Article Text

Download PDFPDF

Epidemiology and causation of coronary heart disease and stroke in India
  1. R Gupta1,
  2. P Joshi2,
  3. V Mohan3,
  4. K S Reddy4,
  5. S Yusuf5
  1. 1
    Fortis Escorts Hospital, Jaipur, India
  2. 2
    Government Medical College, Akola, India
  3. 3
    Madras Diabetes Research Foundation, Chennai, India
  4. 4
    Public Health Foundation of India, New Delhi, India
  5. 5
    Population Health Research Institute, McMaster University, Hamilton, Canada
  1. Dr R Gupta, Fortis Escorts Hospital, JLN Marg, Malviya Nagar, Jaipur 302017, India; rajeevg{at}


Cardiovascular diseases are major causes of mortality and disease in the Indian subcontinent, causing more than 25% of deaths. It has been predicted that these diseases will increase rapidly in India and this country will be host to more than half the cases of heart disease in the world within the next 15 years. Coronary heart disease and stroke have increased in both urban and rural areas. Case–control studies indicate that tobacco use, obesity with high waist:hip ratio, high blood pressure, high LDL cholesterol, low HDL cholesterol, abnormal apolipoprotein A-1:B ratio, diabetes, low consumption of fruits and vegetables, sedentary lifestyles and psychosocial stress are important determinants of cardiovascular diseases in India. These risk factors have increased substantially over the past 50 years and to control further escalation it is important to prevent them. National interventions such as increasing tobacco taxes, labelling unhealthy foods and trans fats, reduction of salt in processed foods and better urban design to promote physical activity may have a wide short-term impact.

Statistics from

The World Health Organisation (WHO) reports that in the year 2005 cardiovascular diseases caused 17.5 million (30%) of the 58 million deaths that occurred world wide.1 Cardiovascular diseases such as coronary heart disease (CHD) and strokes are the largest causes of death in developing countries and are one of the main contributors to disease burden.2 Between 1990 and 2020 these diseases are expected to increase by 120% for women and 137% for men in developing countries as compared with 30–60% in developed countries.3 It has been projected that by the year 2010 60% of the world’s patients with heart disease will be in India.2 Age-standardised cardiovascular disease death rates (per 100 000) in middle-aged subjects (30–69 years) are low in developed countries such as Canada (120) and Britain (180) and high in developing countries Brazil (320), China (280), Pakistan (400), Nigeria (410), Russia (680) and India (405).1 Moreover, in India about 50% of CHD-related deaths occur in people younger than 70 years compared with only 22% in the West.2 In developing countries 94% of deaths from stroke occur in people aged <70 years in contrast to 6% in developed countries.2 This article summarises the evidence which shows that in India there is a fully developed epidemic of cardiovascular diseases such as CHD and stroke. Several studies from India show that the disease burden estimated by disease prevalence studies is increasing in both urban and rural populations, the risk factors for this epidemic are similar to those elsewhere in the world, and there has been a substantial increase in these risk factors in India in recent years.


According to the Global Burden of Diseases Study in India, in the year 1990 CHD caused 0.62 million deaths in men and 0.56 million deaths in women (total 1.18 million) and strokes were responsible for 0.23 million deaths in men and 0.22 million deaths in women (total 0.45 million).3 By the year 2000 CHD had led to 1.59 million deaths and stroke to 0.60 million deaths.3 Mortality from these conditions is predicted to increase rapidly and the absolute numbers of CHD cases in India to overtake those of the established market economies and China while stroke mortality would also increase (table 1).

Table 1 Cardiovascular deaths in India, China and developed countries in millions (Global Burden of Diseases Study)10

Leading major cause groups of deaths during 1984 to 1998 have been reported by the Registrar General of India.4 Trends show that there has been a significant decline in the proportion of deaths from infectious diseases from 22% to 16%, whereas mortality from cardiovascular diseases has increased from 21% to 25%. The reliability of mortality data has been questioned on issues of medical classification of deaths as a large number of deaths are recorded as senility or old age and a major cause in this group is cardiovascular disease, hence the current sources may underestimate the deaths due to cardiovascular diseases.

Gajalakshmi et al evaluated the causes of deaths in subjects aged 25–69 years using a verbal autopsy in urban Chennai.5 From 1995 to 1997 there were 72 165 deaths in Chennai, of which 5388 (7%) were due to suicides, violence and accidents. Verbal autopsy to identify clearly the cause of death was performed in 48 357 of the remaining 66 777 (72.4%). Cardiovascular diseases were the largest group with 18 680 deaths (38.6%) followed by cancer (8.7%), tuberculosis (5.8%) and respiratory causes (3.5%). Unspecified medical causes accounted for 19 825 (41.0%) deaths.

