Association of Blood Pressure and Cardiovascular Mortality in India: Mumbai Cohort Study

nature publishing group See REVIEWER COMMENTARY page 3 Association of Blood Pressure and Cardiovascular Mortality in India: Mumbai Cohort Study Mangesh S. Pednekar 1, Rajeev Gupta 2 and Prakash C. Gupta 1 Background To determine all-cause and circulatory system (cardiovascular) related mortality in subjects with different grades of hypertension, we performed a prospective study. Methods A total of 148,173 individuals aged 35 years were recruited in Mumbai, India in years 1991 1997. Clinical history and anthropometric data were obtained and hypertension-categorized using US 7th Joint National Committee guidelines into normal, prehypertension, stage-i, and stage-ii. These subjects were followed to ascertain vital status from 1997 to 3. Multivariate analysis was performed using Cox proportional analyses and adjusted hazard ratios (HRs), 95% confidence intervals (CIs) determined for mortality in various hypertension grades. Results At baseline, hypertension was in 47.3% men and 45.7% women, while prehypertension in 4.8% men and 35.9% women. In total, 13,261 persons died during average 5.5 years follow-up of whom 9,259 deaths were matched and coded using International Classification of Diseases-. Compared to those with normal blood pressure (BP), all cause mortality (HR, 95% CI) was significantly greater in stage-ii (men 1.41, 1.31 1.52; women 1.46, 1.3 1.64). Circulatory system deaths were significantly more in stage-ii (men 2.5, 1.77 2.39; women 2.6, 1.62 2.61) as well as stage-i (men 1.31, 1.14 1.52; women 1.39, 1. 1.77). Subjects with stage-ii hypertension had greater risk of death from hypertensive heart disease (men 2.77, 1.75 4.4; women 3.4, 1.73 5.35), ischemic heart disease (men 1.87, 1.54 2.28; women 1.85, 1.29 2.65), and cerebrovascular diseases (men 3., 2.42 5.5; women 3.9, 1.77 5.39). Conclusions In urban Indian subjects, compared to normal BP stage-ii hypertension is associated with increased risk of all-cause mortality, while both stage-ii and stage-i hypertension with circulatory system related mortality. Am J Hypertens 9; 22:76-84 9 American Journal of Hypertension, Ltd. Cardiovascular diseases are one of the more important causes of deaths in India. 1 Proportionate mortality due to cardio vascular disease in a nationally representative sample of >, subjects was 17% overall and 23% in middle-age. 1 The World Health Organization reports that age-standardized cardiovascular disease death rates (per,) in middleaged subjects (3 69 years) are low in developed countries such as Canada (1) and Britain (18) and high in developing countries Brazil (3), China (28), India (45), and Russia (68) (ref. 2). There is also a substantial degree of premature mortality 3 and in India % of coronary heart disease related deaths occur in persons < years of age as compared to 22% in the developed countries while 94% of stroke deaths occur at < years of age compared to 6% in developed countries. 4 Major cardiovascular risk factors can be controlled and it leads to healthier life in the middle and old age and can postpone death by at least years (refs. 4 6). Hypertension is a major cardiovascular risk factor and contributes to >65% of 1 Healis, Sekhsaria Institute for Public Health, Navi Mumbai, India; 2 Department of Medicine, Fortis Escorts Hospital, Malviya Nagar, Jaipur, India. Correspondence: Mangesh S. Pednekar (pednekarmangesh@healis.org) Received 21 March 9; first decision 27 April 9; accepted 24 June 9; advance online publication 23 July 9. doi:.38/ajh.9.131 9 American Journal of Hypertension, Ltd. stroke deaths and 25% of coronary heart disease deaths in India. 7 It is a major health issue 7,8 but there are only small studies that have prospectively studied long-term mortality from high blood pressure (BP). 9 We studied the prevalence of smoking and tobacco use, body mass index and overweight, and hypertension in a cohort of around 1, middle-aged individuals in urban locations of Mumbai, western India. 13 Data on smoking and obesity related mortality have been published. 14 17 This study reports the long-term effects of various levels of BP on all-cause and cardiovascular mortality. A specific feature of this population is a low status of awareness, treatment, and control status of hypertension. 7,12 Methods Recruitment. The Mumbai Cohort Study (MCS) was conducted in the main city of Mumbai, with mortality as the endpoint. A total of 148,173 persons aged 35 years were recruited during 1991 1997. House-to-house interviews were conducted face-to-face using a structured questionnaire. Electoral rolls, organized by area with a polling station of 1, 1, individuals as the smallest geographical unit, were used as the sampling frame. The electoral rolls provided name, age, sex, and address of all the individuals aged 18 years. We excluded polling stations that served upper-middle-class and upper class 76 october 9 VOLUME 22 NUMBER 76-84 AMERICAN JOURNAL OF HYPERTENSION Downloaded from https://academic.oup.com/ajh/article-abstract/22//76/213565 on 26 February 18

