Preventing premature mortality in chronic diseases for South Asians in the UK and beyond

Societies across the world experience different rates of diseases, and within these societies, inequalities exist between social class, gender and ethnicity. Starting around the 1960s, 1 studies began to report differences in coronary disease rates of specific ethnic groups. With advancing economic development, market economy countries attracted migrants from poorer nations, who then had clear inequalities in chronic disease patterns. 2 For example, the cardiovascular risk factors 3 and mortality 4 of minority groups, such as those from the South Asian subcontinent, have been worse than that of the majority population.

Mahatma Gandhi famously remarked that ‘the way in which we treat minorities is the measure of civilization in a society’. 5 Though there has been a remarkable decrease in coronary disease mortality in the developed world since its peak in the early 1970s, 6 it is of concern that smaller declines in some minority groups are actually leading to increasing disparities between them and the majority of the population. 7 What has changed in our understanding since ethnic disparities in coronary disease mortality were initially highlighted? We give some past and future pointers in this field by citing examples in the largest ethnic minority group in the UK – people of South Asian origin.

Biological differences are often first sought in trying to explain ethnic differences in disease patterns. The higher prevalence of diabetes in South Asians (predominately comprising people from India, Bangladesh, Pakistan and Sri Lanka) has been long known, and was thought to be the main driver behind the higher rates of coronary heart disease mortality seen in this population when compared to the majority indigenous populations in the developed societies to which South Asians migrate to. Theories to explain this have historically revolved around insulin resistance. 8

The thrifty gene hypothesis additionally proposes that conditions of food scarcity increased evolutionary selection pressure for efficient metabolism, which then disadvantages those who become exposed to the modern obesogenic environment, the supposed combination facing the migrant from South Asia. The thrifty phenotype hypothesis is a parallel concept, based on fetal under-nutrition rather than genotype, linking low birthweight to later increased coronary risk. A novel explanation proposes that South Asians have a smaller superficial subcutaneous adipose tissue compartment than white people. 9 As babies born to mothers of South Asian descent are smaller and have less peripheral fat, during subsequent accelerated growth immersed in richer diets, this primary compartment could reach its capacity for fat storage rapidly, in turn leading to adverse consequences related to central obesity and dyslipidaemia. Other hypotheses related to mitochrondrial efficiency and the burden of infectious disease have also been mooted. 10, 11 Thus, progress has been made in eliciting biological differences between ethnic groups that may underlie different disease patterns in them.

However biological factors may not account for all of the excess cardiovascular risk. 12 The social determinants of health may equally explain ethnic differences in disease patterns. 13 South Asians are more likely to live in areas with relative social and economic deprivation, 14 and deprivation has been proposed to being as potent a cardiovascular risk factor as a decade or more in age, or the diagnosis of diabetes. 15 However, when considering social determinants, the heterogeneity of this group requires consideration, being composed of different nationalities and religions (e.g. Indians and Bangladeshis, Muslims and Hindus). 15 Within this group, Bangladeshis are comparatively poor, less educated, and more uninformed about heart disease despite having the highest prevalence rates of smoking among its men. 16 In many respects, coronary disease rates in Indians are closer to the reference ‘white’ population than they are to Bangladeshis: this is likely to be related to socioeconomic factors over and above biological mechanisms. 17 Equally, migrants who came to work in manual jobs in industry in the 1960s will be different to migrants who were doctors or others from professional and mercantile backgrounds, and this in turn will influence the future socioeconomic and educational/professional status of their children. Thus, though the study of social determinants of health is complex, studying social differences within the South Asian group may help to further discriminate aetiological factors for disease.

The future need not necessarily be worse. There is evidence of improvements in the epidemiology of coronary heart disease in ethnic minority groups. For example, case fatality among South Asians, contrary to work in the 1990s, appears to be at least as good, if not better than, whites in larger, better-characterized studies in the last few years. 18 South Asians increasingly appear not be to be disadvantaged with respect to access to invasive coronary investigation and knowledge of diabetes and coronary diseases in ethnic minorities appears to be improving. 19 However, continuing surveillance of these positive changes is required, as well as application of interventions that have led to such improvements being incorporated into healthcare systems more broadly.