Joshi et al reported mortality statistics from the Andhra Pradesh Rural Health Initiative.6 The study prospectively evaluated causes of death in a population cluster of about 150 000 subjects, and it was reported that in both men and women cardiovascular diseases were the major causes of deaths. In this study there were 1354 deaths in the first year of follow-up and analysis using a validated verbal autopsy instrument showed that diseases of the circulatory system caused 34% of male and 30% of female deaths.

Mohan et al reported that in urban south Indians, mortality rates are twofold higher in people with diabetes than in non-diabetic subjects, and among diabetic subjects, cardiovascular (52.9%) and renal diseases (23.5%) were the commonest causes of death.7 These studies show that these diseases are a major cause of death in Indian urban and rural locations. The continuing prospective Sample Registration System Verbal Autopsy (SRS-VA) Million Deaths Study in India8 and the Prospective Urban Rural Epidemiology (PURE) Study9 should provide more definitive answers.


The Global Burden of Diseases Studies reported the disability-adjusted life years (DALYs) lost by CHD in India in years 1990 and 2000. In 1990 CHD was responsible for 5.6 million DALYs in men and 4.5 million in women.10 This was projected to increase in years 2000, 2010 and 2020 among men to 7.67, 10.46 and 14.36 and in women to 5.55, 6.55 and 7.66 million, respectively. Data from the World Health Report (2002)11 shows that in South East Asia, a region that also includes other countries with high childhood and high adult mortality (Bangladesh, Bhutan, Maldives, Myanmar, Nepal and North Korea), cardiovascular diseases caused a loss of 35.4 million DALYs, CHD 17.99 million, rheumatic heart disease 2.34 million and cerebrovascular diseases 7.98 million. India comprises 81.2% of population in this region and therefore the DALYs lost in India owing to CHD according to this report would be 14.61 million which is much more than the earlier projections.

In the absence of reliable nationwide prospectively collected morbidity data, estimates of the burden of CHD have been based on indicators from population-based, cross-sectional surveys. Multiple epidemiological studies have been performed in urban and rural populations in India over the past 60 years.12 Table 2 shows the prevalence of CHD in various studies. Limitations of these studies include small and variable sample size, variable and, at times, low response rates, inappropriate diagnostic criteria such as history, non-specific electrocardiographic changes—for example, abnormal ST-T waves, lack of age standardisation, and incomplete reporting of results.12 13 On the other hand most of these studies used similar population-based recruitments, methodology and diagnostic criteria (known CHD, Rose questionnaire angina and/or electrocardiographic Q-ST-T changes) as shown in table 2.

Table 2 Coronary heart disease prevalence studies in India

The age groups evaluated in these studies are variable and therefore we compared studies that included subjects aged ⩾25–30 years to determine secular trends. A high prevalence is consistently seen in studies at urban locations (Chandigarh 6.6%,16 Rohtak 3.6%,17 Delhi 9.7%,18 Varanasi 6.5%,19 Jaipur 9.2%,20 Trivandrum 12.7%,21 Chennai 11.0%,22 Jaipur 9.1%,23 and Goa 12.5%24) as compared with the rural (Haryana 2.1%,26 Haryana 2.7%,28 Punjab 3.1%30 and Rajasthan 4.3%31). There are significantly increasing trends in urban (r2 = 0.60) as well as rural (r2 = 0.31) populations (fig 1). Analyses of prevalence studies in various decades in India also provide significant information about the absolute number of CHD cases.12 Decadal variations indicate that the adult prevalence has increased in urban areas from about 6.5% in the mid-1960s to 7.0% in 1980, 9.5% in 1990 and 10.5% in 2000, while in rural areas it increased from 2% in 1970s to 2.5% in 1980, 4% in 1990 and 4.5% in 2000. This would translate into 4.5 million urban subjects in 1970, 5.6 million in 1980, 9.7 million in 1990 and 14.1 million in the year 2000, and in rural populations into 4.1 million in 1970, 6.4 million in 1980, 11.8 million in 1990 and 15.7 million in 2000.12 Thus, epidemiological studies show that there are at present 29.8 million patients with CHD in this country, which is similar to the numbers obtained by the National Commission on Macroeconomics and Health.35 As epidemiological studies do not identify patients with silent and asymptomatic CHD the actual number of cases may be much greater.

Figure 1 Coronary heart disease (CHD) prevalence (%) in Indian urban and rural subjects aged >30 years as reported in epidemiological studies. The diagnostic criteria used were either known CHD, or angina on the Rose questionnaire or electrocardiographic Q/ST/T wave changes. There is a significantly increasing trend of CHD in both urban and rural populations.


