JOURNAL TRANSCRIPT

---

Journal of Human Hypertension (2004) 18, 775–779 & 2004 Nature Publishing Group All rights reserved 0950-9240/04 $30.00 www.nature.com/jhh

ORIGINAL ARTICLE

Evidence for affluence-related hypertension in urban Brazil L Ala1, G Gill1, R Gurgel2 and L Cuevas1 1

Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, UK; 2Nu´cleo de Po´s-Graduac¸a˜o em Medicina –Federal University of Sergipe, Aracaju-Sergipe, Brazil

Cardiovascular diseases are a leading cause of mortality, and systemic hypertension is a major risk factor. There is an increasing prevalence of hypertension in urban areas of developing countries, due to lifestyle changes associated with economic transition and urbanisation. This study aimed to describe the prevalence and identify risk factors for hypertension in an urban area of South America (Coroa do Meio district in Aracaju, Sergipe State, north-east Brazil) and to examine intraurban hypertension prevalence differences. A cross-sectional survey of 400 adults aged 25 years and over was carried out. Information about health and lifestyle was obtained from a structured interview, followed by assessment of blood pressure (BP) and anthropometry. There were 31.8% (95% confidence interval 27.3–36.6%) participants with hypertension (de-

fined as a systolic BP 4140, diastolic BP 490 mmHg, or on antihypertensive medication). Hypertension was independently associated with older age, central obesity (greater waist-to-hip ratio), shorter height and residing in a high socio-economic residential area. Of the four neighbouring areas, hypertension prevalence was 52% in the area of highest income and education, compared with 19, 24 and 34% in the other three areas. The high prevalence of hypertension in this population, and the strong independent association with relative affluence, demonstrates the need for effective primary prevention of hypertension, targeted at modifiable risk factors. Journal of Human Hypertension (2004) 18, 775–779. doi:10.1038/sj.jhh.1001750 Published online 24 June 2004

Keywords: Brazil; risk factors; blood pressure measurement; urbanisation

Introduction Cardiovascular diseases are the leading cause of premature death in industrialised countries1 and the World Health Organisation (WHO) predicts that by the year 2020 the mortality from cardiovascular disease in low- to middle-income countries will exceed deaths from infectious and parasitic diseases in all regions except sub-Saharan Africa.2 Paradoxically, some of the reasons for the upsurge in cardiovascular diseases relate to positive factors such as the decline in childhood mortality and infections.3 The increasing life expectancy and lifestyle changes brought about by urbanisation and industrialisation present an ideal setting for cardiovascular disease epidemics with high cholesterol, diabetes, reduced physical activity, increased salt, tobacco and alcohol consumption4 and hypertension,5 which together with foetal and infant malnutrition result in an unfavourable profile of risk factors.6 Moreover, new generations are subjected to these factors for a longer period, and Correspondence: Dr GV Gill, Liverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA, UK. E-mail: [email protected] Received 25 September 2003; revised 20 February 2004; accepted 27 February 2004; published online 24 June 2004

therefore developing cardiovascular disease at a younger age.7 While there has been a small decline in the prevalence of hypertension and coronary disease in some industrialised countries,8 low- to middleincome countries have shown a steady increase over the last few decades,9 with a gradient towards higher prevalence in urban areas and groups with higher salary grades.10–12 This study describes the prevalence of hypertension in an urban adult population in north east (NE) Brazil and identifies the most common risk factors, and explores differences in prevalence rates and risk factors in different socioeconomic areas within the urban study area.

Methods Geography

We performed a cross-sectional survey in Coroa do Meio in Aracaju, NE Brazil. This is a relatively new low- to middle-income residential area built on partly reclaimed land. The area has a 12 500 population, one health centre and is divided into four health ‘macro-areas’. Macro-area 053 was the first-quarter developed as a residential area and

Affluence and hypertension in urban Brazil L Ala et al 776

macro-areas 043, 024 and 061 are less developed with poor housing in areas marked for future development. About 70% of the residents have migrated from outside the city, a process beginning in the early 1970s. The area thus represents an accessible and compact urban development of mixed affluence. Sample size and sampling