Hypertension and Mortality in Urban Indians articles housing complexes because of security issues (i.e., they were essentially gated communities ). For a selected polling station, all eligible people (aged 35 years) listed on its electoral roll were interviewed by trained field supervisors by using handheld computers (electronic diaries) in local languages (Marathi) but the information was recorded in English. The study satisfies all the criteria regarding the ethical treatment of human subjects, especially those formulated by the Indian Council of Medical Research. Participatory oral consent was obtained from all participants at the time of recruitment. Details regarding the recruitment procedures have been published previously. Data sources. The baseline survey included the measurements of BP in sitting position and were taken to the nearest 2 mm Hg using periodically calibrated mercury sphygmomanometers. The mean of the two measurements, taken 5 min apart, computed for both systolic and diastolic BP, was noted. Prevalence of hypertension in a subgroup of MCS has been reported previously. 12 Follow-up. An active house-to-house follow-up was conducted on average 5.5 years after the baseline survey. The field supervisors were provided with the list of names and addresses of all cohort members and were instructed to revisit. If the person was alive and available, a face-to-face re-interview was conducted. If the person was reported to have died, the date and place of death were recorded with extra questioning and care. Permanent migration while the subject was alive from the study area was considered as withdrawal from the study, and the date of migration was noted. The re-interviews were conducted during 1997 3. The results of follow-up have been reported earlier. 14 17 Cause of death. The deaths recorded during the follow-up of MCS were linked with the data set obtained from the municipal corporation death registers. In Mumbai, almost all the deaths are registered and medically certified. 18 For matched deaths, the underlying cause of death was derived from the cause information copied from the corporation death registers and then coded according to the International Classification of Diseases- guidelines. Cause-specific analyses were performed for various circulatory system related deaths (International Classification of Diseases- codes I 99) such as hypertensive heart diseases (I 15), ischemic heart diseases (I 25), and cerebrovascular diseases (I 69). For 1,685 randomly selected matched deaths, an independent field check was performed and matching was found to be nearly % accurate. Statistical analysis. Methodological details regarding estimation of person-years of follow-up, anthropometric measurements, and information collected from the structured questionnaire have been published elsewhere. 17 The association between various BP categories and the risk of all-cause and various circulatory system related deaths are presented as hazard ratios (HRs) and 95% confidence intervals (CIs) derived from multivariate Cox proportional hazards regression modeling 19 using SPSS 13. (SPSS, Chicago, IL). The response variable, death, was coded as a dichotomous variable (yes/no), and the time to event (or censoring) was regarded as a continuous variable. Hypertension levels were defined according to the US 7th Joint National Committee report into normal (systolic BP <1 and diastolic BP <8 mm Hg), prehypertension (systolic BP 1 139 and/or diastolic BP 8 89 mm Hg), stage-i (systolic 14 159 and/or diastolic 9 99 mm Hg), and stage-ii (systolic 1 and/or diastolic mm Hg). was further subdivided into high-normal (systolic BP 1 129 and/or diastolic BP 8 84 mm Hg) and prestage (systolic BP 13 139 and/or diastolic BP 85 89 mm Hg). If the systolic and diastolic pressure reading belonged to different categories, the higher of the two readings was used to assign the BP category. In contrast to multiple BP recordings on several days as recommended by US 7th Joint National Committee, we measured BP several times in a single visit as reported in earlier Indian studies. 7,8 Respondents were broadly classified as having never used tobacco, using smokeless tobacco only, being a smoker (which may include few smokers plus smokeless tobacco users). Age, education, religion, mother tongue, tobacco use, body mass index, and hypertension (abstracted from baseline data) were fitted as independent variables in the final model. 