronutrient values for these foods were taken from published sources.19 20 21 22 Average intake from the two recalls was used in data analysis. Average weekday and average weekend-day intakes of beer, palm wine, wine, hard liquor, and ogogoro (locally brewed hard liquor) were assessed with detailed questions. Alcohol data were converted to grams of absolute ethanol per week.
Fasting blood samples were drawn with a 15-mL plain red-topped vacuum tube. A drop of whole blood was removed from the tube immediately after the blood draw and was used for measurement of whole blood glucose with the use of a One-Touch II glucometer (Lifescan, Johnson & Johnson). The calibration of the glucometer was checked daily with a standard test strip supplied by the manufacturer. Tubes of blood were held at ambient temperature for 1 to 2 hours until serum was separated by centrifugation for 15 minutes at ambient temperature. Serum was aliquoted and placed on ice for 3 to 5 hours until transportation to storage at -20°C. Within 3 months samples were hand carried on dry ice to the University of Pittsburgh, where they were stored at -70°C. Fasting insulin was measured at the University of Pittsburgh.23
Detailed verbal and written instructions were provided for a 24-hour urine collection. After measurement of urine volume, urine aliquots were placed on ice for 3 to 5 hours and then stored at -20°C. Twenty-four-hour urinary sodium and potassium was measured at the University of Pittsburgh with flame photometry.
Civil Service Staff, Junior Staff (nonprofessional staff, salary grades 1 through 6), and Senior Staff (professional and administrative staff, salary grades 7 through 16) were used as markers of socioeconomic status.
Statistical Analysis
Men and women and high and low BMI groups were compared with
t tests. Means adjusted for age and other factors were
calculated and tested by ANOVA. Spearman correlations were calculated
between BMI and BP and other variables. To examine the possibility
of a threshold relationship between BP and BMI, we plotted BP across
sex-specific quartiles, sextiles, octiles, and deciles of BMI with
the goal of finding the largest number of groups providing a relatively
smooth curve for the estimation of the threshold point. Curves for
octiles and smaller numbers of groups were relatively smooth in men,
and octiles were chosen for presentation. Sextiles were
used for women because of the smaller sample size. Among men there
appeared to be a threshold around the median BMI of 21.5
kg/m2. To explore the region of the threshold, we performed
a series of regressions using BMI cut points at 19, 20, 21, 21.5, 22,
23, 24, 25, and 26 kg/m2. The slopes between systolic BP
and BMI in the groups above and below each of these cut points were
examined. The slope closest to zero was observed in the group below the
BMI cut point of 21.5 kg/m2. The slope in this group
(B=0.25) did not differ from zero (P=.072), whereas the
slope in the group above this cut point did differ from zero (B=1.26,
P=.002). Change in BP across age groups and across BMI
quantiles, adjusted for age, was tested by ANOVA and the test of
linearity. With the use of the median value of each independent
variable to form high and low groups, the relationship of BMI and
fasting insulin, BMI and waist-hip ratio, and BMI and physical
activity to BP was analyzed by two-way ANCOVA adjusted for
age.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Of 592 censused men, 42 (7%) refused to participate and 46 (8%) were absent from work at the time of the study. Of the 504 (85%) that participated, 3 subjects outside the age range of 20 to 64 years and 1 subject with skeletal abnormalities that precluded accurate height measurement were excluded, resulting in 500 male subjects considered here. Among the 378 censused women, 10 pregnant women were excluded. Of the 368 remaining, 35 (9%) refused to participate, 28 (8%) were absent from work, and 305 (83%) participated, of whom 303 were in the age range and 299 had complete height and weight data.
Ninety-eight percent of men and 86% of women originated from the Edo group of tribes. Education beyond secondary school was reported by 31% of men and 25% of women. Forty-two percent of men and 40% of women were Senior Staff. Mean age was 42.3 years among men and 38.6 years among women.
