Prevalence, awareness, treatment, control and risk factors of hypertension in Korea: the Ansan study Inho Jo a , Younjhin Ahn a , Jungbok Lee b , Kyung Rim Shin c , Hong Kyu Lee a,d and Chol Shin b Objectives To determine prevalence, awareness, treatment, and control of hypertension, and its risk factors in an urban Korean population. Design and setting A cross-sectional survey in Ansan-city, Korea. Subjects and methods Population-based samples of people aged 18±92 years in Ansan-city, Korea, were selected, yielding 2278 men and 1948 women, and their blood pressures were measured using a highly standardized protocol. Hypertension was de®ned as a systolic BP > 140 mmHg or diastolic BP > 90 mmHg or reported treatment with antihypertensive medications, and subclassi®ed according to 1999 WHO-ISH guidelines. Isolated systolic hypertension (ISH) de®ned as a systolic BP > 140 mmHg and diastolic BP < 90 mmHg was also examined. Data were strati®ed by age and sex. Results The overall prevalence of hypertension in this study was 33.7%. Among these, 64.9% had Grade 1 hypertension, 22.5% Grade 2, and 12.5% Grade 3. Age- speci®c prevalence of hypertension increased progressively with age, from 14.19% in 18 to 24 year-olds to 71.39% in those 75 years or older. Hypertension prevalence was signi®cantly higher in men (41.5%) than in women (24.5%) (P < 0.001). Isolated systolic hypertension had signi®cantly lower prevalence (4.33%) within the population, although in the elderly aged 55 years or more it rose by 11.13%. Overall, 24.6% of hypertensive individuals were aware that they had high blood pressure, as much as 78.6% were being treated with antihypertensive medications, and 24.3% were under control. Hypertension awareness as well as treatment and control rates varied by sex, with women higher in all three rates. Multivariate analysis revealed that age, body mass index and abdomen circumference were signi®cantly associated with prevalence of hypertension both in men and women. Conclusions Hypertension is highly prevalent in Korea. Despite the high rate of treatment, the rates of awareness and control are relatively low, suggesting the nationwide demand for preventing and controlling high blood pressure in Korea in order to avert an epidemic of cardiovascular disease. J Hypertens 19:1523±1532 & 2001 Lippincott Williams & Wilkins. Journal of Hypertension 2001, 19:1523±1532 Keywords: awareness, control, hypertension, Korea, prevalence, risk factor, treatment a Division of Cardiovascular Research, Department of Biomedical Sciences, National Institute of Health, Seoul, Korea, b Ansan Health Center, Korea University Hospital, Ansan-city, Kyonggi-do, Korea, c College of Nursing Science, Ewha Womans University, Seoul, Korea, and d Division of Endocrinology and Metabolism, Department of Internal Medicine, College of Medicine, Seoul National University, Seoul, Korea. Sponsorship: This work was supported in part by National Institute of Health, Korea grant no. 334±6113±211±207±00 (I.J.) and by Korea University Institutes of Medical Science grant 2000±n6 (C.S.). Correspondence and requests for reprints to Chol Shin, M.D. F.C.C.P., Ph.D. Ansan Health Center, Korea University Hospital, 516 Gojan-dong, Ansan-city, Kyonggi-do, (425±707), Korea. Tel: 82 31 412 5602; fax: 82 31 412 5604; email: chol-shin@hanmail.net Received 28 December 2000 Revised 16 March 2001 Accepted 11 April 2001 Introduction The improved control of infectious and parasitic dis- eases and the sharp decrease in infant mortality during recent decades have dramatically changed the health pro®le of many developing countries [1]. This is re¯ected in an increase in life expectancy and the emergence of cardiovascular diseases (CVD) as a lead- ing cause of morbidity and mortality in such countries [2]. In fact, CVD including stroke are still the leading cause of all death in Korea [3], which accounted for 23.3% in 1999, although in 1983 it was responsible for as high as 27.8% of nation's mortality. Recognizing the rapid emergence of CVD as a clinical and public concern in Korea, the nationwide Korea Health and Nutrition Survey was initiated in 1998 to assess the prevalence of selected life-style-related dis- eases and their risk factors including hypertension. No of®cial results from Korea Health and Nutrition Survey have been, however, published to date. High blood pressure (BP) is one of the most important risk factors for CVD, and it has been shown that the reduction of highly or moderately elevated BP levels results in a decrease in stroke and myocardial infarction rates [4,5]. One of the cornerstones of the primary prevention of CVD since the early 1970s has focused on screening and early antihypertensive drug treatment Original article 1523 0263-6352 & 2001 Lippincott Williams & Wilkins of subjects presenting high BP [6]. Although commu- nity-based or national programs for detection and treat- ment of hypertension have been carried out in many countries [7±14], only a few studies of hypertension prevalence have been done in Korea [15±17]. Further- more, these studies were con®ned to analyses from 8 to 10-year-old surveys. In order to better assess the current status of hypertension screening and manage- ment, we recorded measurements of hypertension pre- valence, awareness, treatment and control in a randomized Korean population. Materials and methods Subjects The study area from which the participants were recruited is situated in a southwestern part of Seoul named Ansan-city, Kyonggi-do in South Korea, and is characterized by a newly industrializing town from a farming county. The city comprises 250 000 population aged more than 18 years. The target study sample was 4700 subjects aged more than 18 years who were randomly selected based upon address codes from June 1999 to April 2000. All were visited and interviewed by well-trained nursing students. A total of 427 subjects (9.09%) refused to participate in the survey. Further- more, we failed to collect suf®cient information