19 Adjusted HR, 95% CI, and survival curves were estimated for different grades of hypertension, stratified by gender. Results We recruited 148,173 subjects in the study during 1991 1997 in the defined locations comprising 88,658 men and 59,515 women. BP data were available in 87,459 men and 59,368 women (Table 1). Baseline characteristics show that illiteracy was more among women (45.3%) than men (16.9%). Only 15.8% men and 5.9% women had more than secondary level education. High prevalence of overweight or obesity (body mass index 25 kg/m 2 ) was noted in both men (.2%) and women (29.5%). Prevalence of any tobacco use was also high (men.% and women 59.7%). Prevalence of hypertension was 47.3% in men (stage-i 34.2%; stage-ii 13.1%) and 45.7% in women (stage-i 29.%; stage-ii 16.7%). Additionally, prehypertension was observed among 4.8% men and 35.9% women. Less than % men and women had normal BP level. The mean follow-up was 5.5 years, which translated into 774,129 person-years. Of the total recruited subjects, 7,265 could not be traced; the most common reason was the demolition of their residential buildings (6,452 subjects). No differences in baseline variables were observed in subjects whose data were available as compared to those lost to follow-up (see Supplementary Table S1 online), Among the remaining 14,98 subjects, 13,261 persons died while 127,647 were alive (of which 25,777 subjects had migrated outside study area) at the end of follow-up period. Of the total 13,261 deaths, 11,249 died within study area and among those died within study area 9,259 deaths were coded using International Classification of Diseases- (Figure 1). 14 Details regarding the matching and coding of underlying causes of deaths have been AMERICAN JOURNAL OF HYPERTENSION VOLUME 22 NUMBER october 9 77 Downloaded from https://academic.oup.com/ajh/article-abstract/22//76/213565 on 26 February 18

Hypertension and Mortality in Urban Indians Table 1 Baseline demographic details of the study cohort Variables Age groups (n = 87,459) % (95% CI) (n = 59,368) % (95% CI) 35 39 9.6 (9.4 9.7) 24.9 (24.5 25.2) 4 44 8. (7.8 8.2) 16.9 (16.6 17.2) 45 49 26.6 (26.4 26.9) 15.4 (15.1 15.7) 54 16.1 (15.9 16.4) 13.3 (13. 13.6) 55 59 12.2 (12. 12.4) 9.9 (9.6.1) 64 11. (.8 11.2) 9.2 (8.9 9.4) 65 69 7.6 (7.4 7.7) 5.1 (4.9 5.2) 8.9 (8.7 9.1) 5.5 (5.3 5.6) Educational status Illiterate 16.9 (16.7 17.2) 45.3 (44.9 45.7) Primary 37.8 (37.5 38.1) 35. (34.7 35.4) Middle 29.5 (29.2 29.8) 13.8 (13.5 14.1) Secondary 9.4 (9.2 9.6) 4.3 (4.1 4.4) College 6.4 (6.2 6.5) 1.6 (1.5 1.7) Body composition 62.4 (62. 62.7) 51.4 (51. 51.8) Thin 9.2 (9. 9.4) 9.1 (8.9 9.3) Very thin 4. (3.9 4.1) 4.3 (4.2 4.5) Extremely thin 4.3 (4.2 4.5) 5.7 (5.6 5.9) Overweight 17.5 (17.3 17.8) 22.6 (22.2 22.9) Obese 2.7 (2.5 2.8) 6.9 (6.7 7.1) Tobacco use Nonuser 3. (29.7 3.3) 4.3 (39.9 4.7) Smokeless 38.6 (38.3 38.9) 59.3 (58.9 59.7) Smoker 16.3 (16.1 16.6).3 (.2.3) Both 15.1 (14.8 15.3).2 (.1.2) Hypertension groups 11.9 (11.7 12.1) 18.5 (18.2 18.8) 4.8 (4.5 41.1) 35.9 (35.5 36.2) Stage-I 34.2 (33.9 34.5) 29. (28.6 29.3) Stage-II 13.1 (12.9 13.4) 16.7 (16.4 17.) CI, confidence interval. published. 14 17 The age-adjusted all-cause mortality rates per, person years in women ranged from 1,67 (prehypertension) to 1,587 (stage-ii hypertension) and in men from 1,737 to 2,535, respectively. Cox regression analysis of mortality for all-cause and for various circulatory system causes after adjustment for age and multiple baseline variables reveal significant differences in various BP groups (Figures 2 and 3). At every time point greater all-causes mortality was observed in subjects with stage-ii hypertension as compared with other groups. Similarly, higher risk of dying from circulatory system diseases was observed in subjects with stage-i as well as stage-ii hypertension as compared with other groups (Figure 3). Cox regression Re-interviewed = 9,282 Migrated = 25,777 Not available for re-interview = 11,588 Status alive = 127,647 Mumbai Cohort n = 148,173 After average 5.5 years of follow-up Vital status known = 14,98 Within study area = 11,249 Status died = 13,261 Matched with BMC death register = 9,259 Unmatched = 1,99 Loss to follow-up = 7,265 Outside study area = 2,12 Figure 1 Flow diagram of house-to-house follow-up of the study subjects. 14 1..98.96.94.92.9.88.86 1..98.96.94.92.9.88.86 I using systolic and diastolic BP as a continuous variable (with increment of mm Hg) for circulatory system related deaths, demonstrated 1.2% increased risk of deaths for systolic BP (HR = 1.12; CI = 1.11 1.14) and 1.5% for diastolic BP (HR = 1.15; CI = 1.12 1.17). Similar risk pattern was I Figure 2 Multivariate adjusted survival curves for all-cause mortality in the study cohort. 78 october 9 VOLUME 22 NUMBER AMERICAN JOURNAL OF HYPERTENSION Downloaded from https://academic.oup.com/ajh/article-abstract/22//76/213565 on 26 February 18