Data for men and women are presented in Table 1. Despite lower BMI, smaller waists and hips, lower fasting insulin, and higher physical activity, BP was higher in men than in women. Men were older; had higher alcohol intake; higher caloric, carbohydrate, protein, and fat intakes; and higher potassium excretion. Sodium excretion and fasting glucose did not differ between men and women. All differences between men and women remained significant after adjustment for age.
|
BP and BMI increased across age groups in both men and women (Fig 1). BP was lower in all age groups in women but increased more steeply across age groups in younger women than in younger men and approached that of men in older age groups. In the youngest age group BMI was the same in men and women but rose sharply in women up to the age group of 35 to 44 years, whereas in men BMI rose steadily across age groups but at a much slower rate, remaining lower than in women.
|
Spearman correlations between risk factors and BP are shown in Table 2. Among men, age and several fat-related variables—BMI, waist girth, waist-hip ratio, and insulin—were related to BP. Minutes walking to work was negatively related to BP. Alcohol intake, caloric and protein intakes, and sodium and potassium excretions were not related to BP. Age, BMI, and waist girth were related to BP among women, whereas fasting insulin and other variables were not related. The Spearman correlations between BMI and waist girth, waist-hip ratio, and fasting insulin were 0.81, 0.43, and 0.46 (all P<.001), respectively, in men and 0.84, 0.21, and 0.44 (all P<.001) in women.
|
When age-adjusted systolic and diastolic BPs were plotted across octiles (men) or sextiles (women) of BMI (Fig 2), there appeared to be an S-shaped relationship. Among men systolic and diastolic BP values were more or less flat across the first (mean BMI, 17.3 kg/m2) through fourth (mean BMI, 21.2 kg/m2) octiles. BP rose steeply across the fourth to seventh (mean BMI, 24.6 kg/m2) octiles, increasing 11 mm Hg for systolic and 8 mm Hg for diastolic BP, and then did not increase between the seventh and eighth (mean BMI, 28.4 kg/m2) octiles. In women a similar though less strong pattern was observed. Systolic BP increased 8 mm Hg and diastolic 6 mm Hg from the second (mean BMI, 20.9 kg/m2) to fourth (mean BMI, 24.8 kg/m2) sextile of BMI. There was no increase below the second sextile and no increase and even an apparent decrease from the fourth to sixth (mean BMI, 31.7 kg/m2) sextile.
|
BP was also examined across quantiles of fasting insulin, waist-hip ratio, waist girth, and physical activity (minutes walking to work) (Fig 3). The pattern of the relationship of BP with insulin and waist girth was very similar to that with BMI in men. BP and waist-hip ratio were not strongly related. BP generally decreased with increasing physical activity. Among women systolic BP was not related to fasting insulin, waist-hip ratio, or waist girth (Fig 3). Diastolic BP increased in the midpart of the distributions of fasting insulin, waist-hip ratio, and waist girth, but the relationships were not strong. In contrast to men, BP did not decrease with increasing physical activity in women.
|
The median BMI in men, 21.5 kg/m2, was used as an estimate of the lower threshold. Half of the men and about one third of the women were below this BMI threshold. Men below the threshold are compared with those above the threshold in Table 1. In addition to higher weight and BP, men above the BMI threshold were older; had higher waist and hip girths, waist-hip ratio, fasting insulin, protein intake, and sodium excretion; and spent less time walking to work. Alcohol intake was not higher. Women above the threshold (Table 1) also had higher waist and hip girths, waist-hip ratio, fasting insulin, and sodium excretion compared with those below the threshold. In addition, fasting glucose, caloric intake, and potassium excretion were increased. These women reported higher total physical activity but did not differ from those below the threshold in minutes walking or biking to work.
The relationship of BP, BMI, waist girth, and insulin in the low and high BMI groups across age groups in men and women is shown in Fig 4. Among men in the low BMI group, BP, BMI, waist girth, fasting insulin, and physical activity did not increase across age groups, whereas waist-hip ratio increased linearly. In contrast, in women in the low BMI group BP increased considerably, whereas BMI, waist girth, waist-hip ratio, and physical activity did not change significantly and fasting insulin decreased across age groups. Above the BMI threshold, BP, BMI, waist girth, and waist-hip ratio increased linearly across age groups, whereas fasting insulin remained constant in men and physical activity did not change significantly. In women BP increased across age groups, whereas BMI, fasting insulin, and waist-hip ratio did not increase. Waist girth increased modestly but significantly, physical activity decreased across age categories in this group.
|
Fig 5 shows age-adjusted systolic BP by median BMI and median fasting insulin, median BMI and median waist-hip ratio, and median BMI and median physical activity in men and women. High insulin was not related to BP if BMI was low. If BMI was high, high insulin was associated with higher BP in men and women, but the relationship was not statistically significant. There was a significant interaction between BMI and waist-hip ratio such that in the low BMI group BP was lower in those subjects with high waist-hip ratio, whereas the highest BP was observed in subjects with high BMI and high waist-hip ratio in both men and women. A high level of physical activity was associated with lower BP in both the low and high BMI groups in men. The opposite pattern was observed in women, among whom high levels of physical activity were associated with higher BP in both low and high BMI groups.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The findings in this study support the hypothesis that a threshold exists below which there is little relationship between weight and BP. In this population the threshold was located at a BMI of approximately 21.5 kg/m2. This finding was more apparent in men, among whom half were below the BMI threshold, whereas only one third of the women were below this point. The relationship between BMI and BP was not confounded by alcohol intake, caloric or macronutrient intakes, or sodium or potassium excretions.