regard- ing a detailed history and examination from 47 other subjects. Complete data for the analysis of hypertension and possible risk factors in this study were, therefore, available and based on the remaining 4226 subjects (2278 men, 1948 women). Questionnaires and blood pressure measurement The survey team consisting of nursing students was well trained in the details of questionnaires and meas- urements before starting the survey. In particular, all the surveyors were encouraged to read BP readings within a 2 unit error on their sequential measurements during practice. A team of two surveyors visited each home and interviewed the volunteer subjects. Only one person aged more than 18 years within the households was asked to participate in an interview. While one surveyor interviewed the study subject, the other asked another individual within the household for con®rma- tion of the data. Detailed information included sex, weight (portable scales: SECA, Hamburg, Germany; calibration monthly), height (steel ruler), abdomen circumference, alcohol intake, smoking, household in- come, occupation, marital status, years of education, driving, and BP. BP was measured according to World Health Organization-International Society of Hyper- tension (WHO-ISH) guidelines [18] using Bauman- ometer 1 (WA Baum Co. Inc. Copiague, New York, USA). Three BP readings were taken on either arm with 5 min intervals. Subjects were seated and caffeine substances and smoking were restricted for 30 min before the measurement. Data presented here are the average of three measurements. De®nitions Hypertension was de®ned as a systolic BP > 140 mmHg or diastolic BP > 90 mmHg according to WHO-ISH criteria [18] or a subject's currently receiv- ing antihypertensive treatments to control hyper- tension, regardless of the measured level. We further classi®ed the severity of hypertension as normal (systo- lic BP , 140 mmHg and diastolic BP , 90 mmHg), Grade 1 (140±159 for systolic BP/90±99 mmHg for diastolic BP), Grade 2 (160±179/100±109 mmHg), and Grade 3 (> 180/110 mmHg). Isolated systolic hyper- tension (ISH) was de®ned as a systolic BP > 140 mmHg and diastolic BP , 90 mmHg. Body mass index (BMI) was calculated from the simple equation (body weight in kg divided by height in m 2 ). Smoking was classi®ed as current, ex- (not for at least 1 year), and non-smoker. Alcohol intake was classi®ed as drin- ker and non-drinker from self-administered question- naire. Awareness of hypertension was de®ned as a subject who reported a history of previous diagnosis of hypertension or high BP. Treatment of hypertension was de®ned as a subject who currently uses prescribed antihypertensive drugs. Control of hypertension was de®ned as a subject who was receiving treatment for hypertension and kept BP normal (systolic BP , 140 mmHg and diastolic BP , 90 mmHg). Data analysis All the data of this study were summarized as mean and standard deviation for continuous variables and as frequency and percentage for categorical variables. Intergroup comparison for risk factors among subjects with or without hypertension was performed using a ÷ 2 test for distribution and t-test for continuous variables. The signi®cance of various risk factors was calculated by logistical regression analysis where the odds ratio and 95% con®dence intervals (CI) were calculated using multivariate analysis. All the analyses were con- ducted using the statistical software package SAS V6.12 [19] and a probability value of P , 0.05 was considered to be signi®cant. Results General characteristics The mean age of the study participants was 45.3 Æ 17.9 (mean Æ SD) years (range 18±92 years). The mean age (46.5 Æ 17.2) of the men was slightly, but statistically signi®cantly higher than that (44 Æ 18.6) of the women. All anthropometric data tested in this study was sig- ni®cantly higher in men than in women. A total of 21% of the population had attended at least high school. Generally, men had higher education levels than wo- men. The majority had a partner (75.1% of the men and 65.7% of the women) and occupation (77.9% of the 1524 Journal of Hypertension 2001, Vol 19 No 9 men and 74.2% of the women). Other detailed life style characteristics for the study population are shown in Table 1. Blood pressure BP distributions were approximately Gaussian, with a rightward skew (Fig. 1). Mean systolic and diastolic BP (mmHg) was 128.9 Æ 18.0 and 83.6 Æ 12.9 for men; and 121.1 Æ 18.1 and 77.3 Æ 12.6 for women, respectively (Table 1). Figure 2 displays the mean systolic and diastolic BP by age and sex. Mean systolic BP rose progressively across entire age range. The slope of the rise in systolic BP was bigger in women than in men. Although systolic BP in women was signi®cantly lower before the age of 54 (P , 0.01), it appeared that there was no signi®cant difference between 55 and 74 years of age (P . 0.05), and even a reversal was noted after 75 (P , 0.05). The pattern for mean diastolic BP was similar at a younger age (before the age of 45), but the average level of diastolic BP plateaued in men after the age of 45 and in women after the age of 55, respec- tively. Mean diastolic BP was almost the same in both men and women after the age of 55 years. Prevalence of hypertension Table 2 presents the age±sex-speci®c and age±sex- adjusted prevalence of hypertension. Overall, 33.7% of all studied population were hypertensive. Hypertension appeared to be signi®cantly more common in Korean men (41.53%) than women (24.54%) (P , 0.001). Hypertension prevalence increased progressively with age. In the youngest age group (18±24 years) hyper- tension was present in 14.19% of the population, whereas the prevalence rate was 71.39% in the oldest age group (75±92 years). National estimates of the hypertension prevalence were calculated using the available census data in 1995 [20], resulting in 35.8, 21.6 and 28.6% in age-adjusted men, women, and age± sex-adjusted total population of Korea, respectively (Table 2). Using subclassi®cation