According to recent WHO estimates, developing countries account for 85% of deaths from stroke world wide.1 Stroke death rates per 100 000 in selected countries among people aged 30–69 years is high in developing countries (Nigeria 122, Tanzania 118, India 100, China 96, Pakistan 84 and Brazil 82) as compared with <20 in UK and Canada.36 Disability due to strokes in these countries is almost seven times that in developed countries.36 The Global Burden of Diseases Study reports that stroke is common in India, although stroke mortality is currently lower than in China (table 1).10 Prevalence data for stroke in India are scanty and the available studies suffer from multiple biases that are similar to those for CHD.13 A low prevalence in community studies might also be due to low survival after an acute event, and many cases of transient ischaemic attacks and complete recovery from strokes may be missed on inquiry hence these prevalence rates might be underestimates. Table 3 shows the major epidemiological studies in India that determined the community stroke prevalence. The crude prevalence rates of stroke appear to be higher in urban populations than in rural subjects, but there are location-based differences as seen by a very high prevalence among Parsis in Mumbai (842/100 000)46 as compared with the Mumbai general population (220/100 000)38 Evaluation of secular trends in stroke in India is not possible owing to the small numbers of studies.

Table 3 Stroke prevalence studies in India

The incidence of stroke has been reported by a study from West Bengal in India and was significantly greater in rural subjects than in the urban population.47 In a 5-year prospective study among 20 842 rural subjects the age-adjusted incidence rate was 262/100 000 a year,48 while in a study among 50 291 urban subjects the incidence was 105/100 000.39 These studies have excluded deaths from stroke and the data are likely to be underestimates. The Global Burden of Diseases Study reported an estimated population-based annual stroke incidence in India of 89/100 000 in 2005, which is projected to increase in 2015 to 91/100 000 and in 2030 to 98/100 000.10 This is compounded by high stroke mortality in India. In hospital-based studies 30-day stroke mortality in Mumbai was 32% in 1963–8, which declined to 12% in 1978–82,49 but still remains significantly greater than in the developed countries where short-term mortality rates vary from 5% to 10%.36

In most parts of the world about 70% of strokes are due to ischaemia, of which 25% are cardioembolic, 27% are haemorrhagic and 3% are of unknown cause.36 An important difference in stroke epidemiology in developing countries is the greater occurrence of haemorrhagic strokes. In China 17.1–39.4% strokes are due to intracerebral haemorrhage. In other low- and middle-income countries the incidence of haemorrhagic stroke varies from 29% to 57% in Africa, 19% to 46% in Asia and 19% to 43% in South America compared with 10% to 20% in North America and Western Europe.49 Only a few studies in India have reported statistics such as these. A population-based study reported ischaemic stroke in 68% and haemorrhagic in 32%.50 This is consistent with data in other developing countries, although more and larger studies are needed to determine trends in stroke subtypes in India.


There have been multiple hypotheses to explain the increase in cardiovascular diseases in India. Studies in emigrants indicated that South Asians had higher rates of CHD, but no higher rates of risk factors than the local population.51 Genetic factors were implicated. These studies suffered from multiple biases, the major being the “healthy survivor” bias, as survivors of acute coronary event that reached these hospitals were younger, more educated, affluent and had risk factors that were not considered significant with the available knowledge, thus perpetuating the allegory that there was something unique among the Indians. Most important was that these studies did not study the relationship between the risk factors and CHD and, moreover, did not measure several of the risk factors.

Epidemiological studies and risk factors

There is a strong correlation between urbanisation and increase in the risk of cardiovascular disease in Indian subjects. Coronary risk factors that are more prevalent in Indian urban subjects may explain their excess cardiovascular disease.12 Padmavati reported that the prevalence of CHD was significantly greater in urban subjects in Delhi than in rural subjects around Delhi.15 They also reported that coronary risk factors such as hypertension and high cholesterol levels were more prevalent in the urban subjects. Similar urban–rural differences were reported by Gupta and Malhotra,17 Chadha et al18 and the Indian Council of Medical Research studies52 who evaluated urban–rural risk factor differences. All these studies reported that multiple lifestyle factors (sedentariness, dietary calorie and fat intake) as well as physiological factors (weight, body mass index (BMI), waist–hip ratio (WHR), blood pressure, high total and low-density lipoprotein (LDL) cholesterol, low high-density lipoprotein (HDL) cholesterol, total/HDL cholesterol ratio, triglycerides and diabetes) were significantly more prevalent in the urban populations. These studies also noted that smoking, which is an established cardiovascular risk factor, was more common in rural subjects and therefore could not provide conclusive guidance.

In the mid-1990s we systematically compared the prevalence of multiple coronary risk factors in rural and urban subjects in Rajasthan and also correlated these risk factors with the prevalence of CHD.23 31 We found that the prevalence of risk factors was similar to those found in the studies reported above. We then determined the importance of various risk factors in CHD using logistic regression and reported that smoking was an independent risk factor in both rural and urban subjects. Other major risk factors were obesity, high WHR, hypertension and lipid abnormalities. This showed that standard cardiovascular risk factors were important in Indians. Emigrant versus native South Asian case–control studies have reported a greater prevalence of risk factors such as diabetes, impaired glucose tolerance and other lipid abnormalities to explain the greater prevalence of CHD in emigrant South Asians.5355

The prevalence of stroke seems to be similar in urban and rural populations, but only a limited number of comparable studies exist. Smoking and hypertension are well-known stroke risk factors and explain the greater stroke incidence among the Chinese population.36 It is well known that smoking and tobacco use is greater among the rural populations in India.56 Recent studies have reported a high prevalence of hypertension in rural subjects in different parts of the country.57 Thus, stroke in rural populations may be explained by these standard risk factors, although results of continuing studies using either a cross-sectional design (eg, INTERSTROKE Study)50 or a prospective design (namely, PURE Study)9 are awaited to evaluate accurately stroke determinants in India.