A threshold effect for BMI and BP is implicit in the contrast between primitive populations, with low BMI and no increase in BP with age,4 5 and urban populations, in which both BMI and BP generally increase across age groups.6 7 8 9 10 However, longitudinal data confirming a threshold, or cross-sectional data from which a threshold can be inferred, are sparse. The study of Yi migrants in China24 does suggest a similar threshold effect for BP at a BMI of approximately 21.5 kg/m2. A threshold effect for BP and other cardiovascular risk factors was identified at less than 15 mm subscapular skinfold (mean BMI, 21.1 kg/m2) among white women in the Framingham Offspring study.25 A search for a low BMI threshold for BP in most Westernized black populations is not possible because the great majority of the subjects are well above the proposed threshold. The present study, in which BMI was less than 21.5 kg/m2 in half of the men and one third of the women, provides the first data, though only cross-sectional, suggesting a threshold for BMI and BP in a black population. This is relevant to Westernized black populations because of the genetic background they share with this population.10
Among the population above the BMI threshold, even though the subjects were still quite lean by Western standards, weight-related factors were important correlates of BP, particularly in men. Age, BMI, and waist girth were the strongest correlates of BP in men and women. The determination of whether weight (or central weight) or weight-related metabolic factors that may influence BP more directly, eg, serum insulin, made the most important contribution to BP was difficult because of lack of linearity and multicolinearity. The study of the corelationships of BP, BMI, fasting insulin, and central weight distribution with age permitted further dissection of the multicolinearity among the latter variables with regard to BP. Cross-sectional examination of BP and BMI across age groups among the men below the BMI threshold found the same pattern as in earlier cross-sectional studies of very lean rural men,4 5 namely, that neither BP nor BMI increases with age (Fig 4). Despite no change in waist circumference or insulin, waist-hip ratio increased sharply and was unrelated to BP. In contrast, among men above the BMI threshold, BMI, waist girth, waist-hip ratio, and BP increased across age groups as in cross-sectional studies of Westernized populations. However, fasting insulin did not increase across age groups, providing indirect evidence that the relationship of BP to BMI and central weight was not mediated by increasing insulin insensitivity.
In contrast to men, BP increased across age groups even in the very lean group of women, although none of the weight-related variables increased. BP and waist girth increased across age groups in the heavier group of women, but as in men fasting insulin did not change. Thus, despite the higher BMI observed in women than men, there was no evidence for an important role for insulin insensitivity determining BP. There appeared to be unidentified age-related factors that influence BP in women.
Insulin metabolism and the increase in insulin insensitivity that generally accompanies obesity have frequently been hypothesized as aspects of a major mechanism by which obesity influences BP. This is supported by a number of studies in predominantly white populations that have found independent relationships between insulin measures and BP, although some studies are not consistent with this finding (reviewed in References 11 and 1211 12 ). Only a few studies examining the relationship between insulin and BP have been reported among blacks. In black children in the Bogalusa Heart Study, baseline fasting insulin measured at ages 5 through 11 years was not related to subsequent systolic BP over 6 years, although in white children this relationship was present independent of weight and other risk factors.15 In young African American adults, fasting insulin levels and insulin resistance were higher in borderline hypertensive than in normotensive individuals independent of weight.16 The relationship of insulin resistance to BP was stronger in subjects with a BMI less than 28 kg/m2. Above a BMI of 28 kg/m2, insulin resistance was related only to BMI and not to BP in this small study. In a large study of predominantly nonobese young adults, fasting insulin levels were associated with BP in blacks and whites after adjustment for BMI and other cardiovascular risk factors.13 In a study comparing Pima Indian, white, and black normotensive obese adults, fasting plasma insulin and insulin resistance were related to BP only among whites but not among Pima Indians or blacks.14 The current study also did not find an independent relationship between insulin and BP in an adult black population, in this case, a lean population. Although sparse in the upper end of the BMI distribution, the data suggest an upper plateau in the relationship with BP that may or may not continue across higher levels of BMI. Part of the inconsistency between studies may depend on whether BMI distribution of the population covers a BMI range that is primarily linearly related to BP or is a primarily plateau range. In addition, the data from the current study suggest that there may be a disassociation between BMI and insulin across age groups, at least in some BMI ranges. Thus, varying age distributions between populations could also contribute to inconsistencies between studies. The inconsistency may also be due to statistical instability of models resulting from the multicolinearity between insulin-related measures and BMI and the relative biological and measurement instabilities of insulin-related measures compared with BMI. On the other hand, there may not be a direct relationship between insulin metabolism and BP, but rather each is related to obesity. The occurrence of plateaus in the relationship between insulin and BP, if confirmed in other studies, suggests that insulin is not directly regulating BP. A recent review of animal and human studies concluded that chronic hyperinsulinemia does not affect BP regulation.12