Participants were selected by systematic random sampling. An initial household was selected at random and every sixth household was visited. Adults aged 25 years or over were investigated, excluding pregnant women. Interviews were conducted between 0800 and 1700 h during weekdays, and also weekends and public holidays to enrol working individuals. If there was no response to a call, or if potential subjects were absent, then the adjacent house was visited until a response was obtained. This only occurred in seven houses, or 2% of the total and the strategy ensured an effective 100% response rate. The sample size was based on estimated hypertension prevalence of 30%13–15 with an error margin of 75%. A sample size of 400 participants was chosen to investigate risk factors with odds ratios (OR) over 2.5 and a power of 80%. This sample size was also adequately powered for logistic regression analysis.

were referred to their local health centre, and all participants received an information brochure on hypertension. The body mass index (BMI) was defined as the body weight in kilograms divided by the square of height in metres. The waist-to-hip ratio (WHR) was calculated as the ratio of the waist to the hip circumference, and the abdomen to thigh ratio (ATR) as the ratio of the abdominal to the midthigh circumferences. Statistical analysis

The prevalence of hypertension by age was calculated as percentages with 95% confidence intervals (95% CI). Associations between risk factors and hypertension were analysed, and the crude odds ratios and w2 were calculated using 2
2 tables. These were obtained using a backward, stepwise logistic regression approach. Linear regressions were used to assess correlations between anthropometrical values and blood pressure. P-values of below o0.05 were considered statistically significant for the multivariate analysis Ethical approval

Approval for the study was obtained from the Ethical Committee of the Liverpool School of Tropical Medicine, and the Ethics Committee of the Federal University of Sergipe, Brazil.

Measurements

Interviews were conducted with structured questionnaires. Blood pressures (BP) were taken from the left arm, with suitably sized cuffs using a mercury sphygmomanometer (UNITEC). All measurements were made based upon World Health Organisation (WHO) and International Society of Hypertension (ISH) guidelines.16 Phase I and V Korotkoff sounds were taken as systolic and diastolic pressures, respectively, and read to the nearest 2 mmHg. A single BP level was taken, and all measurements were made by one observer. Hypertension was defined as a systolic or diastolic BP over 140 or 90 mmHg, respectively, as well as and those on antihypertensive medication.16 Weight was measured to the nearest 0.5 kg using a Seca portable scale, and height was measured with a stadiometer to the nearest 0.5 cm. Waist and hip circumference were taken at the narrowest section of the trunk between the costal margins and the iliac crest, and the widest section of the buttocks over the greater trochanters, respectively. Abdominal circumference was obtained as the widest part of the abdomen between the umbilicus and the symphysis pubis, and mid-thigh circumference half-way between the lateral epicondyle and the greater trochanter. Circumferences were read to the nearest 0.5 cm with the subject wearing light clothing and standing upright. Newly diagnosed hypertensive patients Journal of Human Hypertension

Results In all, 260 (65%) of the 400 participants were female and 140 male. Males had a mean age of 45715 (s.d.) and females 42714 years. In total, 315 (79%) participants were born outside the city and had lived in Aracaju for a mean of 18712 years. Most participants had a marital partner (73%) and low levels of education (see Table 1). There were 92% of women who had been pregnant, with a mean of 4.5 pregnancies each. The mean BMI was 25.6 for males

Table 1 Characteristics of the study participants (n ¼ 400) Age (years) Body mass index (BMI) Waist/hip ratio (WHR) Abdomen/thigh ratio (ATR) Migrants (%) Regular partner Pregnancies (women) Number of children Education (a) Incomplete primary (b) Primary (c) Secondary (d) Tertiary Employment Figures are presented as percentages, or mean7s.d.

43714 25.774.7 0.970.1 1.970.3 79% 73% 4.574.1 3.572.9 65% 16% 15% 4% 45%

Affluence and hypertension in urban Brazil L Ala et al 777

Table 2 Characteristics of the study population by residence in macro-area Area 024 (n ¼ 112)

Area 043 (n ¼ 85)

Area 053 (n ¼ 102)

Area 061 (n ¼ 101)

40713 50% 18711 25.474.6 0.970.1 1.870.2 675 1.971.5 19% (12–27)

43713 68% 20711 25.975.0 0.970.1 1.970.2 474 1.571.2 34% (24–45)