CHD risk factors in case–control studies

Hospital-based case–control studies in the 1960s and 1970s reported that smoking or tobacco use, hypertension, diabetes and cholesterol levels were significantly greater in subjects with CHD.58 Subsequent case–control studies reported that other lipid abnormalities such as high LDL cholesterol and triglycerides, and low HDL cholesterol were also important.5963 Abnormalities of the apolipoprotein system, including high apolipoprotein B100, high lipoprotein(a) and low apolipoprotein A-1 levels, in patients with CHD as compared with controls have also been reported.62 63 A high prevalence of abnormalities of coagulation and platelet functions has also been reported.64 Ethnic susceptibility to CHD among South Asians was initially observed among people who had emigrated to the West Indies, East Indies and Britain.65 McKeigue et al reported that an increased WHR and abnormalities of glucose and lipid metabolisms were major coronary risk factors in South Asians in Britain.53 It was reported that these subjects had a greater WHR and this was associated with a conglomeration of metabolic risk factors—peripheral insulin resistance and hyperinsulinaemia, hypertriglyceridaemia, low HDL cholesterol, borderline raised LDL cholesterol and type 2 diabetes.53 6567

Pais et al studied 200 cases of first myocardial infarction in Bangalore and compared the risk factor profile with that of 200 age-matched controls.68 The most important predictor of acute myocardial infarction was current smoking of both cigarettes and bidis, followed by a history of hypertension and diabetes. Other important risk factors were increased fasting glucose and a high WHR.69 Rastogi et al performed a multicentre case–control study in Delhi and Bangalore to identify important coronary risk factors and reported similar findings.70 71 Cigarette or bidi smoking, BMI >25 kg/m2, WHR >1.0, and a history of hypertension, high cholesterol and diabetes were important risk factors. An inverse association of CHD risk with dietary intake of vegetables, green leafy vegetables, cereals and mustard oil was noted. Patil et al performed a case–control study to identify risk factors for acute myocardial infarction in a rural population of Central India.72 One hundred and eleven consecutive cases of acute myocardial infarction were recruited and compared with 222 controls matched for age and sex. Important risk factors were smoking, raised fasting glucose and a high WHR. Smoking has been identified as the most important risk factor among the young survivors of myocardial infarction in many studies.58

The INTERHEART study was a large case–control study performed in 52 countries of the world with 15 152 cases of first myocardial infarction and 14 820 controls.73 This study identified nine well-known coronary risk factors—abnormal lipids, smoking, hypertension, diabetes, a high WHR, psychosocial factors, low fruit and vegetable consumption, low alcohol consumption, and lack of physical activity—as accounting for more that 90% cases of acute myocardial infarction world wide. An important finding of this study was a younger age of occurrence of acute myocardial infarction in South Asians.74 The mean (SD) age of occurrence of a first myocardial infarction among 1732 participants from South Asia was 53.0 (11.4) years as against 58.8 (12.2) years in other countries. Although the mean age of myocardial infarction was lower in South Asians than in subjects from other countries, the risk factors were similar. Before adjustment for the nine INTERHEART risk factors there was a higher probability of patients who were younger than 40 years in the South Asian group (p = 0.001), but after adjustment for these risk factors the difference was attenuated and not significant (p = 0.27) (fig 2).

Figure 2 Proportionate probability of acute myocardial infarction in South Asian (SA) subjects as compared with subjects from rest of the world (ROW). There is greater probability of acute myocardial infarction in young South Asian subjects as compared with people from other countries (p = 0.001). However, the difference disappears when the probability is adjusted for the nine INTERHEART risk factors (p = 0.27). Adapted from Joshi et al.74

The risk factors that were found to be important in the overall INTERHEART cohort were also important in the South Asian cohorts (table 4).74 Some harmful risk factors were more common in South Asians (raised Apo B/Apo A-1 ratio and diabetes) and all the risk factors occurred at a younger age in this group. Bidi smoking as well as use of non-smoked tobacco, which is peculiar to the Indian subcontinent, also emerged as an important risk factor.74 In the INTERHEART Study bidi smoking was associated with age and sex adjusted odds ratio (OR) of 2.89 (95% confidence interval (CI) 2.11 to 3.96) in the overall cohort and was same in South Asians (OR = 2.73, 95% CI 1.90 to 3.92).75 Participants who were current smokers of cigarettes or bidis and who chewed tobacco had an OR of 4.09 (95% CI 2.98 to 5.61) while those who chewed tobacco were also at increased risk (OR = 2.23, 95% CI 1.41 to 3,52). These data are especially relevant for India where tobacco smoking and chewing are widely prevalent both in men and women.76