Central weight deposition, as assessed by waist-hip ratio, was not related to BP independent of BMI in men and was only weakly related in women. This agrees with studies of Nigerians and African Americans across a wide range of age, BMI, and waist-hip ratio. Among college-age Nigerian and African American men and women,26 27 very weak correlations between BP and waist-hip ratio were found in both populations. Mean waist-hip ratios in these groups were less than 0.82 in men and 0.77 in women. Waist-hip ratio was very similar in Nigerians and African Americans despite differences in mean BMI (20.4 versus 24.8 kg/m2, respectively, in men and 20.5 versus 22.3 kg/m2, respectively, in women).27 Mean waist-hip ratios similar to those in these studies were observed in the large CARDIA study of young adult African American men (mean BMI, 24.6 kg/m2) and women (mean BMI, 25.8 kg/m2).28 A very small but significant correlation between waist-hip ratio was observed with systolic BP but not with diastolic BP independent of percent body fat. Adjustment for fasting insulin had very little effect on the correlation. In a much heavier large population of adult black men (mean BMI, 26.3 kg/m2; waist-hip ratio, 0.89) and women (mean BMI, 29.3 kg/m2; waist-hip ratio, 0.85) waist-hip ratio was not related to diastolic BP in men or women after adjustment for percent body fat.29 A borderline significant relationship with systolic BP was observed in women but not in men. In another adult black population similar to the previous population in levels of BMI and waist-hip ratio, waist-hip ratio was not related to the occurrence of hypertension in men or women after adjustment for BMI, glycosylated hemoglobin, age, and family history of high BP.3 In an elderly population of blacks, again with similar levels of BMI and considerably higher levels of waist-hip ratio, the occurrence of hypertension was not related to waist-hip ratio.30 An index of fat distribution, based on waist and estimated hip girths in the 1960-62 Health Examination Survey, was associated with hypertension in black women but not in black men after adjustment for age and ponderosity.31 Thus, among black adults, central weight, as assessed by waist-hip ratio, appears to make little contribution to BP and hypertension beyond the influence of weight, although central weight may be somewhat more important in women than in men.
Contrary to expectation, energy intake was not related to BP. In contrast, energy expenditure was related to both weight and BP. Higher levels of physical activity, as measured by minutes spent walking or biking to work daily, were independently related to both lower weight and lower BP in men. Among women, high and low weight groups did not differ in physical activity. The relationship between physical activity and BP was not significant, but the trend was opposite that expected; ie, higher physical activity was associated with higher BP in women. No data are available to explain this association. However, women generally did not walk long distances to work if they could avoid it. Thus, the anecdotal impression emerging at data collection was that the women who walked long distances to work were experiencing a more difficult and stressful lifestyle.
Differences in optimal weight set points or in sensitivity to central weight gain may contribute to differences in hypertension rate between ethnic groups or between men and women. In the BMI distribution included in the present study there is no strong evidence implicating weight-related factors in women. However, among men the data suggest a low weight threshold. It is unknown whether this apparent set point observed among African men is applicable to African American men. However, it would seem to be a prudent hypothesis. Overweight in the US population is defined as BMI greater than or equal to the 85th percentile among 20- to 29-year-olds.32 For men this cutoff is at a BMI of 27.8 kg/m2 or 85.1 kg (188 lb) for a man 175 cm (69 in) tall. If similar cutoffs are appropriate for African American and Nigerian men, our data suggest that this cutoff is about 20 kg (44 lb) above optimal weight. Between ages 35 and 64 years, approximately 85% to 90% of US black men have a BMI higher than 21.5 kg/m2.32 A downward revision of criteria for overweight could occur after validation of the relationship with BP across a wide range of BMI values and confirmation from longitudinal studies that maintenance of low BMI is truly optimal with regard to BP and does not increase morbidity and mortality. The data are consistent with the hypothesis that optimal intervention strategies should be directed toward prevention of weight gain beginning in early adulthood or earlier. Increased physical activity is likely to be a key component of these interventions.