Hypertension and Mortality in Urban Indians articles 1..98.96 I The risk pattern remained statistically unchanged across tobacco categories, except for circulatory system related deaths and within circulatory system for cerebrovascular deaths. At stage-ii hypertension, smokeless tobacco users had higher risk of death than corresponding nonusers (for circulatory system deaths; HR = 2.49 vs. 1.43 and for cerebrovascular disease deaths; HR = 5.1 vs. 1.4), while at both stage-i and stage-ii smokers had higher risk of cerebrovascular disease deaths than corresponding nonusers (HR = 2.71 vs..69 and HR = 4.92 vs. 1.4, respectively). 1..98.96 observed for various circulatory system related deaths such as hypertensive heart disease, ischemic heart disease, and cerebro vascular diseases (Figure 4). Adjusted HRs in various BP groups with subjects with normal BP taken as reference are shown in Tables 2 and 3. The all-cause mortality was significantly higher in men (HR = 1.41; CI = 1.31 1.52) and women (HR = 1.46, CI = 1.3 1.64) with stage-ii hypertension, whereas, no significant difference was observed in subjects with stage-i hypertension. In contrast, subjects with prehypertension (mainly subjects with highnormal BP: men HR =.91, CI =.84.98; women HR =.74, CI =.65.85) appeared at lower risk. Increased risk of circulatory system related deaths was observed in subjects with stage-i (men HR = 1.31, CI = 1.14 1.52; women HR = 1.39, CI = 1. 1.77) as well as stage-ii (men HR = 2.5, CI = 1.77 2.39; women HR = 2.6, CI = 1.62 2.61) hypertension, in contrast, women with high-normal BP were at lower risk (HR =., CI =.52.95). Within circulatory system related deaths, increased risk of hypertensive heart diseases, ischemic heart diseases, and cerebrovascular diseases was observed in subjects with stage-ii hypertension. Similarly, increased risk of hypertensive heart diseases (women), ischemic heart diseases (men), and cerebrovascular diseases (men) was observed in subjects with stage-i hypertension (Tables 2 and 3). We also estimated adjusted HR in men for various BP categories stratified according to the different tobacco habits (nonuser, smokeless tobacco users, and smokers) (Table 4). I Figure 3 Adjusted survival curves for mortality from circulatory system diseases (cardiovascular diseases) in Mumbai. Discussion MCS shows that among urban Indians stage-ii hypertension is associated with increased all-cause mortality, while both stage-i and stage-ii hypertension are associated with increased circulatory system related mortality. Subjects with high-normal BP had lower all-cause mortality (mainly due to causes other than circulatory system), whereas subjects with prehypertension had similar risk of all-cause mortality to that of normal BP. In contrast, men with prehypertension had increased risk of cerebrovascular disease deaths. The mortality risk of hypertension with all-cause mortality remains statistically unchanged even after stratification by tobacco habit, except for deaths due to cerebrovascular diseases. At stage-ii hypertension deaths due to cerebrovascular diseases for both smokers as well as smokeless tobacco users had higher risk than corresponding risk for nonusers. MCS has multiple limitations and strengths. A major limitation is the measurement of BP in a single visit. Regression dilution bias and regression-to-the-mean are well-known issues in hypertension epidemiology and all the prospective studies should be corrected for these factors. 21,22 However, due to logistic reasons many international studies have been performed using methodology that is similar to MCS. 6 Moreover, recent studies suggest that isolated high BP readings ( whitecoat hypertension ) carry adverse prognosis similar to established hypertension and therefore our observations are important. 23 We acknowledge the possibility of misclassification of US 7th Joint National Committee stages in MCS due to single visit BP measurements and the lack of correlation of prehypertension with cardiovascular mortality suggests that this may be true. Second, we obtained cause-of-death information from local death registries. Cause-of-death registries are often imprecise in India and this could be important in MCS. On the other hand, the Mumbai registry is one of the oldest and most efficient systems of mortality ascertainment and therefore should be most reliable data from this country. 17 Third, drug therapy can substantially influence mortality.,24,25 However, in the MCS the baseline hypertension awareness and therapy status was low, 12 which is similar to studies reported from other regions of India. 7 Low level of knowledge about therapy status in the study population is unlikely to influence outcomes. However, awareness of high BP would have increased after the baseline survey. 26 Fourth, we do not have data on other cardiovascular risk factors, such as diabetes and lipid abnormalities, which are also major predictors of cardiovascular mortality. 3 AMERICAN JOURNAL OF HYPERTENSION VOLUME 22 NUMBER october 9 79 Downloaded from https://academic.oup.com/ajh/article-abstract/22//76/213565 on 26 February 18