Table 4 Risk factors for acute myocardial infarction in South Asians in the INTERHEART Study74

These case–control studies have potential biases, but appear plausible given that their findings are consistent among themselves and also with the findings of cohort studies conducted in Western populations. Prospective cohort studies in South Asian populations are highly desirable. Cohort studies also have their limitations—for example, they are not good methods for studying acute precipitators such as stress and variations in environment due to air pollution or temperature. On the other hand, cohort studies are useful in studying the effects of chronic exposures, assuming that one or a few measures accurately characterise the exposure. We believe that a carefully conducted case–control study is just as valuable as a cohort study. The population attributable risk calculated in a case–control study is an estimate of the proportion of cases that would be avoided if the exposure is removed. This statistic (as in the INTERHEART Study) combines information about the strength of association of the relative risk with information about the prevalence of exposure, and is important to estimate modifiable environmental exposures. Additionally, it is difficult to include enough cases of young patients with a myocardial infarction in a cohort study or in other populations where the incidence of the event is low. Continuing prospective studies such as the PURE Study9 involving 28 500 people from five locations in India and an additional 4000 subjects from Pakistan and Bangladesh (n = 32 500) is likely to provide more important information.

Stroke risk factors

Stroke is a non-homogeneous condition caused by either extracranial/intracranial vascular atherothrombotic disease or intracranial haemorrhagic lesions.77 Differences in cardiovascular risk factors can partly influence the pathological types of stroke as well as subtypes of ischaemic stroke across populations. Risk factors for each type of stroke differ—atherosclerosis risk factors (as in CHD) predominate in the former, whereas hypertension and smoking are the known leading causes of the latter. Leading stroke risk factors are raised blood pressure, smoking, high cholesterol, low fruit and vegetable intake, physical inactivity and alcohol excess.78 However, there is evidence that these risk factors are associated with differing attributable risks for stroke as compared with CHD, and some established CHD risk factors such as high cholesterol are of uncertain importance for stroke.50

Hypertension is an important and consistent conventional risk factor for stroke. The Prospective Studies Collaboration meta-analysis of over one million subjects79 reported a log-linear relationship of blood pressure and stroke. An Asia-Pacific regional systematic review on blood pressure and stroke correlation also reported similar results.80 Correlation of hypercholesterolaemia with stroke is not clear and a meta-analysis of 45 prospective cohort studies reported no significant association between total cholesterol and ischaemic stroke, although a trend was evident.81 Trials of LDL cholesterol lowering with statins report a 20% relative risk reduction for stroke, which is much less than for CHD (>50%). Diabetes has been reported to be an independent risk factor for ischaemic stroke.

Lifestyle factors such as smoking and tobacco use have been reported to be important for both ischaemic and haemorrhagic strokes.50 The relationship between excessive salt intake and hypertension was established by the INTERSALT Study, but relationship between salt intake and stroke is variable.82 In countries with a high salt intake there is a clear association, whereas in countries with a low sodium intake it is variable.50 Other dietary factors, such as fruit and vegetable consumption, are associated with a reduced risk of stroke and excessive fried foods and fat intake with an increased risk.77 Excessive alcohol intake is also a risk factor for stroke. Folate supplementation has been reported to be associated with a reduced risk of stroke in a meta-analysis.83 Regular physical activity is also protective. No population-based prospective studies exist in India, and data on stroke risk factors still rely on small case–control studies.38 49 In certain geographical areas socioeconomic factors, ethnicity and race are important risk factors for stroke.78

It is also important to realise that specific infections common in India may also contribute to stroke. These infections include malaria, neurocysticercosis, leptospirosis and viral haemorrhagic fevers. Conditions such as sickle cell anaemia, and snake bites are other prevalent disorders that can produce a stroke-like picture owing to intracerebral haemorrhage.84 Cerebral venous sinus thrombosis is a common cause of stroke among women in India, especially during the postpartum period.85

In India only small case–control studies to identify risk factors have been performed. Dalal reported that diabetes, hypertension, tobacco use and low haemoglobin levels were important risk factors in a study in Mumbai.38 A study in urban subjects in Kolkata reported that hypertension was the most important risk factor, whereas another study in West Bengal rural subjects reported that existing heart disease, hypertension and smoking were important.47 All these studies were small and larger studies that are currently in progress such as the INTERSTROKE Study50 should be able to identify risk factors more accurately.