of BP according to WHO-ISH guidelines, 64.9% of hypertensive popu- lation excluding antihypertensive drug takers had Grade 1 BP; 22.5% Grade 2, 12.5% Grade 3, as shown in Table 3. Hypertension severity varied signi®cantly until the age of 54, showing increases in both Grade 2 and 3 BPs compared with decrease in Grade 1 BP. After the age of 55, however, the prevalence of each Grade BP remained relatively constant. Grade 1 BP reading was most prevalent (89.1%) between the age of 18±24 years. No signi®cant difference was found in the severity of sex-speci®c hypertension. Table 4 shows the age- and sex-speci®c prevalence of ISH. The overall prevalence of ISH in the studied population was 4.33%, and it signi®cantly (P , 0.001) increased with age. Furthermore, the prevalence of ISH rose signi®cantly from 1.36% in all population aged 18±54 years to Table 1 General characteristics of study population Men (n  2278) Women (n  1948) All (n  4226) Mean Æ SD Age (years) ÃÃà , y 46.5 Æ 17.2 44.0 Æ 18.6 45.3 Æ 17.9 SBP (mmHg) ÃÃà 128.9 Æ 18.0 121.1 Æ 18.1 125.3 Æ 18.4 DBP (mmHg) ÃÃà 83.6 Æ 12.9 77.3 Æ 12.6 80.7 Æ 13.1 Height (cm) ÃÃà 169.3 Æ 6.1 157.3 Æ 6.4 163.8 Æ 8.6 Weight (kg) ÃÃà 66.2 Æ 9.3 55.0 Æ 8.2 61.0 Æ 10.4 Abdomen circumference (cm) ÃÃà 86.9 Æ 10.6 82.2 Æ 13.4 84.7 Æ 12.2 BMI (kg/m 2 ) ÃÃà 23.1 Æ 2.8 22.3 Æ 3.3 22.7 Æ 3.1 Number of subjects (%) Years of education (%) à < 12 years 1650 (72.4) 1673 (85.9) 3323 (78.6) . 12 years 628 (27.6) 275 (14.1) 903 (21.4) Marital status (%) à Single 567 (24.9) 669 (34.3) 1236 (29.3) Couple 1711 (75.1) 1279 (65.7) 2990 (70.7) Occupation (%) à Yes 1775 (77.9) 1446 (74.2) 3221 (76.2) No 503 (22.1) 502 (25.8) 1005 (23.8) Household income (US$/month) (%) à , 800 824 (36.2) 600 (30.8) 1424 (34.0) 800±1600 908 (39.9) 797 (40.9) 1705 (40.7) 1600±2500 331 (14.5) 360 (18.4) 691 (16.5) > 2500 203 (8.9) 164 (8.4) 367 (8.8) Alcohol intake (%) à Yes 1490 (65.5) 648 (33.4) 2138 (50.8) No 783 (34.5) 1290 (66.6) 2073 (49.2) Smoking status (%) à Current smoker 1308 (57.5) 116 (2.7) 1424 (33.7) Ex-smoker 361 (15.8) 16 (0.4) 377 (8.9) Nonsmoker 609 (26.7) 1816 (43.0) 2425 (57.4) SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index. y Means of variables were signi®cantly different between men and women, ÃÃà P , 0.001. Distributions of subjects were signi®cantly different among variable categories and between gender, à P , 0.05. Hypertension in Korea Jo et al. 1525 11.13% in those aged 55±92 years. In the elderly (after the age of 55), the prevalence of ISH in men and women was 10.56 and 11.92%, respectively, showing that ISH appeared to be more common in women than in men in this age group, however, no gender differ- ence in the prevalence of ISH was found (P  0.512). Hypertension awareness, treatment, and control Table 5 showed that approximately a quarter (24.6%) of the studied population who could be classi®ed as hypertensive, were aware of their high BP. Rates of awareness were higher in women (33.5%) than men (20.1%) (P , 0.001). The estimated percentage of hypertensive individuals receiving pharmacological treatment in Korea was 78.6%. A higher proportion of women (85.0%) were being treated than men (73.2%) (P , 0.01). The estimated percentage of hypertensive individuals whose BPs were under control (systolic BP , 140 mmHg and diastolic BP , 90 mmHg) was 24.3%. Better control rates were noted in women (27.9%) than men (20.9%), but no signi®cant difference was found (P  0.354). Risk factors associated with hypertension The demographic, anthropometric and lifestyle charac- teristics of subjects with or without hypertension were presented in Table 6. The prevalence of hypertension in both genders was associated with greater age (P , 0.001). Hypertension was also signi®cantly asso- ciated with higher body mass index (BMI), greater abdomen circumference, years of education, marital status, occupation, driving, smoking, and lower house- hold income in both genders. Alcohol intake was associated with hypertension in women (P , 0.01), but not in men. When age-adjusted, BMI and abdomen circumference in men had a strong dose±response relationship over their entire ranges as shown in Table 7. In women, however, these relationships were only true at BMI > 26 kg/m 2 and abdomen circumference . 95 cm. Hypertension was signi®cantly less prevalent in those who earned the highest incomes in both genders (P , 0.01). Interestingly, alcohol intake was associated Fig. 1 Bar graphs showing distribution of (a) systolic and (b) diastolic blood pressures in the Ansan adult population. 0 200 400 600 800 1000 1200 Number of subjects Ͻ60 60Ϫ 70Ϫ 80Ϫ 90Ϫ 100Ϫ 110Ϫ 120Ϫ 130Ϫ 140Ϫ 150Ϫ 160Ϫ 170Ϫ 180Ϫ 190Ϫ 200Ϫ SBP (mmHg) (a) 0 200 400 600 800 1000 1200 1400 1600 Number of subjects Ͻ40 40Ϫ 50Ϫ 60Ϫ 70Ϫ 80Ϫ 90Ϫ 100Ϫ 110Ϫ 120Ϫ 130Ϫ 140Ϫ DBP (mmHg) (b) SBP-M SBP-F DBP-M DBP-F 50 70 90 110 130 150 BP (mmHg) 18–24 25–34 35–44 45–54 55–64 65–74 75–92 Age group (years) c c bc c b c a b a a a a a a d d d d d d c c b b a a a a * *** *** *** *** *** *** *** *** Fig. 2 Line graph shows systolic blood pressure (SBP) and diastolic blood pressure (DBP) by age and sex (M, male; F, female) among the Ansan adult population. Different letters means signi®cantly different group at P  0.0001 by analysis of variance (ANOVA) test. à P , 0.01; ÃÃà P , 0.001) in SBP and DBP between men and women of same age group. Table 2 Prevalence of hypertension by age and sex group (%) Age group Men Women All 18±24 25.0 6.2 14.2 25±34 30.1 6.7 18.3 35±44 34.8 9.6 22.2 45±54 42.6 28.2 36.9 55±64 51.5 57.1 51.8 65±74 62.0 59.0 59.4 75±92 66.2 76.6 71.4 Crude prevalence ÃÃà , y 41.5 24.5 33.7 National estimate 35.8 21.6 28.6 § Hypertension, diastolic blood pressure > 90 mmHg or systolic blood pressure > 140 mmHg. y Crude prevalence was signi®cantly different between men and women, ÃÃà P , 0.001. § Age-sex-adjusted hypertension prevalence by the total population of Korea in 1995. 