50714* 59% 24715** 26.275.0 0.970.1 1.970.3 876* 4.475.7* 52% (42–62)*

41714 64% 18713 25.274.5 0.970.1 1.870.3 574 1.871.5 24% (16–33)

Age (years) Born in rural area (%) Years after migration BMI WHR ATR Years at school Minimum salary*** Hypertension (%)

Data for macro-area 053 significantly different from other areas (*Po0.01, and **o0.05) The salary figures assume that 1.0 is the national minimum salary***

Table 3 Prevalence of hypertension by age and sex Age band (years) 25–34 35–44 45–49 55–64 65–74 475 Total

Males (%) 6/43 6/32 13/26 10/21 9/14 2/4 46/140

(14%) (19%) (50%) (48%) (64%) (50%) (33%)

Females (%) 5/88 15/87 19/34 21/28 13/15 8/8 81/260

(6%) (17%) (56%) (75%) (87%) (100%) (31%)

and 25.7 for females. Males were a mean 10 cm taller than females and had higher WHRs (Po0.01 for both). Abdomen to thigh ratio (ATR) and WHR were correlated with BMI, although ATR changed more with age than the WHR (see Table 1). The prevalence of hypertension was 31.8% (95% CI 27.3–36.6) and macro-area 053 had the highest prevalence (52.0%) with a mean systolic BP of 131 mmHg. The highest mean diastolic BP however was found in macro-area 043 (82 mmHg). The characteristics of the macro-areas are shown in Table 2. Participants of macro-area 053 were older, had lived in Aracaju for longer and had better education and income. The prevalence of hypertension increased with age. Overall, 46 males (33%) and 81 (31%) females were hypertensive, and hypertension was more prevalent in young males than young females, though this trend reversed after 45 years of age (see Table 3). Hypertensive patients were older, more likely to have diabetes, higher BMI, WHR and ATR, but less likely to have a partner and shorter height (see Table 4). Female hypertensive patients had a higher number of pregnancies. Low education, lack of significant physical activity at work and being born in rural areas were associated with hypertension. Factors not associated with hypertension included family history, alcohol consumption, smoking, not partaking in physical leisure activities and salary level. Significant risk factors at the univariate level were fitted into a logistics regression to remove confounding influences. An age over 45 years, WHR

over 0.85, height over 160 cm and living in macroarea 053 were independently associated with hypertension.

Discussion The prevalence of hypertension (31.8%) in this population is similar to reports from other developing countries using similar diagnostic criteria.9,14,17 It has been reported that hypertension affects up to 40% of the adult Brazilian population and is higher in the urban areas.13 Most studies however come from the prosperous southern part of the country.13 Catanduva (Sao Paulo state), for example (a municipality with high socioeconomic profile), had a prevalence of 32%,14 Bambui reported 45% in adults15 and Pelotas 20%.18 In contrast, a small study of the Xavante Indians in rural central Brazil described a prevalence of 5% in males and 8% in females. In comparison, it is estimated that 10–20% of whites (Caucasians) and 25–35% of blacks (AfroCaribbeans) in Britain,19 16–29% of Australians and 15–23% of New Zealanders are hypertensive. A number of previous studies have shown that rural–urban migration increases markedly the risk of hypertension.10–12,20 A particularly interesting finding of our study is that even within high-prevalence urban areas, hypertension rates can vary according to socioeconomic status. Thus, macro-area 053 had a hypertension prevalence rate of 52%, compared with rates of 19–34% in the other three areas. Incomes and educational levels were significantly higher in this area, and our statistical analysis showed that geographical location here was an independent risk factor for hypertension. Although the prevalence of hypertension in our study increased with age, an important number (9%) of hypertensives were less than 35 years old. This is in keeping with findings from developing countries where cardiovascular diseases tend to develop at a younger age21 as urban transition and westernisation of lifestyles are more likely to affect the young who migrate and become exposed to risk factors. Similar to Whelton’s findings,22 hypertension in our female Journal of Human Hypertension

Affluence and hypertension in urban Brazil L Ala et al 778

Table 4 Risk factors for hypertension reaching statistical significance Risk factor Age445 years Diabetes BMI 425.0 WHR 40.85 ATR 42.0 Height o1.60 m Sedentary work Resident area 053 From rural area Local resident 414 years No partner 44 pregnancies Incomplete 11 education