|
Acknowledgments |
|---|
This work was supported in part by research grant HL-44413 from the National Institutes of Health.
|
Footnotes |
|---|
Reprint requests to Clareann H. Bunker, PhD, Department of Epidemiology, A542 Crabtree Hall, University of Pittsburgh, 130 DeSoto St, Pittsburgh, PA 15261. E-mail bunkerc+@pitt.edu.
Received May 12, 1995; first decision June 14, 1995; accepted June 14, 1995.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Stamler J. Epidemiologic findings on body mass and blood pressure in adults. Ann Epidemiol. 1991;1:347-362. [Medline] [Order article via Infotrieve]
2.
Folsom AR, Burke GL, Byers CL, Hutchinson RG, Heiss G,
Flack JM, Jacobs DR Jr, Caan B. Implications of obesity for
cardiovascular disease in blacks: the CARDIA and ARIC
studies. Am J Clin Nutr. 1991;53:1604S-1611S.
3.
Lackland DT, Orchard TJ, Keil JE, Saunders DE Jr,
Wheeler FC, Adams-Campbell LL, McDonald RH, Knapp RG. Are race
differences in the prevalence of hypertension explained by body mass
and fat distribution? A survey in a biracial population.
Int J Epidemiol. 1992;21:236-245.
4. Kaminer B, Lutz WP. Blood pressure in bushmen of the Kalahari Desert. Circ Res. 1960;12:289-295.
5. Shaper AG, Wright DH, Kyobe J. Blood pressure and body build in three nomadic tribes of Northern Kenya. E Afr Med J. 1969;46:273-281. [Medline] [Order article via Infotrieve]
6. Johnson, TO. Prevalence of overweight and obesity among adult subjects of an urban African population sample. Br J Prev Soc Med. 1970;24:105-109. [Medline] [Order article via Infotrieve]
7. Johnson TO. Arterial blood pressures and hypertension in an urban African population sample. Br J Prev Soc Med. 1971;25:26-33. [Medline] [Order article via Infotrieve]
8. Oviasu VO, Okupa FE. Arterial blood pressure and hypertension in Benin in the equatorial forest zone of Nigeria. Trop Geogr Med. 1980;32:241-244. [Medline] [Order article via Infotrieve]
9. Lang T, Pariente P, Salem G, Tap D. Social, professional conditions and arterial hypertension: an epidemiological study in Dakar, Senegal. J Hypertens. 1988;6:271-276. [Medline] [Order article via Infotrieve]
10. Bunker CH, Ukoli FA, Nwankwo MU, Omene JA,Currier GW, Holifield-Kennedy L, Freeman DT, Vergis EN, Yeh LLL, Kuller LH. Factors associated with hypertension in Nigerian civil servants. Prev Med. 1992;21:710-722. [Medline] [Order article via Infotrieve]
11. Weir MR, Tuck ML. Essential hypertension in blacks: is it a metabolic disorder? Am J Kidney Dis. 1993;4(suppl 1):58-67.
12.
Hall JE. Renal and
cardiovascular mechanisms of hypertension in
obesity. Hypertension. 1994;23:381-394.
13.
Manolio TA, Savage PJ, Burke GL, Liu K, Wagenknecht LE,
Sidney S, Jacobs DR Jr, Roseman JM, Donahue RP, Oberman A.
Association of fasting insulin with blood pressure and lipids in young
adults: the CARDIA Study.
Arteriosclerosis. 1990;10:430-436.
14. Saad MF, Lilloja S, Nyomba BL, Castillo C, Ferraro R, De Gregorio M, Ravussin E, Knowler WC, Bennett PH, Howard BV, Bogardus C. Racial differences in the relation between blood pressure and insulin resistance. N Engl J Med. 1991;324:733-739. [Abstract]
15. Jiang X, Srinivasan SR, Bao W, Berenson GS. Association of fasting insulin with longitudinal changes in blood pressure in children and adolescents: the Bogalusa Heart Study. Am J Hypertens. 1993;6:564-569. [Medline] [Order article via Infotrieve]
16.
Falkner B, Hulman S, Kushner H.
Insulin-stimulated glucose utilization and borderline hypertension
in young black adults. Hypertension. 1993;22:18-25.