1526 Journal of Hypertension 2001, Vol 19 No 9 with hypertension only in men, not in women, after age adjustment (Table 7). In comparison, as shown in Table 6, a signi®cant relationship with alcohol was observed only in women, not in men, without age adjustment. No signi®cant relationships with all other lifestyles were observed in both genders after age adjustment (Table 7). Multivariate logistic regression analysis showed that regardless of age and sex, in men, age, BMI, abdomen circumference and alcohol intake were independent risk factors and showed strong association with hypertension (Table 8). In women, however, alcohol intake lost its signi®cance. Only weak, but still signi®cant association was observed with BMI in women (P  0.07). No independent associations with hypertension were found with years of education, occupation, household income and smoking in both genders. Discussion Our results showed that the overall prevalence of hypertension, estimated to be 33.7%, exceeds those of many countries [8,21±24]. This is especially noteworthy because this study included a comparatively younger ( > 18 years old) age group than other studies. In many Table 3 The severity of hypertension by age and sex according to World Health Organization-International Society of Hypertension (WHO- ISH) criteria (%) Age categories 18±24 25±34 35±44 45±54 55±65 65±74 75±92 Total Hypertension 65 161 250 176 218 312 242 1424 Severity Grade 1 (mild) 57 (89.1) 121 (76.1) 168 (70.0) 89 (58.9) 95 (59.4) 127 (58.3) 89 (56.7) 46 (64.9) Grade 2 (moderate) 6 (9.4) 26 (16.4) 45 (18.8) 39 (25.8) 41 (25.6) 58 (26.6) 44 (28.0) 259 (22.5) Grade 3 (severe) 1 (1.6) 12 (7.5) 27 (11.3) 23 (15.2) 24 (15.0) 33 (15.1) 24 (15.3) 144 (12.5) Men 48 132 191 111 122 139 64 807 Grade 1 (mild) 41 (85.4) 100 (75.8) 130 (68.1) 63 (56.8) 74 (60.7) 78 (56.1) 38 (59.4) 524 (64.9) Grade 2 (moderate) 6 (12.5) 22 (16.4) 36 (18.8) 27 (24.3) 31 (25.4) 40 (28.8) 15 (23.4) 177 (21.9) Grade 3 (severe) 1 (2.1) 10 (7.6) 25 (13.1) 21 (18.9) 17 (13.9) 21 (15.1) 11 (17.2) 106 (13.1) Women 16 27 49 40 38 79 93 342 Grade 1 (mild) 16 (100) 21 (77.8) 38 (77.6) 26 (65.0) 21 (55.3) 49 (62.0) 51 (54.8) 222 (64.9) Grade 2 (moderate) 0 4 (14.8) 9 (18.4) 12 (30.0) 10 (26.3) 18 (22.8) 29 (31.2) 82 (24.0) Grade 3 (severe) 0 2 (7.4) 2 (4.0) 2 (5.0) 7 (18.4) 12 (15.2) 13 (14.0) 38 (11.1) Subtotal y 64 159 240 151 160 218 157 1149 Use antihypertensive drugs 1 2 10 25 58 94 85 275 Grade 1 blood pressure (BP) indicates 140±159 mmHg systolic blood pressure (SBP) or 90±99 mmHg diastolic blood pressure (DBP); Grade 2, 160±179 mmHg SBP or 100±109 mmHg DBP; Grade 3, > 180 mmHg SBP or > 110 mmHg DBP. P , 0.001 for all differences. y Excludes those reporting treatment with antihypertensive drugs. Table 4. Prevalence of isolated systolic hypertension by age and sex group (%) Age group Men Women All 18±24 2.00 0.78 1.31 25±34 2.01 2.05 0.46 0.63 1.25 1.36 35±44 1.23 0.18 0.71 45±54 3.72 2.21 3.14 55±64 5.18 7.14 5.70 65±74 14.75 10.56 y 9.09 11.92 y 12.4 11.13 75±92 13.25 18.1 15.9 Crude prevalence § 4.92 3.64 4.33 Isolated systolic hypertension (ISH), diastolic blood pressure (DBP) , 90 mmHg and SBP > 140 mmHg. y No gender difference in the prevalence of ISH was found in population aged more than 55 years (P  0.512). § Crude prevalence was different between men and women, but not signi®cant (P , 0.051). Table 5 Hypertension prevalence, awareness, treatment and control in men and women Men Women All P Prevalence 946/2278 (41.5%) 478/1948 (24.5%) 1424/4226 (33.7%) , 0.001 Awareness 190/946 (20.1%) 160/478 (33.5%) 350/1424 (24.6%) , 0.001 Treatment 139/190 (73.2%) 136/160 (85.0%) 275/350 (78.6%) , 0.01 Control 29/139 (20.9%) 38/136 (27.9%) 67/275 (24.3%) NS Awareness, the number of hypertensive persons who were diagnosed before; Treatment, the number of hypertensive persons who uses anti-hypertension drugs; Control, the number of persons who keep blood pressure normal (diastolic blood pressure , 90 mmHg and systolic blood pressure , 140 mmHg) among treatments. NS, not signi®cant (P  0.354). Hypertension in Korea Jo et al. 1527 other countries, however, the prevalence of hyper- tension was also reported to be higher than that in the present data [25±27]. These differences may re¯ect the effects of dynamic interactions between genetic, socio- cultural, demographic and economic factors. From this point of view, each country's own data should be of great importance to later develop a hypertension sur- veillance system at the global level. The prevalence of hypertension shown in this study was comparable to those from National Health and Nutrition Examination Survey (NHANES) I [7], western India [9], and south- eastern Spain [28]. In Korea, however, earlier surveys reported signi®cantly lower prevalence [15±17], sug- gesting an upward trend in the prevalence of hyper- tension for 1990±1992 in Korea. A similar trend was observed in the prevalence of hypertension in China during 1980±1991 [23]. In comparison, a signi®cant downward trend in mean systolic BP and in the prevalence of hypertension was detected in US [7] during 1960±1991 and Finland [26] during 1982±1997. The differences in trends among these are of great importance, but yet to be fully understood. One possible explanation is that highly developed countries have already provided good surveillance systems to control and prevent high BP, while Korea is only currently initiating one. An interesting ®nding of the present study is that the prevalence of hypertension in men is signi®cantly (P , 0.001) higher than in women, as shown in Table Table 6 Prevalence of risk factors of hypertension in men and women à Men Women Hypertensive Normotensive Hypertensive Normotensive Risk factors (n  946) (n  1332) (n  478) (n  1470) Age (years) 51.7 Æ 17.4 42.8 Æ 16.1 62.2 Æ 17.6 38.1 Æ 14.6 Age group 18±24 4 (24.5%) 151 16 (6.2%) 242 25±34 134 (29.9%) 314 27 (6.2%) 407 35±44 197 (34.6%) 372 53 (9.5%) 502 45±54 126 (42.6%) 170 50 (27.6%) 131 55±64 157 (50.8%) 152 61 (54.5%) 51 65±74 185 (60.7%) 120 127 (57.7%) 93 75±92 98 (64.9%) 53 144 (76.6%) 44 BMI (kg/m 2 ) 23.8 Æ 2.86 22.6 Æ 2.72 23.5 Æ 3.72 21.9 Æ 3.05 BMI group , 20 81 (27.1%) 218 82 (16.1%) 428 20±22 179 (32.7%) 369 95 (19.0%) 404 22±24 246 (40.9%) 355 112 (25.1%) 334 24±26 241 (47.4%) 268 74 (30.0%) 173 26±28 130 (61.3%) 82 56 (42.4%) 76 > 28 69 (63.3%) 40 59 (51.8%) 55 Abdomen circumference (cm) 90.6 Æ 8.78 84.9 Æ 8.40 91.1 Æ 11.9 80.1 Æ 10.2 Years of education y < 12 years 715 (43.3%) 935 452 (27.0%) 1221 . 