Major risk factors dependent on the demographic and societal transition are physical inactivity, excess dietary calories and fat intake, smoking and tobacco use, being overweight and obesity, high blood pressure, diabetes, cholesterol levels, the metabolic syndrome and psychosocial stress. There is epidemiological evidence that the levels of many of these risk factors are increasing in India. The rapid urbanisation and globalisation is increasing mechanisation leading to sedentariness and this, coupled with recently found affluence, is increasing tobacco, calorie and fat intake, leading to increases in weight, glucose, blood pressure and unfavourable lipid profiles.


Smoking and tobacco use became a fashion statement in the past century in India. This phase was characterised by a large increase in its consumption driven by prevalent attitudes and cultural changes. The British studies and US Surgeon General’s reports on the harm of smoking and tobacco use led to a gradual decline in smoking in developed countries. On the other hand, populations of low- and middle-income countries have been increasing their cigarette consumption since about 1970.86 In these countries the per capita annual consumption of cigarettes increased from 800/year to more than 1200 from 1970–2 to 1990–2. In India, tobacco use increased by 36% over the same period.87 According to World Development Report, in India, the per capita tobacco consumption was 0.7/kg/year in 1974–6 and 0.8/kg/year in 1980 and was projected to increase to 0.9 kg/year by 2000 in contrast to developed countries where it was poised to decline from 2.9/kg/year in 1974 to 1.8 in 2000.88

The prevalence of smoking and tobacco use varies in different regions of India. In the second National Family Health Survey high smoking and tobacco use was reported among men and women in northeastern and northern Indian states while a low prevalence was observed in Punjab and Maharashtra.56 The habit of smoking is more prevalent in rural subjects (30–60%) than in the urban population (10–30%).76 Although smoking is low among women, use of non-smoked tobacco is high.76 There is an urgent need to curb the tobacco epidemic in India. Important policy steps in this direction are faithful implementation of the WHO Framework Convention for Tobacco Control initiative in India and strict application of the Indian Tobacco Control Act.76 Most of the recent Indian epidemiological studies have reported an inverse association between smoking/tobacco use and educational status,56 89 and it seems that improving the literacy levels of the population is an efficient method to decrease tobacco consumption.

Dietary changes and obesity

Per capita consumption of major fats and oils has increased in India in the past 30 years.90 In 1958 it was 5.62 kg a year which increased to 5.79 in 1961, 5.23 in 1966, 5.85 in 1971, 5.21 in 1976, 6.48 in 1981 and 6.97 in 1986. This consumption is much lower than in European Union countries (38.98), USA (39.72), Canada (34.83) and Japan (19.84). Consumption of oils and hydrogenated oils is increasing similarly. This may be associated with increasing obesity, although no national data are available. The Jaipur Heart Watch (JHW) studies reported significantly escalating trends in obesity and high WHR in an Indian urban population. These studies were performed in 1992–4 (JHW-1),20 1999–2001 (JHW-2),23 2003–4 (JHW-3)91 and 2005–6 (JHW-4).92 Age-adjusted prevalence in subjects aged 20–59 years shows that obesity (BMI ⩾25 kg/m2 and high WHR >0.95) has increased significantly (fig 3). There was a significant correlation of increasing socioeconomic status (literacy levels) with increasing obesity and truncal obesity.93 These studies show that in India increasing socioeconomic status is associated with increasing obesity in contrast to high- and middle-income countries, where poverty is associated with greater obesity.94

Figure 3 Trends in age-adjusted prevalence of various risk factors in Jaipur Heart Watch (JHW) studies among urban subjects aged 20–59 years in India. These studies were performed in 1992–4 (JHW-1),20 1999–2001 (JHW-2),23 2003–4 (JHW-3)91 and 2005–6 (JHW-4).92 There are significantly escalating trends in prevalence of obesity (BMI ⩾25 kg/m2), high WHR (>0.95), hypertension and hypercholesterolaemia (cholesterol ⩾5.10 mmol/l) in both men and women.


A review of studies of the prevalence of hypertension in India has shown a high prevalence in both urban and rural areas.95 Indian urban population-based studies using WHO guidelines for diagnosis (known hypertension or blood pressure ⩾160 mm Hg systolic or 95 mm Hg diastolic, or both) have shown increasing hypertension among adults aged ⩾20 years from about 5% in the 1960–70s to 11–15% in the late 1990s.57 The prevalence of hypertension using recent criteria (blood pressure ⩾140/⩾90 mm Hg) has been reported among some urban Indian populations. Gupta et al reported hypertension in Jaipur in 30% of men and 33% of women,20 Joseph et al reported it in 31% of men and 41% of women in Trivandrum,96 whereas Mohan et al reported a crude prevalence rate of 21% in Chennai.97 In Mumbai, Anand reported hypertension in 34% of middle-class executives,98 whereas Gupta et al reported hypertension in 44% of men and 45% of women in Mumbai.99 We determined trends in the age-adjusted prevalence of hypertension in JHW studies in subjects aged 20–59 years (fig 3). A significant escalation of hypertension prevalence was seen in both men and women. These findings are in accord with those of many developed countries where it has been reported that, at any one time, about half of the population have high blood pressure.100