Hypertension and Mortality in Urban Indians Hypertensive heart disease 1. 1..996.996.992.992 I I Ischemic heart disease 1. 1..996.996.992.988.992 I.984.98 I Cerebrovascular disease 1. 1..996.996.992 I.992 I Figure 4 Adjusted survival curves for mortality from various circulatory system diseases in men and women: hypertensive heart disease, coronary heart disease, and cerebrovascular disease in Mumbai, India. On the other hand smoking and tobacco usage are also major vascular risk factors in India and MCS has shown significant interaction of tobacco with BP. 3 Fifth, MCS excluded polling stations comprising upper-middle class and upper-class housing complexes that were not accessible due to security issues. Similarly, the study excluded homeless persons, such as footpath dwellers, as they were generally excluded from the voter s list. Therefore, MCS may not be truly representative 8 october 9 VOLUME 22 NUMBER AMERICAN JOURNAL OF HYPERTENSION Downloaded from https://academic.oup.com/ajh/article-abstract/22//76/213565 on 26 February 18

Hypertension and Mortality in Urban Indians articles Table 2 Person-years, number of deaths, adjusted hazard ratios, and 95% confidence intervals for major causes of death in men by different categories of BP Total High normal Prestage Stage-I hypertension Stage-II hypertension Person-years 53,666 183,298 128,78 55,2 149,424 57,4 All deaths No. of deaths 1,152 3,56 1,921 1,135 3,38 2,73 Hazard ratios 1..94.91**.99 1.5 1.41*** 95% CI.87 1..84.98.91 1.7.98 1.12 1.31 1.52 Circulatory system No. of deaths 233 854 521 333 1,71 825 Hazard ratios 1. 1.13 1.11 1.18 1.31*** 2.5*** 95% CI.98 1.31.95 1.29.99 1.39 1.14 1.52 1.77 2.39 Hypertensive diseases No. of deaths 23 77 41 36 6 112 Hazard ratios 1. 1.5.91 1.26 1.3 2.77*** 95% CI.65 1.68.55 1.53.74 2.15.82 2.6 1.75 4.4 Ischemic heart disease No. of deaths 136 527 338 189 656 454 Hazard ratios 1. 1.16 1.19 1.11 1.31*** 1.87*** 95% CI.96 1.4.97 1.45.89 1.38 1.9 1.59 1.54 2.28 Cerebrovascular diseases No. of deaths 35 148 78 197 196 Hazard ratios 1. 1.37 1.16 1.73*** 1.71*** 3.*** 95% CI.95 1.99.78 1.74 1.15 2.61 1.19 2.47 2.42 5.5 Other circulatory system No. of deaths 39 2 64 38 112 63 Hazard ratios 1..95.92 1. 1.4 1.26 95% CI.65 1.38.62 1.38.63 1.57.72 1.52.83 1.91 Causes other than circulatory system No. of deaths 58 1,256 811 445 1,24 666 Hazard ratios 1..88**.86***.91.96 1.15** 95% CI.79.97.77.96.8 1.3.86 1.6 1.3 1.3 Hazard ratios are adjusted for age, education, religion, mother tongue, tobacco habit, and body mass index. CI, confidence interval. *P <.5. **P <.1. ***P <.1. of the Mumbai population. The strengths of MCS include a population-based nature of the cohort, very large sample size that is much more than many of the earlier studies in the developing countries, and use of recent BP cutoffs to identify importance of different grades of hypertension. International Society of Hypertension has estimated global burden of high BP related diseases. 26 Worldwide, 7.6 million deaths (about 13.5% of the global total) have been attributed to high BP. It was also estimated that 54% of stroke and 47% of ischemic heart disease deaths are attributable to high BP. High BP is responsible for 9.6% of deaths in South Asia. The deaths attributable to high BP ranged from 4.% in Sub-Saharan Africa to 35.% in Europe and Central Asia. Using HRs and proportion of deaths from Tables 2 and 3 we estimated proportion of population attributable 27 deaths due to high BP (stage-i and stage-ii). In MCS, high BP was responsible for 7.9% male and 11.5% female all-cause deaths. Also, of the total circulatory system deaths, high BP was responsible for 22.7% male and 3.2% female deaths. Among men within various circulatory system deaths, high BP was responsible for 38.5% of cerebrovascular disease deaths, 3.2% of hypertensive diseases deaths, and.7% of ischemic heart disease deaths. Among women these proportions were 45.1, 47.2, and 23.1%, respectively. Our estimates on proportions of attributable deaths are consistent with those made by International Society of Hypertension 26 AMERICAN JOURNAL OF HYPERTENSION VOLUME 22 NUMBER october 9 81 Downloaded from https://academic.oup.com/ajh/article-abstract/22//76/213565 on 26 February 18