12 years 231 (36.8%) 397 26 (9.45%) 249 Marital status Single 753 (44.0%) 958 240 (18.8%) 1039 Couple 193 (34.0%) 374 238 (35.6%) 431 Occupation With 675 (38.0%) 1100 187 (12.9%) 1259 Without 271 (53.9%) 232 291 (58.0%) 211 Driving Yes 508 (37.1%) 862 43 (10.9%) 351 No 438 (48.2%) 470 435 (28.0%) 1119 Alcohol intakeé6 Yes 630 (42.3%) 860 111 (17.1%) 537 No 316 (40.1%) 472 367 (28.2%) 933 Smoking status Current smoker 518 (39.6%) 790 44 (37.9%) 72 Ex-smoker 177 (49.0%) 184 5 (31.3%) 11 Nonsmoker 251 (41.2%) 358 429 (23.6%) 1387 Household income (US$/ month) , 850 398 (47.6%) 438 252 (40.2%) 375 850±1700 347 (38.2%) 561 165 (20.7%) 632 1700±2550 127 (38.4%) 204 50 (13.9%) 310 > 2550 [1] 74 (36.5%) 129 11 (6.7%) 153 à Distributions of hypertensives and normotensives in both genders were signi®cantly different (P , 0.001) by all variables except years of education and alcohol intake. y Men, P , 0.05; women, P , 0.01. § Men, not signi®cant; women, P , 0.01. 1528 Journal of Hypertension 2001, Vol 19 No 9 2, which contributed to high overall prevalence. It is interesting, but not totally new, since Jee et al. [16] have already reported that the prevalence of hypertension was 28.9% in Korean men and 15.9% in Korean women. Furthermore, an interesting result when working on a Bulgarian population study, showed a signi®cantly high- er prevalence of age-adjusted hypertension in men (59.1%) than that in women (36.2%) (P , 0.001) [14]. The reasons for these large gender differences resulting from the present study have not been determined with certainty to date. One possible explanation is that the men had a higher BMI and greater abdomen circumfer- ence as shown in Table 1, which can contribute to higher prevalence of hypertension. Despite the signi®- cant gender difference of overall hypertension preva- lence, the general pattern of an increasing average systolic BP has been recognized as in previous studies (Fig. 2) [8,14]. Also, the pattern of the age-related increase in diastolic BP followed by a plateau from middle-age has been the norm in most surveys, [14]. The prevalence of hypertension classi®ed by severity was dramatically changed by age (Table 3). For exam- ple, between 65 and 74 years of age, 56.1% of men had Grade 1; 28.8% Grade 2; 15.1% Grade 3, whereas 62.0% of women had Grade 1; 22.8% Grade 2; 15.2% Table 7 Odds ratio of various risk factors for hypertension in men and women Men Women Risk factors Odds ratio (CI) Age-adjusted odds ratio (CI) Odds ratio (CI) Age-adjusted odds ratio (CI) Age (years) 18±24 1 1 25±34 1.32 (0.90±1.92) 1.00 (0.53±1.90) 35±44 1.63 (1.13±2.35) ÃÃà , } 1.60 (0.89±2.85) 45±54 2.28 (1.54±3.39) ÃÃà 5.78 (3.16±10.5) ÃÃà 55±64 3.18 (2.15±4.71) ÃÃà 18.1 (9.66±33.9) ÃÃà 65±74 4.75 (3.20±7.06) ÃÃà 20.6 (11.7±36.6) ÃÃà 75±84 5.70 (3.58±9.06) ÃÃà ± 49.5 (26.9±90.9) ÃÃà ± BMI (kg/m) , 20 1111 20±22 1.31 (0.96±1.78) 1.61 (1.16±2.23) Ãà 1.23 (0.89±1.70) 1.12 (0.75±1.66) 22±24 1.87 (1.38±2.52) ÃÃà 2.15 (1.57±2.95) ÃÃà 1.75 (1.27±2.41) ÃÃà 1.38 (0.94±2.03) 24±26 2.42 (1.78±3.30) ÃÃà 2.75 (1.99±3.80) ÃÃà 2.23 (1.56±3.20) ÃÃà 1.26 (0.82±1.94) 26±28 4.27 (2.93±6.22) ÃÃà 4.98 (3.37±7.37) ÃÃà 3.85 (2.53±5.84) ÃÃà 1.87 (1.13±3.09) Ãà > 28 4.64 (2.92±7.40) ÃÃà 5.32 (3.28±8.61) ÃÃà 5.60 (3.62±8.67) ÃÃà 2.35 (1.40±3.92) ÃÃà Abdomen circumference (cm) < 75 1111 75±80 1.86 (1.22±2.84) ÃÃà 1.62 (1.05±2.49) à 2.13 (1.40±3.26) ÃÃà 1.37 (0.86±2.17) 80±85 2.24 (1.51±3.32) ÃÃà 1.80 (1.20±2.68) Ãà 3.18 (2.13±4.74) ÃÃà 1.52 (0.97±2.37) 85±90 2.84 (1.93±4.19) ÃÃà 2.14 (1.44±3.19) ÃÃà 3.99 (2.68±5.94) ÃÃà 1.12 (0.70±1.78) 90±95 5.84 (3.94±8.67) ÃÃà 4.04 (2.69±6.08) ÃÃà 6.27 (4.12±9.55) ÃÃà 1.51 (0.92±2.47) 95±100 7.39 (4.83±11.3) ÃÃà 5.01 (3.23±7.77) ÃÃà 14.8 (9.50±23.1) ÃÃà 2.72 (1.61±4.59) ÃÃà . 100 10.6 (6.62±17.0) ÃÃà 6.90 (4.24±11.2) ÃÃà 21.3 (13.7±33.2) ÃÃà 3.12 (1.85±5.28) ÃÃà Years of education < 12 years 1111 . 12 years 0.76 (0.63±0.92) Ãà 1.02 (0.83±1.24) 0.28 (0.19±0.43) ÃÃà 0.82 (0.52±1.30) Marital status Single 1111 Couple 0.66 (0.54±0.80) ÃÃà 0.98 (0.79±1.22) 2.39 (1.94±2.95) ÃÃà 1.24 (0.93±1.64) Occupation With 1111 Without 1.94 (1.56±2.33) ÃÃà 1.00 (0.79±1.28) 9.28 (7.34±11.7) ÃÃà 1.56 (1.08±2.25) Ãà Driving Yes 1111 No 1.58 (1.33±1.88) ÃÃà 1.01 (0.83± 1.23) 3.17 (2.27±4.44) ÃÃà 1.31 (0.90±1.90) Household income ($US/month) , 850 1111 850±1700 0.39 (0.31±0.49) ÃÃà 0.85 (0.63±1.14) 0.39 (0.31±0.49) ÃÃà 0.84 (0.63±1.14) 1700±2550 0.24 (1.71±0.34) ÃÃà 0.92 (0.62±1.38) 0.24 (0.17±0.34) ÃÃà 0.91 (0.61±1.36) > 2550 0.11 (0.06±0.20) ÃÃà 0.42 (0.21±0.82) Ãà 0.11 (0.06±0.20) ÃÃà 0.41 (0.21±0.81) Ãà Alcohol intake No 1111 Yes 1.09 (0.92±1.30) 1.34 (1.11±1.61) Ãà 0.52 (0.41±0.67) ÃÃà 1.23 (0.92±1.66) Smoking status Nonsmoker 1111 Current smoker 0.94 (0.77±1.14) 0.96 (0.78±1.18) 1.98 (1.34±2.92) ÃÃà 0.98 (0.59±1.62) Ex-smoker 1.37 (0.06±1.78) à 1.05 (0.80±1.39) 1.47 (0.51±4.25) 0.44 (0.12±1.52) CI, con®dence interval; BMI, body mass index. §,à P , 0.05, Ãà P , 0.01, ÃÃà P , 0.001. Hypertension in Korea Jo et al. 1529 Grade 3. To what extent well-established risk factors for hypertension, such as BMI, years of education, occupation, alcohol intake and smoking, account for the observed age-speci®c variation in hypertension severity is an important issue worthy of further investigation. As shown in Table 4, ISH was not common (1.36%) in the population aged , 54 years, while its prevalence greatly increased (11.13%) in the elderly (. 