Hypertensive (%)

Normotensive (%)

75 16 71 73 49 69 69 42 68 49 35 38 74

20 9 58 37 13 55 45 18 55 33 23 24 60

OR (95% CI)

P-value

AOR (95% CI)

11.8 5.5 2.8 4.7 6.5 1.9 2.7 2.3 1.7 1.9 1.8 1.9 1.9

o0.01 o0.01 o 0.01 o0.01 o0.01 o 0.01 o0.01 o 0.01 o0.05 o0.01 o0.05 o0.01 o0.01

8.0 (4.6–13.8)

(7.1–19.4) (2.4–13.0) (1.8–4.6) (3.0–7.6) (3.9–11.0) (1.2–2.9) (1.8–4.3) (2.0–5.2) (1.1–2.6) (1.3–3.0) (1.2–2.9) (1.2–3.0) (1.2–3.0)

2.7 (1.5–4.9) 2.4 (1.3–4.1) 2.9 (1.6–5.2)

OR ¼ odds ratio, AOR ¼ adjusted odds ratio. For definition of ‘hypertensive’ and ‘normotensive’, see ‘Methods’ section. The following factors were not significantly associated with hypertension—gender, smoking, alcohol consumption, unemployment, lack of leisure physical activity and family history of either hypertension, stroke or myocardial infarction.

participants exceeded that of males in older age groups. There is however substantial interaction between the risk factors investigated; and only age, WHR, height and living in macro-area 053 were independent predictors of hypertension. Central obesity is a known risk marker for cardiovascular disease23 probably because it is associated strongly with hypertension, dyslipidaemia, glucose intolerance and insulin resistance.24 The WHR and waist circumference are thought to be better at detecting central obesity than BMI.25 In our study, WHR changed less than ATR with increasing age, making it a better indicator of central obesity across the age spectrum. It is probable that people with a higher WHR did less physical work and that females with high WHR had more pregnancies, accounting for the association between number of pregnancies and hypertension. There is an increasing recognition that dietary patterns and lifestyles with increasing obesity are changing in developing countries, posing an increase in the risks of developing hypertension.26 We have shown that WHR, a measure of central obesity was significantly associated with hypertension, stressing the importance of diet and lifestyle modification as an effective preventative measure against this disease. Sichieri et al27 reported that Brazilians who fall into the lower quartile of height are more likely to be hypertensive. Short stature is a marker of poor development, and low birth weight and malnutrition in early life have been linked to increased cardiovascular risk in later life.6 In developed countries it is generally accepted that lower socioeconomic classes are more likely to develop hypertension.28 This is in contrast to developing countries where a higher socioeconomic status has been more consistently implicated as a risk factor for hypertension (perhaps related to increased salt and alcohol intake, reduced exercise and obesity). Journal of Human Hypertension

In conclusion, this urban population of Brazil shows a high prevalence of hypertension, with especially high rates in higher socioeconomic zones. Central fat distribution (increased WHR) is also a significant risk factor for hypertension in this population. A public health campaign to increase risk factor awareness, and efforts to alter diet and lifestyle practices of the community could reduce the threat posed by hypertension, and the future risk of cardiovascular diseases in this community.

References 1 Uemura K, Pisa Z. Trends in cardiovascular disease mortality in industrialised countries since 1950. World Health Stat Quart 1988; 41: 155–178. 2 Wielgosz AT. The decline in cardiovascular health in developing countries. World Health Stat Quart 1993; 46: 90. 3 Alberti G. Non communicable diseases: tomorrow’s pandemics. Bull World Health Org 2001; 79: 907. 4 INTERSALT Cooperative research group. Intersalt: an international study of electrolyte excretion and blood pressure. Results for 24 h urinary sodium and potassium excretion. Br Med J 1988; 297: 319–328. 5 Beilin LJ. Lifestyle and hypertension—an overview. Clin Exp Hypertens 1999; 21: 749–762. 6 Barker DJP, Bull AR, Osmond C, Simmonds SJ. Fetal and placental size and risk of hypertension in adult life. Br Med J 1990; 301: 259–262. 7 Mbewu A. Prevention is both moral and cost effective: Round table discussion. Bull World Health Org 2001; 79: 983. 8 Mosterd A et al. Trends in the prevalence of hypertension, antihypertensive therapy, and left ventricular hypertrophy from 1950 to 1989. N Engl J Med 1999; 340: 1221–1227. 9 Edwards R et al. Hypertension prevalence and care in an urban and rural area of Tanzania. J Hypertens 2000; 18: 145–152.