Hypertension and Mortality in Urban Indians Table 3 Person-years, number of deaths, adjusted hazard ratios, and 95% confidence intervals for major causes of death in women by different categories of BP Total High normal Prestage Stage-I hypertension Stage-II hypertension Person-years 61,593 119,933 85,842 34,91 95,447 53,264 All deaths No. of deaths 458 729 392 337 1,8 1,145 Hazard ratios 1..84***.74***.98 1.3 1.46*** 95% CI.74.94.65.85.85 1.13.92 1.16 1.3 1.64 Circulatory system No. of deaths 93 167 8 87 358 47 Hazard ratios 1..86.* 1.8 1.39*** 2.6*** 95% CI.66 1.11.52.95.8 1.45 1. 1.77 1.62 2.61 Hypertensive diseases No. of deaths 15 26 12 14 79 2 Hazard ratios 1..85.69 1.6 1.85** 3.4*** 95% CI.45 1.61.32 1.47.51 2.21 1.5 3.27 1.73 5.35 Ischemic heart disease No. of deaths 41 83 39 44 144 174 Hazard ratios 1..91.73 1.15 1.18 1.85*** 95% CI.62 1.32.47 1.14.75 1.78.83 1.69 1.29 2.65 Cerebrovascular diseases No. of deaths 16 26 16 71 97 Hazard ratios 1..78.51 1.19 1.69 3.9*** 95% CI.42 1.47.23 1.13.59 2.41.96 2.95 1.77 5.39 Other circulatory system No. of deaths 21 32 19 13 64 34 Hazard ratios 1..83.83.84 1.38 1.1 95% CI.48 1.45.44 1.54.42 1..82 2.32.57 1.8 Causes other than circulatory system No. of deaths 241 337 183 154 43 4 Hazard ratios 1..79***.***.93.9 1.* 95% CI.67.93.58.85.76 1.14.76 1.6 1.1 1.43 Hazard ratios are adjusted for age, education, religion, mother tongue, tobacco habit, and body mass index. CI, confidence interval. *P <.5. **P <.1. ***P <.1. despite the fact that there was a slight difference in the definition of hypertension in the latter. The Seven Countries Study determined relation between BP and coronary heart disease mortality among men in different parts of the world. 28 The age-standardized 25-year mortality (per, person years) from coronary heart disease varied from 22 in the Mediterranean countries to in northern Europe. It was demonstrated that in all the six populations the relative increase in long-term mortality for a given increase in BP was similar. It was reported that with each mm Hg increment of systolic BP the coronary heart disease mortality increased by 17% before adjustment for before subject variability and 28% after adjustment and the risk increased by 13% for a 5 mm Hg increase in diastolic BP. MCS also demonstrated that with mm Hg increment of systolic BP the circulatory system deaths increased by 1.2% and this was 1.5% for diastolic BP after average 5.5 years of follow-up. However, MCS was of a shorter duration than the Seven Countries Study and findings were based on BP measurement on a single visit. In the Asia Pacific Cohort Studies Collaboration, cardiovascular deaths including ischemic heart and strokes were monitored in 37 studies involving >425,325 subjects in various countries of Asia and Australia. 29 A continuous loglinear relationship was observed for all the three conditions with BP to the level of 115 mm Hg systolic. This is in contrast to MCS where systolic BP levels between 1 and 139 mm Hg 82 october 9 VOLUME 22 NUMBER AMERICAN JOURNAL OF HYPERTENSION Downloaded from https://academic.oup.com/ajh/article-abstract/22//76/213565 on 26 February 18

Hypertension and Mortality in Urban Indians articles Table 4 Adjusted hazard ratios and 95% confidence intervals in men for major causes of death for various categories of BP stratified by tobacco habits Categories All deaths Hazard ratios (95% confidence intervals) Nonuser Smokeless tobacco users Smokers 1. 1. 1..93 (.8 1.8).93 (.83 1.4).95 (.85 1.6) Stage-I hypertension 1.11 (.96 1.29) 1.3 (.92 1.15) 1.3 (.93 1.15) Stage-II hypertension 1.37 (1.16 1.61)*** 1.43 (1.27 1.62)*** 1.42 (1.26 1.59)*** Circulatory system (ICD codes I 99) 1. 1. 1..98 (.75 1.29) 1.23 (.96 1.59) 1.15 (.91 1.46) Stage-I hypertension 1.13 (.87 1.48) 1.46 (1.13 1.88)*** 1.31 (1.3 1.66)* Stage-II hypertension 1.43 (1.7 1.9)** 2.49 (1.93 3.21)*** 2.21 (1.73 2.82)*** Hypertensive diseases (ICD codes I 15) 1. 1. 1..92 (.39 2.16) 1.23 (.56 2.73) 1.1 (.45 2.27) Stage-I hypertension 1.29 (.57 2.96) 1.17 (.53 2.58) 1.41 (.64 3.) Stage-II hypertension 2.7 (.88 4.87) 2.99 (1.39 6.45)*** 3.39 (1.54 7.47)*** Ischemic heart diseases (ICD codes I 25) 1. 1. 1. 1.25 (.85 1.82) 1.13 (.8 1.58) 1.14 (.85 1.53) Stage-I hypertension 1.49 (1.3 2.17)* 1.38 (.99 1.92) 1.16 (.86 1.56) Stage-II hypertension 1.8 (1.21 2.67)*** 2.3 (1.44 2.86)*** 1.84 (1.35 2.51)*** Cerebrovascular diseases (ICD codes I 69) 1. 1. 1. (1).93 (. 1.72) 1.61 (.88 2.93) 1.49 (.72 3.11) Stage-I hypertension (2).69 (.37 1.32) 2.3 (1.12 3.66)** 2.71 (1.34 5.48)*** Stage-II hypertension (3) 1.4 (.53 2.3) 5.1 (2.79 8.99)*** 4.92 (2.4.)