55 years). This result supported again that ISH is the most common form of hypertension in the elderly [29,30]. The prevalence of ISH in the elderly Korean could not be directly compared with other previous ®ndings, since each study employed different de®nitions (diasto- lic and systolic cut-off points) and age distribution of the population studied. Several other factors besides de®nitions and age group studied were gender, social class, the number of blood pressure readings performed, geographical location, and treatment status [30,31]. Because of these limitations, the prevalence of ISH varied from study to study; ranged from 1% in Israel to 24% in Norway when the criterion of ISH de®ned as a systolic BP > 160 mmHg and diastolic BP , 90 or 95 mmHg was used [31]. When the lowest limit of systolic BP was adjusted as 140 mmHg, the prevalence of ISH in the elderly Korean was comparable to that from the 1992±1993 Mexican study [32]. Furthermore, it was also found to be signi®cantly lower than those in the studies by Multiple Risk Factor Intervention Trial Research Group [33], Psaty et al. [34], and Kocemba et al. [35]. Despite signi®cantly lower prevalence from this study, however, it is likely that ISH will become a truly major medical and social problem since popula- tions in Korea are becoming older. Insigni®cant or less gender difference in the prevalence of ISH in the elderly Korean may re¯ect the fact that ISH is primarily a characteristic of older ages at a time when the male and female systolic pressure difference is minimal or reversed [36]. Although 33.7% of the total population were hyper- tensive, only one-quarter (24.6%) of hypertensive in- dividuals were aware of their condition (Table 5). This low rate of awareness is somewhat unexpected, since Korean National Health Insurance Policy has provided most Koreans with medical examinations including BP measurement, every other year. This suggested, in part, delinquent measurement of hypertension during exam- ination. As much as 65% awareness was reported from the population-based study in Barbados [37]. A recent study in Austria advised improved strategies to achieve better community control of hypertension since they found that the intensive blood pressure education campaign had only a temporary effect on improving blood pressure awareness [38]. Despite the low rate of awareness in Korea, the diag- nosis of hypertension led to medical treatment in about three-quarters (78.6%) of patients. This is not surpris- ing, since hypertensive patients were within easy access of inexpensive antihypertensive medication in Korea. Contrary to treatment, the control of hypertension by antihypertensive drug therapy was rather low, with only one-quarter (24.3%) of treated patients achieving an adequate BP control. One possible explanation might be a lack of aggressiveness in treating the patients [39]. Another accepted reason for not achieving goal BP is poor patient compliance with the antihypertensive medication. According to many studies, only 50±60% of patients adhere well to the prescribed medication [40,41]. It has been suggested that interventions fo- cused on better patient compliance should be persona- lized and multifactorial, but no detailed studies are so far available [42]. The other possible reason may be the side-effects of the antihypertensive drugs currently available. The low ef®cacy of antihypertensive treat- ment in the community raises an important question of its cost-effectiveness. Indeed, an ineffective treatment will greatly increase the costs induced by the diseases related to hypertension (stroke and coronary heart diseases). The impact on health expenditure of the failure to adequately control BP by drug treatment should thus be considerable. A signi®cantly higher rate of awareness, treatment and control was prominent in women as shown in Table 5. Since men usually could afford to have systematic clinical examinations, it was assumed that the levels of hypertension awareness, treatment and control rates were relatively higher in men than in women. Interest- ingly, however, the results from the present study showed that women achieved better awareness, treat- ment and control than men. The reason for this result is yet to be fully understood. Another interesting ®nd- ing in the present study is the signi®cant deviation from classical `rule of halves', which is currently con- troversial, to be applied [43]. Table 8 Multivariate logistic regression analysis for association of various risk factors with prevalence of hypertension in Ansan-city, Korea Men Women Odds ratio (CI) Odds ratio (CI) Age 1.26 (1.17±1.37) ÃÃà ,§ 1.81 (1.62±2.03) ÃÃà BMI 1.18 (1.08±1.30) Ãà 1.09 (0.98±1.21) Abdomen circumference 1.27 (1.17±1.37) ÃÃà 1.13 (1.03±1.23) à Years of education 0.97 (0.78±1.21) 0.97 (0.61±1.55) Marital status 1.16 (0.90±1.48) 1.15 (0.84±1.58) Occupation 1.08 (0.84±1.48) 1.34 (0.89±2.01) Household income 1.02 (0.91±1.13) 0.88 (0.74±1.03) Driving 1.20 (0.96±1.50) 1.16 (0.79±1.70) Alcohol intake 1.32 (1.08±1.61) Ãà 1.23 (0.91±1.66) Smoking status 0.94 (0.85±1.05) 0.98 (0.76±1.26) CI, con®dence interval; BMI, body mass index. §,à P , 0.05, Ãà P , 0.01, ÃÃà P , 0.001. 1530 Journal of Hypertension 2001, Vol 19 No 9 Studies from developing and developed countries have shown that obesity [9,44±46], smoking [9,46], alcohol intake [18,46], and years of education [47±49] are associated with the prevalence of hypertension. The present data are in accord with an association of smoking, obesity and less years of education with hypertension in both genders, but an association of alcohol intake only in women (Table 6). However, age- adjustment nulli®ed the signi®cant associations of years of education, marital status, driving, and smoking with hypertension (Table 7). It is dif®cult to understand why our data showed no signi®cant association between smoking and age-adjusted hypertension in both gen- ders. Since smoking, a well-known risk factor for hyper- tension [9,46], is relatively popular in Korea, more detailed studies should be awaited for con®rmation of this data. Both non-working women and alcohol con- suming males had signi®cant