Affluence and hypertension in urban Brazil L Ala et al 779

10 Poulter NR et al. The Kenyan Luo migration study: observations on the initiation of a rise in blood pressure. Br Med J 1990; 300: 967–972. 11 Mbanya J, Minkoulou E, Salah J, Balkau B. The prevalence of hypertension in rural and urban Cameroon. Int J Epidemiol 1998; 27: 181–185. 12 Olatunbosun ST, Kaufman JS, Cooper RS, Bella AF. Hypertension in a black population: prevalence and biosocial determinants of high blood pressure in a group of urban Nigerians. J Human Hypertens 2000; 14: 249–257. 13 Olmos RD, Lotufo PA. Epidemiologia da hipertensa˜o arterial no Brasil e no mundo. Rev Brasileira Hipertens 2002; 9: 21–23. 14 Freitas OC et al. Prevalence of hypertension in the urban population of Catanduva, in the state of Sao Paulo, Brazil. Arquivos Brasileiros Cardiol 2001; 77: 16–21. 15 Lima-Costa MF et al. The Bambui health and aging study (BHAS): prevalence of risk factors and the use of preventive health care services. Revista Panamen˜a de Salud Publica 2001; 9: 219–227. 16 Chalmers J. 1999 World Health Organisation—International Society of Hypertension Guidelines for the Management of Hypertension. J Hypertens 1999; 17: 151–185. 17 Aubert L et al. Knowledge, attitudes and practices on hypertension in a country in epidemiological transition. Hypertension 1998; 31: 1136–1145. 18 Piccini RX, Victora CG. Systemic arterial hypertension in an urban area in southern Brazil: prevalence and risk factors. Rev Saude Publ 1994; 28: 261–267. 19 Raleigh VS. Diabetes and hypertension in Britain’s ethnic minorities: implications for the future of renal services. Br Med J 1997; 314: 209.

20 Steyn K, Kazenellenbogen JM, Lombard CJ, Bourne LT. Urbanisation and the risk for chronic diseases of lifestyle in the black population of the Cape Peninsula, South Africa. J Cardiovasc Risk 1997; 4: 135–142. 21 Murray CJL, Lopez AD. Global Comparative Assessments in Health Sector. World Health Organisation: Geneva, WHO, 1994. 22 Whelton PK. Primary prevention of hypertension: rational, approaches, realities and perspective. J Hum Hypertens 1996; 1(Suppl 8): s47–s50. 23 Dobbelsteyn CJ, Joffres MR, MacLean DR, Flowerdew G. A comparative evaluation of waist circumference, waist to hip ratio and body mass index as indicators of cardiovascular risk factors. The Canadian heart health surveys. Int J Obes Relat Metab Disord 2001; 25: 652–661. 24 Kaplan NM. The deadly quartet. Upper body obesity, glucose intolerance, hypertriglyceridemia and hypertension. Arch Int Med 1989; 149: 1514–1520. 25 Taylor RW, Jones IE, Williams SM, Goulding A. Evaluation of waist circumference, waist to hip ratio, and the conicity index as screening tools for high trunk fat mass, as measured by dual energy X ray absorptiometry in children aged 3–19 years. Am J Clin Nutr 2000; 72: 490–495. 26 Monteiro CA et al. The nutrition transition in Brazil. Eur J Clin Nutr 1995; 49: 105–113. 27 Sichieri R, Siqueira KS, Pereira RA, Ascherio A. Short stature and hypertension in the city of Rio de Janeiro, Brazil. Public Health Nutr 2000; 3: 77–82. 28 Marmot MG, Shipley MJ. Do socioeconomic differences in mortality persist after retirement? 25 year follow up of civil servants from the first Whitehall study. Br Med J 1996; 313: 1177–1180.

Journal of Human Hypertension