*** Hazard ratios are adjusted for age, education, religion, mother tongue, and body mass index. ICD, International Classification of Diseases. *P <.5. **P <.1. ***P <.1. were not associated with increased cardiovascular mortality and only stage-i and stage-ii hypertension were important. Greater all-cause mortality in subjects with normal BP as compared to those with high normal or prehypertension is an intriguing finding in this study. This could be due to the fact that we measured BP on only 1 day and variable whitecoat effect has led to misclassification of subjects at lower end of BP distribution. Our findings are not comparable to other studies in India as none exist. High-normal BP or prehypertension has been identified as independent cardiovascular risk factor. The Framingham study 3 reported that subjects with prehypertension (men HR = 1.6, women HR = 2.5) as well as high-normal BP (men HR = 1.3) were associated with increased mortality. In MCS, men with prestage (HR = 1.73) and high-normal (HR = 1.16) BP had increased risk of deaths from cerebrovascular diseases. However, many of the subjects in the prehypertension category might be having white-coat hypertension. 24 It may be noted that follow-up period in MCS was 5.5 years whereas this was 12 years in the Framingham study, 3 but both the studies highlight the importance of high-normal BP and prehypertension. In addition to smoking, smokeless tobacco use has also emerged as an important contributor of mortality (Table 4). A recent study from Sweden, reported an increased risk of fatal ischemic stroke associated with current snuff use (relative risk = 1.72), whereas no increased risk was noted for hemorrhagic stroke. 31 MCS also demonstrated the association of BP categories with various causes of deaths stratified by tobacco use. At stage-ii hypertension, deaths due to cerebrovascular diseases for both smokers as well as smokeless tobacco users had higher risk than corresponding risk for non-tobacco users. This finding may be very important for the populations like in India where the prevalence of both tobacco use and hypertension is high. Hypertension is a major public health problem; its control can prevent cardiovascular mortality. The populationwide Canadian Hypertension Education Program Outcomes Research Taskforce study 32 reported that despite increasing AMERICAN JOURNAL OF HYPERTENSION VOLUME 22 NUMBER october 9 83 Downloaded from https://academic.oup.com/ajh/article-abstract/22//76/213565 on 26 February 18

Hypertension and Mortality in Urban Indians incidence of hypertension in the Canadian population from years 1995 to 5, the mortality rates had declined, the relative decrease being more in men ( 22.2%) as compared to women ( 7.3%). This decline has been attributed to increased prevalence of diagnosed hypertension and better treatment and control status. The BP Lowering Treatment Trialists Collaboration also reported significant decrease in all-causes, all-cause cardiovascular and specific cardiac disease mortality by BP treatment. 33 Thus, the lessons from MCS are to increase focus on hypertension awareness, its treatment and better control for prevention of adverse cardiovascular outcomes. Supplementary material is linked to the online version of the paper at http://www.nature.com/ajh Acknowledgments: This study was conducted in collaboration with the International Agency for Research on Cancer, Lyon, France (Collaborative Research Agreement DEP/89/12), the Clinical Trial Service Unit of the University of Oxford, Oxford, United Kingdom, and the World Health Organization Geneva, Switzerland. All of these entities provided funding for the study. The authors wish to acknowledge the cooperation of the Municipal Corporation of Greater Mumbai (BMC) for providing access to the cause-of-death information. Disclosure: The authors declared no conflict of interest. 1. Million Death Study Investigators. Preliminary Report on Causes of Death in India 1 3. Registrar General of India: New Delhi, 7. 2. World Health Organization. Preventing Chronic Diseases: A Vital Investment. World Health Organization: Geneva, 5. 3. Gupta R, Joshi P, Mohan V, Reddy KS, Yusuf S. Epidemiology and causation of coronary heart disease and stroke in India. Heart 8; 94:16 26. 4. Gaziano T, Reddy KS, Paccaud F, Horton S, Chaturvedi V. Cardiovascular disease. In: Jamison DT, Breman JG, Measham AR, Alleyne G, Claeson M, Evans DB, Jha P, Mills A, Musgrove P (eds), Disease Control Priorities in Developing World. Oxford University Press: Oxford, 6, pp. 645 662. 5. Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F, McQueen M, Budaj A, Pais P, Varigos J, Lisheng L; INTERHEART Study Investigators. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet 4; 364:937 952. 6. Ezzati M, Lopez AD, Rodgers A, Murray CJL. Comparative Quantification of Health Risks. Global and Regional Burden of Disease Attributable to Major Risk Factors. World Health Organization, Geneva, 4. 7. Gupta R. Trends in hypertension epidemiology in India. J Hum Hypertens 4; 18:73 78. 