positive associations with age-adjusted hypertension in this study (P , 0.01). The signi®cant, but negative association of highest house- hold income (> US$2550/month) with age-adjusted prevalence of hypertension in both genders re¯ects that individuals from high socio-economic status are likely to prevent or control hypertension. This ®nding is consistent with observations made in developed coun- tries [50,51]. The association of diabetes mellitus with hypertension is well known [46], and should be consid- ered as a risk factor in the study. However, only 14 individuals from the study population (n  4226) re- ported their history in self-administered questionnaires, and therefore may be considerably underdetected. Therefore we did not analyse this variable in the current study. Obese persons are prone to develop hypertension and the present data showing the strong dose±response relationship between BMI and hyper- tension con®rm this association. Average BMI index (23.8 kg/m 2 for men; 23.5 kg/m 2 for women) observed in hypertensives in the present study is usually consid- ered within desirable limits in developed countries [52], therefore, lifestyle guidelines for the hypertension prevention in Korea should be different from those developed countries. The limit of BMI index (22.6 kg/ m 2 for men; 21.9 kg/m 2 for women) observed in normo- tensives in this study may be given due consideration to de®ne safe limits for the risk factor for the preven- tion of hypertension in Korean. Although many risk factors are to be considered, our data clearly indicated that BMI and abdomen circumference are independent and preventable predictors of hypertension in Koreans, although slightly less signi®cant in women (Table 8). Therefore, it is widely recognized that the future programmes for the primary prevention of hypertension should concentrate on obesity in Korea. In summary, the results from the present Ansan study show a higher prevalence of hypertension than has been reported in previous Korean studies. Our ®ndings also suggest strong associations between hypertension and BMI and abdomen circumference. There was no signi®cant relationship between hypertension and smoking and this remains to be further studied. The extremely high prevalence of hypertension coupled with the disturbingly low prevalence rates of awareness and control has important implications for health care providers, public health of®cials, and health policy- makers in Korea. This study identi®es an urgent need for nationwide efforts to prevent and control hyper- tension, in order to avert an epidemic of BP-related atherosclerotic cardiovascular diseases. Acknowledgements We thank Drs Namhan Cho and Sangmee Ahn Jo for their critical readings, Ms Sun Mi Lee for project coordination, and Ms Jooyoung Lee for secretarial assistance. The opinions expressed in this article are those of the authors and do not necessarily imply endorsement by his or her employer or the funding agency. References 1 Jamison DT, Mosley WH, Measham AR, Bobadilla JLL. (editors) Disease control priorities in developing countries. Oxford, UK: Oxford University Press; 1993. 2 Whelton PK, Brancati FL, Appel LJ, Klag MJ. The challenge of hyper- tension and atherosclerotic cardiovascular disease in economically devel- oping countries. High Blood Pressure and Cardiovascular Prevention 1995; 4:36±45. 3 Korean National Statistical Of®ce (KNSO). 1999 Annual report on the cause of death statistics. Daejeon, Korea: KNSO; 2000. 4 Neaton JD, Grimm RH Jr, Prineas RJ, Stamler J, Grandits GA, Elmer PJ, et al. Treatment of mild hypertension study. Final results. Treatment of mild hypertension Study Research Group. JAMA 1993; 270:713±724. 5 Dollery C. Hypertension trial results: consensus and con¯icts. J Hum Hypertens 1995; 9:403±408. 6 WHO Technical Report series. Prevention of coronary heart disease. Report of a WHO Expert Committee. Geneva: World Health Organiza- tion; 1982. 7 Burt VL, Cutler JA, Higgins M, Horan MJ, Labarthe D, Whelton P, et al. Trends in the prevalence, awareness, treatment, and control of hyper- tension in the adult US population: data from the health examination surveys, 1960 to 1991. Hypertension 1995; 26:60±69. 8 Ibrahim MM, Rizk H, Appel LJ, el Aroussy WE, Helmy S, Sharaf Y, et al. Hypertension prevalence, awareness, treatment and control in Egypt: Results from the Egyptian National Hypertension Project (NHP). NHP Investigative Team. Hypertension 1995; 26:886±890. 9 Gupta R, Guptha S, Gupta VP, Prakash H. Prevalence and determinants of hypertension in the urban population of Jaipur in western India. J Hypertens 1995; 13:1193±1200. 10 Colhoun HM, Dong W, Poulter NR. Blood pressure screening, manage- ment and control in England: results from the health survey for England 1994. J Hypertens 1998; 16:747±752. 11 Stergiou GS, Thomopoulou GC, Skeva Il, Mountokalakis TD. Prevalence, awareness, treatment, and control of hypertension in Greece: the Didima study. Am J Hypertens 1999; 12:959±965. 12 Du X. Prevalence, treatment, control, and awareness of high blood pressure and the risk of stroke in northwest England. Prev Med 2000; 30:288±294. 13 Fuentes R, Ilmaniemi N, Laurikainen E, Tuomilehto J, Nissinen A. Hyper- tension in developing economies: a review of population-based studies carried out from 1980 to 1998. J Hypertens 2000; 18:521±529. 14 Stein AD, Stoyanovsky V, Mincheva V, Dimitrov E, Hodjeva D, Petkov A, et al. Prevalence, awareness, treatment and control of hypertension in a working Bulgarian population. Eur J Epidemiol 2000; 16:265±270. 15 Kim JS, Jones DW, Kim SJ, Hong YP. Hypertension in Korea: a national survey. Am J Prev Med 1994; 10:200±204. 16 Jee SH, Appel LJ, Suh I, Whelton PK, Kim IS. Prevalence of cardio- Hypertension in Korea Jo et al. 1531 vascular risk factors in South Korean adults: results from the Korea Medical Insurance Corporation (KMIC) Study. Ann Epidemiol 1998; 8:14±21. 