17. Dischinger P, DuChene AG. Quality control aspects of blood pressure measurements in the Multiple Risk Factor Intervention Trial. Control Clin Trials. 1986;7:137S-157S. [Medline] [Order article via Infotrieve]
18. Kriska AM, Bennett PH. An epidemiological perspective of the relationship between physical activity and NIDDM: from activity assessment to intervention. Diabetes Metab Rev. 1992;8:355-372. [Medline] [Order article via Infotrieve]
19. Oyenuga JA. Nigeria's Foods and Feeding-Stuffs. Ibaden, Nigeria: Ibaden University Press; 1968.
20. Platt BS. Tables of Representative Values of Foods Commonly Used in Tropical Countries. London, England: Her Majesty's Stationery Office; 1962. Privy Council, Medical Research Council Special Report Series No. 302.
21. Food Composition Table for Use in Africa. Bethesda, Md: US Dept of Health, Education, and Welfare; Food and Agriculture Organization of the United Nations; 1968.
22. Nutritive Value of American Foods in Common Units, Agriculture Handbook No. 456. Washington, DC: US Dept of Agriculture, 1988.
23.
Herbert V, Lau S, Gottlieb CW. Coated charcoal
immunoassay of insulin. J Clin Endocrinol. 1965;25:1375-1384.
24.
He J, Klag MJ, Whelton PK, Chen JY, Qian MC, He
GQ. Body mass and blood pressure in a lean population in
southwestern China. Am J Epidemiol. 1994;139:380-389.
25. Garrison RJ. Healthy adiposity in women: the Framingham offspring study. J Am Coll Nutr. 1993;12:357-362. [Abstract]
26. Adams-Campbell LL, Nwankwo M, Ukoli F, Omene J, Haile GT, Kuller LH. Body fat distribution patterns and blood pressure in black and white women. J Natl Med Assoc. 1990;82:573-576. [Medline] [Order article via Infotrieve]
27. Adams-Campbell LL, Wing R, Ukoli FA, Janney CA, Nwankwo MU. Obesity, body fat distribution, and blood pressure in Nigerian and African-American men and women. J Natl Med Assoc. 1994;86:60-64. [Medline] [Order article via Infotrieve]
28.
Folsom AR, Burke GL, Ballew C, Jacobs DR Jr, Haskell
WL, Donahue RP, Liu K, Hilner JE. Relation of body fatness and
its distribution to cardiovascular risk factors in
young blacks and whites: the role of insulin. Am J
Epidemiol. 1989;130:911-924.
29. Croft JB, Strogatz DS, Keenan NL, James SA, Malarcher AM, Garrett JM. The independent effects of obesity and body fat distribution on blood pressure in black adults: the Pitt County study. Int J Obes. 1993;17:391-397.
30. Stevens J, Gautam SP, Keil JE. Body mass index and fat patterning as correlates of lipids and hypertension in an elderly, biracial population. J Gerontol. 1993;48:M249-M254. [Abstract]
31. Gillum RF. The association of body fat distribution with hypertension, hypertensive heart disease, coronary heart disease, diabetes and cardiovascular risk factors in men and women aged 18-79 years. J Chronic Dis. 1987;40:421-428. [Medline] [Order article via Infotrieve]
32. Najjar MF, Rowland M. Anthropometric reference data and prevalence of overweight, United States, 1976-80. In: Vital Health Stat [11]. Washington, DC: US Government Printing Office; 1987. National Center for Health Statistics, DHHS publication (PHS) 87-1688.
This article has been cited by other articles:
 |
|
 |
|
  J. Addo, L. Smeeth, and D. A. Leon Hypertension In Sub-Saharan Africa: A Systematic Review Hypertension, December 1, 2007; 50(6): 1012 - 1018. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. H. Kuller Epidemic Hypertension in Sub-Saharan Africa Hypertension, December 1, 2007; 50(6): 1004 - 1005. [Full Text] [PDF]
|
|
|
|
 |
|
 P. B. Gorelick Cerebrovascular Disease in African Americans Stroke, December 1, 1998; 29(12): 2656 - 2664. [Full Text] [PDF]
|
|
|
|
|
|
J. S. Kaufman, M. C. Asuzu, J. Mufunda, T. Forrester, R. Wilks, A. Luke, A. E. Long, and R. S. Cooper Relationship Between Blood Pressure and Body Mass Index in Lean Populations Hypertension, December 1, 1997; 30(6): 1511 - 1516. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||