Lipid levels and dyslipidaemias

The epidemiological studies of cholesterol measurement in India are hampered by the lack of uniform assay that has resulted in large variations. However, in recent studies cholesterol levels measured by enzyme-based assays have shown a progressive increase. In 1982, Gandhi et al101 estimated cholesterol lipoproteins levels in 201 urban Delhi subjects and reported mean (SD) serum cholesterol of 160 (29) mg/dl (4.10 (0.75) mmol/l) in men and 150 (25) mg/dl (3.90 (0.65) mmol/l) in women.101 In a more recent study Reddy et al reported mean (SD) cholesterol levels of 177 (40) mg/dl (4.60 (1.05) mmol/l) in men and 176 (40) mg/dl (4.55 (1.05) mmol/l) in women in Delhi.102 Pradeepa et al have demonstrated that even within an urban environment, there appears to be a significant difference in the lipid levels, dyslipidaemia being more common and severe in the middle-income group than in the low-income group.103 High levels of small dense LDL are more important risk markers, and a study showed greater levels among emigrant South Asians than in US-based Caucasians.64 The INTERHEART Study reported that the ratio of apolipoprotein (apo) B to apo A-1 was an important risk marker for acute myocardial infarction.73 Higher levels were seen in South Asian cases (61.5%) than in subjects from other countries (48.3%).74 Limited information exists about the changing time trends in prevalence of dyslipidaemia in Indian subjects from methodologically comparable studies. The Jaipur Heart Watch studies reported that there is a significant increase in total cholesterol, LDL cholesterol and triglycerides and a decline in HDL cholesterol in both men and women at all age groups.93 Age-adjusted prevalence of high total cholesterol (⩾5.17 mmol/l) among subjects aged 20–59 years shows increasing trends in both men and women (fig 3). The increase in total cholesterol in urban Indians contrasts with the falling mean cholesterol level in the USA.2


The first systematic investigation of diabetes in India was performed by the Indian Council of Medical Research Task Force on diabetes in the 1970s.104 The population aged >14 years was screened using a post 50 g glucose load, and capillary blood glucose >4.40 mmol/l (>170 mg/dl) was taken to show the presence of diabetes. A total of 34 194 subjects were screened and the prevalence of diabetes was 2.1% in urban subjects and 1.5% in rural populations.

Recent studies in large cities in northern and southern India report that the prevalence of diabetes among adults (⩾20 years) ranges from 8% to 15%.105 106 From Chennai a significant increase in the prevalence of type 2 diabetes among adults has been reported (fig 4).107 When similar diagnostic criteria (known diabetes and/or fasting and post-glucose load hyperglycaemia) were used, the prevalence of age-adjusted diabetes among adults in urban Chennai increased from 8.3% in 1988–9, to 11.6% in 1994–5, 13.5% in 2000 and 14.3% in 2003–4. An increase in the prevalence of diabetes has also been reported from rural Tamilnadu.108 Secular trends in the prevalence of diabetes show that there is a slow increase in Indian rural subjects as compared with the urban population.105

Figure 4 Increasing age-adjusted prevalence (%) of diabetes and impaired glucose tolerance in Chennai in South India.107

Although the prevalence of diabetes is low in rural populations (2–4%),105 there is evidence of a high burden of impaired glucose tolerance.106 Studies from rural Tamilnadu in the early 1990s showed a low prevalence of diabetes and a high prevalence of impaired glucose tolerance.108 Over time, while the prevalence of diabetes increased that of impaired glucose tolerance declined associated with increasing obesity. This study shows that there is a significant pool of potentially diabetic subjects in rural Indian populations and rapid lifestyle changes can lead to diabetes. The high prevalence of diabetes reported from rural Andhra Pradesh in a recent study109 seems to confirm this observation.

Other factors

Many small studies from India have evaluated unconventional cardiovascular risk factors, including lipid subfractions, platelet functional abnormalities, inflammatory markers, homocysteine and thrombotic factors. Larger studies are needed. Barker’s hypothesis that focuses on the adverse long-term cardiovascular impact of fetal undernutrition and low birth weight has been proposed as a cardiovascular risk factor.110 Studies have reported that neonates with a low birth weight have higher insulin levels and insulin resistance than normal weight neonates.111 This trend persists into early and mid-childhood.112 113 In the UK it has been observed that South Asian children have a lower birth weight and greater insulin resistance than Caucasians.114 The long-term prognostic impact of this finding has been studied in the New Delhi Birth Cohort Study, which showed that an increase in birth weight early in life resulted in greater insulin resistance and more diabetes in low birth weight children at 30 years of age.115 More studies are needed to evaluate the long-term cardiovascular consequences of this finding.

