A major long-term study of half a million adults in Chennai, India, has found that social inequality remains a powerful predictor of premature death, even among people who neither smoke nor drink alcohol. The study, conducted by researchers at the Epidemiological Research Centre in Chennai in collaboration with Oxford Population Health, also shows that tobacco and alcohol use substantially worsen these inequalities among men. The study is published in Nature Health.
The study followed 458,609 men and women aged 35 and over, recruited between 1998 and 2001 and tracked until 2020. Researchers examined how education level (a key marker of social inequality) alongside smoking and alcohol use, influenced the likelihood of dying prematurely (before age 70).
Over the study period, more than 52,000 participants died prematurely. Among people who neither smoked nor drank alcohol, there were strong social inequalities in mortality. People with no formal education had nearly three times the risk of dying in middle age compared with those who had more than 11 years of schooling. This pattern was consistent in both men and women.
Dr Vendhan Gajalakshmi, co-first author and principal investigator of the study at the Epidemiological Research Centre in Chennai, said:
These findings show that social disadvantage alone is a major determinant of early death, even in the absence of smoking and alcohol.
However, among men, smoking and alcohol use were common and further widened these inequalities. Nearly two-thirds (65%) of men with no schooling reported smoking or drinking, compared with just 21% of those with the highest education levels. On average, male smokers began in early adulthood and smoked around 11 cigarettes or bidis (small, hand-rolled cigarettes of tobacco wrapped in a dried leaf) per day, while those who drank consumed the equivalent of 19 units of alcohol per week.
When the combined effects of education, smoking, and alcohol use were analysed, both habits were found to independently increase mortality risk. Men who smoked had about 25% higher risk of death than those who never smoked, while those who drank alcohol had about 50% higher risk compared with non-drinkers. Men who both smoked and drank had nearly double the risk of dying before age 70 compared with those who did neither. Importantly, the study estimates that about half of all deaths in middle age among men who both smoked and drank can be attributed directly to these habits.
The findings highlight the dual impact of social and behavioural factors on health. While reducing smoking and alcohol use could substantially lower mortality, the persistence of large inequalities among non-smokers and non-drinkers underscores the need to address broader social determinants of health.
Sarah Lewington, Professor of Epidemiology and Medical Statistics at Oxford Population Health and senior author of the study, said:
This research shows that tackling tobacco and alcohol use is essential, but not sufficient. Reducing social inequalities must also be a central part of strategies to prevent premature deaths.
The study provides some of the most detailed evidence to date from a large urban Indian population and has important implications for public health policy, particularly in low- and middle-income countries facing rising burdens of non-communicable diseases.
Professor Prabhat Jha, Head of Department at Oxford Population Health and Principal Investigator for the Indian Million Deaths Study said:
This landmark, 20-year prospective study - the largest of its kind yet done in India - demonstrates the immense value of long-term tracking. It provides vital insights into the social and behavioural drivers of premature mortality in a globally significant nation.