Developed countries such as the UK will continue to attract migrant populations. What other research is relevant for the future to ensure that ethnic disparities significantly reduce, or indeed, disappear? Larger, classical epidemiological studies with improved ethnic codification will always be required and are in the pipeline, but some of the answers may lie in trials of culturally sensitive interventions (such as dietary and exercise methods), which can span disparities in language and culture. 20 The application of genetic epidemiology to large prospective cohort studies has already begun to identify novel plausible targets in the pipeline for study in ethnic minority populations. 21

Many of the answers may come from global ‘ethnic majorities’. Large-scale studies are being conducted in the South Asian subcontinent, and these answers will be potentially be applicable both there and in minority ethnic groups in the UK. 22 Endeavours such as registry data collection will also help to make assessments of treatment and outcome. 23 Obviously, large populations have their own patterns of disease – for example, in India there is a strong association between tuberculosis and tobacco smoking deaths. 24 There remains however a relative dearth of prospective studies that include biomedical repositories in South Asia – for example, the Indian Genome Variation Consortium is the largest, involving genotypic analysis of only 15,000 participants, undermining the number of potential hypotheses it can test. 25 Although there are larger cohorts available (such as Chennai and Mumbai), 26, 27 they have not been able to undertake biochemical sampling. Furthermore, follow-up of outcomes in these populations, in particular related to death certification, can be difficult, requiring new approaches, such as verbal autopsies.

By involving populations that have not previously been studied extensively, such studies would allow more reliable assessments of the real importance of various causes of death, rather than extrapolating from populations in developed countries. The range of hypotheses to be studied could be substantially widened if samples (of blood, urine, etc.) were stored for future analyses, which could involve not only factors already of interest but also those that will in the future become of interest. Such analyses are being carried out in a number of ongoing large scale studies throughout the world, for example, in the UK Biobank study and the Kadoorie Study of Chronic Disease in China (both comprising approximately half a million participants). 28, 29 Other, large studies such as INTERHEART have demonstrated that exposure-outcome relationships may be more similar than they are different across global settings, with abnormal lipids, smoking, hypertension, diabetes accounting for most of the risk of myocardial infarction worldwide in both sexes and at all ages in all regions throughout the world although they are limited by their retrospective case control design. 30 Further validation of such work is being undertaken in studies in South Asia such as the Pakistan Risk of Myocardial Infarction (PROMIS). 31 Given that one sixth of global deaths occur in India alone, creating the infrastructure for such prospective studies ideally allied with biomedical repositories in South Asian cohorts is of importance in the future. 32 Such research will also have significant implications for the healthcare of ethnic minority communities in the UK.

Beyond aetiological study, large randomized trials and comprehensive meta-analyses have demonstrated the importance of medical therapies such as aspirin and statin treatments. 33, 34 These medications are now available throughout the world off-patent, but many patients, especially those in poorer nations such as those from the South Asian subcontinent, do not have access to them. They could be combined to create a ‘polypill’ and could potentially be given at low cost (less than $1 per month). It has been proposed to use such a polypill therapy for people without evidence of disease (i.e. primary prevention), although the balance of benefit and risk would be significantly more beneficial for those with evidence of vascular disease (secondary prevention). 35 As well as the developed world, prospective trials are now taking place in India and other South Asian countries, where they may reap the largest rewards. 36 The results of these studies could potentially save thousands of premature deaths (and even more co-morbidity) in the South Asian subcontinent, but these patients are a challenge to enrol in clinical trials and, indeed, implement in real-life practice. 37 Beyond medication, interventions to reduce adverse lifestyle behaviours also require a novel approach in developing countries – to reduce smoking rates for example requires an appreciation of the effects of trade liberalization 38 and global marketing 39 on increased tobacco consumption beyond strategies to target the individual. The WHO Framework Convention on Tobacco Control 40 is an example of a legally binding health treaty that may lead to large benefits in populations and consequently, individuals, if implemented globally.

What has been learned in cardiovascular disease over the last 20 years may equally be applied to other fields of medicine and public health, including cancer and other chronic diseases. 41 Epidemiological transitions within minority ethnic groups are undergoing rapid change in other areas, for example, in excess alcohol consumption. Although countries such as India have a reputation for abstinence, public health effects of alcohol dependence and binge drinking are beginning to emerge. 42 Studies from the UK have demonstrated that Indian populations here are at higher risk of the sequelae of excess alcohol, such as liver cirrhosis, making biomedical and epidemiological research into such areas a priority. 43, 44 They also demonstrate the need to disaggregate different populations in ethnic minority research – for example, Sikhs drink alcohol more than other South Asian groups but when overall rates are averaged out, this fact is hidden. The cultural differences between an Indian Sikh and a Pakistani Muslim are too large to be ignored, with one abhorring alcohol and indulging in tobacco, and vice versa, with resultant chronic diseases at a population level.

The bad news is that ethnic differences in chronic disease, such as that reported in South Asians in the UK, continues to be an area of considerable health inequality. The good news is that research in coronary disease in South Asians in the UK suggests that such health inequalities are potentially reversible. However, a multidisciplinary approach will be required to ascertain the impact of ethnicity and treat disadvantaged groups in other fields. Further research, ideally with large-scale studies from the populations in South Asia itself, are required to allow clinicians to fulfil the ambition of preventing premature mortality due to chronic disease, both in the UK and beyond.

Footnotes

DECLARATIONS

Competing interests None declared

Funding None

Ethical approval Not applicable

Guarantor NB

Contributorship Both authors contributed equally