Cardiovascular risk factor antecedents at childhood and adolescence have been sparsely studied in India.116 In the Indian cohort of Global Youth Tobacco Survey a variable prevalence of tobacco consumption was reported in different regions, with high use in eastern and central Indian states.117 From Delhi it has been reported that smoking among children is rapidly increasing, although a study reported that within a period of 2 years the habit declined.118 This was attributed to better health awareness and tobacco control initiatives. In affluent Indian children there seems to be an epidemic of obesity while among children of low socioeconomic status low body weight is widely prevalent.119 A high prevalence of high blood pressure, high cholesterol levels and high glucose levels in children has been reported from a few centres while many centres report a low prevalence of these conditions.120 121 The role of genes and gene–environment interactions in the pathogenesis of cardiovascular diseases also need to be studied.


The increasing burden of coronary heart disease and stroke emphasises the importance of containing the epidemic of cardiovascular disease in India as well as combating its impact and minimising its toll.122 The INTERHEART Study has conclusively demonstrated that the usual cardiovascular risk factors are important in Indians.73 Primordial, primary as well as secondary prevention efforts should be directed towards these factors. At a macrolevel, better social, economic and cultural status correlates inversely with lifestyle factors of smoking, abnormal food patterns and exercise and is recommended for primordial prevention. Public broadcasting systems, television and newspapers have an important role in the dissemination of health-related information among populations. Suitable strategies to impart information to these print and electronic media should be developed locally. In Indians greater literacy levels and awareness have led to a decrease in some cardiovascular risk factors—for example, smoking, in the well educated.89 Public awareness and demand has led to an increase in the number of physical activity centres such as parks and walkways in many urban locations. A new public health education campaign focusing on lifestyle changes (increased physical activity, prudent diet and tobacco cessation) has been launched in Chennai and preliminary results are encouraging.123

The measures that have the greatest impact on population-based prevention are policy initiatives. The initiatives that might have an impact even in the short term include an increase in tobacco taxes, economic and labelling disincentives for unhealthy foods, removal of trans fats especially in processed foods, reduction in salt in home-made and processed foods, and urban design features which promote safe and pleasurable physical activity. These are more likely to have a much greater impact than limited community initiatives which will take a long time to develop and are resource intensive. In India some of these policy changes are already progressing and tobacco legislation has been approved by parliament. Some other national initiatives for prevention and control of cardiovascular diseases are being planned.122

A high-risk intervention approach is also important. The effectiveness of this depends directly on practising doctors and healthcare workers. Doctors and other healthcare providers serving the populations need to be educated as to the specific cultural barriers and opportunities that would encourage implementation of CHD prevention practices. The Indian Council of Medical Research has developed a chronic diseases risk factor surveillance programme for India.124 Pilot surveys have been carried out in various locations in the country using the WHO-Steps methodology, which outlines the sequential measurement of behavioural, physical and biochemical risk factors through its core, expanded and optional modules.125 Preliminary risk factor data in the 15–64-year age group indicate a high rate of multiple cardiovascular risk factors, with a greater prevalence of risk factors in respondents residing in urban areas than in those residing in slum/peri-urban and rural areas.124 Successful adaptation and implementation of the WHO stepwise approach for non-communicable disease risk factor surveillance has paved the way for developing a module for risk factor surveillance in the National Integrated Disease Surveillance Programme (IDSP).124

A few primary prevention efforts have focused on individual risk factors. The Indian Diabetes Prevention Programme (IDPP) evaluated increased physical activity versus metformin for prevention of diabetes in subjects with impaired glucose tolerance.126 At a median follow-up of 30 months, lifestyle modification reduced progression to diabetes by 28.5%, which was same as with metformin (26.4%) and also with combined metformin and lifestyle changes (28.2%). A yoga-based comprehensive lifestyle change has been evaluated in primary and secondary prevention of CHD risks and preliminary results are encouraging.127 For comprehensive cardiovascular primary prevention, The Indian Polycap Study (TIPS) is evaluating the usefulness of various drug combinations for reduction of cardiovascular risks.128 Secondary prevention practices in Indian are woefully inadequate both in acute care settings and in long-tem medical care facilities.129 The influence of the appropriate use of evidence-based treatments on outcomes in patients with CHD or stroke is not known in India. The Heart Outcomes Prevention Evaluation-3 (HOPE-3) Trial will evaluate the effect of statins and/or angiotensin receptor blockers on cardiovascular outcomes including a large sample from India.130 Some national initiatives for prevention and control of cardiovascular diseases in India are being planned.122

In conclusion, cardiovascular diseases, especially CHD and stroke, are major health problems in India. Risk factors for these conditions in Indian subjects are similar to those in populations elsewhere in the world and seem to be escalating. It is imperative that primordial, primary and secondary prevention efforts that have been shown to be effective in other regions of the world131 are translated into active regional and national initiatives in this country.


View Abstract


  • Competing interests: None declared.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.