17 Jones DW, Kim JS, Kim SJ, Hong YP. Hypertension awareness, treatment and control rates for an Asian population: results from a national survey in Korea. Ethn Health 1996; 1:269±273. 18 Chalmers J, MacMahon S, Mancia G, Whitworth J, Beilin L, Hansson L, et al. 1999 World Health Organization-International Society of Hypertension Guidelines for the management of hypertension. Guidelines sub-commit- tee of the World Health Organization. Clin Exp Hypertens 1999; 21:1009±1060. 19 SAS Institute Inc. SAS/SAGT User's Guide, 6.12 edition. Cary, NC, USA: SAS Institute Inc; 1996. 20 Korean National Statistical Of®ce (KNSO). 1995 Population and housing census report, Daejeon, Korea: KNSO; 1998. 21 Burt VL, Whelton P, Roccella EJ, Brown C, Cutler JA, Higgins M, et al. Prevalence of hypertension in the US adult population. Results from the third national health and nutrition examination survey, 1988±1991. Hypertension 1995; 25:305±313. 22 van Leer EM, Seidell JC, Kromhout D. Levels and trends in blood pressure and prevalence and treatment of hypertension in the Nether- lands, 1987±1991. Am J Prev Med 1994; 10:194±199. 23 Wu X, Duan X, Gu D, Hao J, Tao S, Fan D. Prevalence of hypertension and its trends in Chinese populations. Int J Cardiol 1995; 52:39±44. 24 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. 25 Ordunez-Garcia PO, Espinosa-Brito AD, Cooper RS, Kaufman JS, Nieto J. Hypertension in Cuba: evidence of narrow black-white difference. J Hum Hypertens 1998; 12:111±116. 26 Kastarinen M, Salomaa V, Vartiainen E, Jousilahti P, Tuomilehto J, Puska P, et al. Trends in blood pressure levels and control of hypertension in Finland from 1982 to 1997. J Hypertens 1998; 16:1379±1387. 27 Baba S, Pan WH, Ueshima H, Ozawa H, Komachi Y, Stamler R, et al. Blood pressure levels, related factors, and hypertension control status of Japanese and Americans. J Hum Hypertens 1991; 5:317±32. 28 Puras A, Sanchis C, Artigao LM, Divison JA. Prevalence, awareness, treatment and control of hypertension in Spanish population. Eur J Epidemiol 1998; 14:31±36. 29 Black HR. Isolated systolic hypertension in the elderly: lessons from clinical trials and future directions. J Hypertens 1999; 17:S49±S54. 30 Staessen J, Amery A, Fagard R. Isolated systolic hypertension in the elderly. J Hypertens 1990; 8:393±405. 31 Amery A, Fagard R, Guo C, Staessen J, Thus L. Isolated systolic hypertension in the elderly: an epidemiologic review. Am J Med 1991; 90:64S±70S. 32 Arroyo P, Fernandez V, Avila±Rosas H. Overweight and hypertension: data from the 1992±1993 Mexican survey. Hypertension 1997; 30:646±649. 33 Multiple risk factor intervention trial. Risk factor changes and mortality results. Multiple Risk Factor Intervention Trial Research Group. JAMA 1982; 248:1465±1477. 34 Psaty BM, Furberg CD, Kuller LH, Borhani NO, Rautaharju PM, O'Leary DH, et al. Isolated systolic hypertension and subclinical cardiovascular disease in the elderly. Initial ®ndings from the Cardiovascular Health Study. JAMA 1992; 268:1287±1291. 35 Kocemba J, Gryglewska B, Klich A, Grodzicki T. Distribution of blood pressure and prevalence of arterial hypertension in the old population of Cracow. Mater Med Pol 1991; 23:33±35. 36 Dustan HP. Gender differences in hypertension. J Hum Hypertens 1996; 10:337±340. 37 Freeman V, Fraser H, Forrester T, Wilks R, Cruickshank J, Rotimi C, et al. A comparative study of hypertension prevalence, awareness, treatment and control rates in St Lucia, Jamaica and Barbados. J Hypertens 1996; 14:495±501. 38 Schmeiser-Rieder A, Kunze U. Blood pressure awareness in Austria. A 20 year evaluation, 1978±1998. Eur Heart J 2000; 21:414±420. 39 Graves JW. Management of dif®cult-to-control hypertension. Mayo Clin Proc 2000; 75:278±284. 40 Luscher TF, Vetter H, Siegenthaler W, Vetter W. Compliance in hyper- tension: facts and concepts. J Hypertens 1985; 3:S3±S9. 41 Balazovjech I, Hnilica P Jr. Compliance with antihypertensive treatment in consultation rooms for hypertensive patients. J Hum Hypertens 1993; 7:581±583. 42 Rudd P. Clinicians and patients with hypertension: unsettled issues about compliances. Am Heart J 1995; 130:572±579. 43 Marques-Vidal P, Tuomilehto J. Hypertension awareness, treatment and control in the community: is the `rule of halves' still valid? J Hum Hypertens 1997; 11:213±220. 44 Berrios X, Koponen T, Huiguang T, Khaltaev N, Puska P, Nissinen A. Distribution and prevalence of major risk factors of noncommunicable diseases in selected countries: the WHO Inter-Health Programme. Bull WHO 1997; 75:99±108. 45 Singh RB, Beegom R, Ghosh S, Niaz MA, Rastogi V, Rastogi SS, et al. Epidemiological study of hypertension and its determinants in an urban population of north India. J Hum Hypertens 1997; 11:679±685. 46 Menzie Centre for Population Research, University of Tasmania, Australia. In: WHO Collaborating Centre for Population-based Cardiovascular Disease Prevention Programme (editors): Pro®le of cardiovascular dis- eases, diabetes mellitus and associated risk factors in the western Paci®c region. Manila, Phillippines: WHO, Regional Of®ce for the Western Paci®c; 1999. 47 Kaplan GA, Keil JE. Socioeconomic factors and cardiovascular disease. A review of the literature. Circulation 1993; 88:1973±1998. 48 Singh RB, Sharma JP, Rastogi V, Niaz NA, Singh MK. Prevalence and determinants of hypertension in the Indian social class and heart survey. J Hum Hypertens 1997; 11:51±56. 49 van Rossum CTM, van de Mheen H, Witteman JCM, Hofman A, Mackenbach JP, Grobbee DE. Prevalence, treatment, and control of hypertension by sociodemographic factors among the Dutch elderly. Hypertension 2000; 35:814±821. 50 Pappas G, Queen S, Hadden W, Fisher G. The increasing disparity in mortality between socio-economic groups in the United States, 1960 and 1986. N Engl J Med 1993; 329:103±109. 51 Rose G, Marmot MJ. Social class and coronary heart disease. Br Heart J 1981; 45:13±19. 52 National Cholesterol Education Program. Second Report of the Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). Circulation 1994; 89:1333±445. 1532 Journal of Hypertension 2001, Vol 19 No 9