8. Gupta R, al-odat NA, Gupta VP. Hypertension epidemiology in India: metaanalysis of year prevalence rates and blood pressure trends. J Hum Hypertens 1996; :465 472. 9. Gupta R, Gupta KD, Kasliwal P. Influence of mild to moderate treated hypertension on 9 11 year mortality in patients with pre-existing coronary heart disease. J Hum Hypertens 1992; 6:313 316.. Gupta PC. Survey of sociodemographic characteristics of tobacco use among 99,598 individuals in Bombay, India using handheld computers. Tob Control 1996; 5:114 1. 11. Shukla HC, Gupta PC, Mehta HC, Hebert JR. Descriptive epidemiology of body mass index of an urban adult population in western India. J Epidemiol Community Health 2; 56:876 88. 12. Gupta PC, Gupta R, Pednekar MS. Hypertension prevalence and blood pressure trends in 88 653 subjects in Mumbai, India. J Hum Hypertens 4; 18:97 9. 13. Pednekar MS, Gupta PC, Shukla HC, Hebert JR. Association between tobacco use and body mass index in urban Indian population: implications for public health in India. BMC Public Health 6; 6:. 14. Pednekar MS. The impact of Tobacco Use and/or Body Composition on Adult Mortality in Urban Developing Country Population. Acta Universitatis Tamperensis 134, Tampere University Press, Tampere, 8. <http://acta.uta.fi/ pdf/978-951-44-7431-6.pdf>. 15. Gupta PC, Pednekar MS, Parkin DM, Sankaranarayanan R. Tobacco associated mortality in Mumbai (Bombay) India. Results of the Bombay Cohort Study. Int J Epidemiol 5; 34:1395 142. 16. Pednekar MS, Gupta PC, Hebert JR, Hakama M. Joint effects of tobacco use and body mass on all-cause mortality in Mumbai, India: results from a populationbased cohort study. Am J Epidemiol 8; 167:33 34. 17. Pednekar MS, Hakama M, Hebert JR, Gupta PC. Association of body mass index with all-cause and cause-specific mortality: findings from a prospective cohort study in Mumbai (Bombay), India. Int J Epidemiol 8; 37:524 535. 18. Jha P, Gajalakshmi V, Gupta PC, Kumar R, Mony P, Dhingra N, Peto R; RGI-CGHR Prospective Study Collaborators. Prospective study of one million deaths in India: rationale, design, and validation results. PLoS Med 6; 3:e18. 19. Cox DR. Regression models and life-tables. J R Stat Soc (B) 1972; 34:187 2.. Bakris,. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, Jones DW, Materson BJ, Oparil S, Wright JT Jr, Roccella EJ; National Heart, Lung, and Blood Institute Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High BP; National High BP Education Program Coordinating Committee. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High BP: the JNC 7 report. JAMA 3; 289:25 2572. 21. Whelton PK. Epidemiology of hypertension. Lancet 1994; 344:1 6. 22. Lewington S, Clarke R, Qizilbash N, Peto R, Collins R; Prospective Studies Collaboration. Age-specific relevance of usual BP to vascular mortality: a metaanalysis of individual data for one million adults in 61 prospective studies. Lancet 2; 3:193 1913. 23. Lloyd-Jones DM, Levy D. Epidemiology of hypertension. In: Black HR, Elliott WJ (eds), Hypertension: A Companion to Braunwald s Heart Disease. Saunders: Philadelphia, 7, pp. 3 14. 24. Kaplan NM, Opie LH. Controversies in hypertension. Lancet 6; 367:168 176. 25. Tu K, Chen Z, Lipscombe LL. Prevalence and incidence of hypertension from 1995 to 5: a population-based study. CMAJ 8; 178:1429 1435. 26. Laws CMM, Hoorn SV, Rodgers A, for the International Society of Hypertension. Global burden of BP related disease, 1. Lancet 8; 371:1513 1518. 27. Rockhill B, Newman B, Weinberg C. Use and misuse of population attributable fractions. Am J Public Health 1998; 88:15 19. 28. van den Hoogen PC, Feskens EJ, Nagelkerke NJ, otti A, Nissinen A, Kromhout D. The relation between blood pressure and mortality due to coronary heart disease among men in different parts of the world. Seven Countries Study Research Group. N Engl J Med ; 342:1 8. 29. Lawes CM, Rodgers A, Bennett DA, Parag V, Suh I, Ueshima H, MacMahon S; Asia Pacific Cohort Studies Collaboration. Blood pressure and cardiovascular disease in the Asia Pacific region. J Hypertens 3; 21:7 716. 3. Vasan RS, Larson MG, Leip EP, Evans JC, O Donnell CJ, Kannel WB, Levy D. Impact of high-normal blood pressure on the risk of cardiovascular disease. N Engl J Med 1; 345:1291 1297. 31. Hergens MP, Lambe M, Pershagen G, Terent A, Ye W. Smokeless tobacco and the risk of stroke. Epidemiology 8; 19:794 799. 32. Tu K, Chen Z, Lipscombe LL. Mortality among patients with hypertension from 1995 to 5: a population-based study. CMAJ 8; 178:1436 144. 33. BP Lowering Treatment Trialists Collaboration. Effects of different BP lowering regimens on major cardiovascular events: results of prospectively designed overviews of randomised trials. Lancet 3; 362:1527 1535. 84 october 9 VOLUME 22 NUMBER AMERICAN JOURNAL OF HYPERTENSION Downloaded from https://academic.oup.com/ajh/article-abstract/22//76/213565 on 26 February 18