The following article in The Hindu is written by Sonalde Desai a Senior Fellow at NCAER & Professor of Sociology at University of Maryland and Debasis Barik an Associate Fellow at NCAER. The article features important evidence gathered from the India Human Development Survey (IHDS) conducted periodically by NCAER in collaboration with the University of Maryland. IHDS is India’s only national longitudinal panel data set and provides considerable opportunities for researchers to generate high quality relevant evidence to inform sound policymaking.
Recently a parliamentary committee declined to extend the size of health warnings on cigarette packets due to lack of independent evidence on the health impacts of smoking on the Indian population. A longitudinal study conducted by the National Council of Applied Economics (NCAER) and University of Maryland shows that in India too smoking kills.
The India Human Development Survey (IHDS) was first conducted in 2004-05. In this survey 41554 households were surveyed in both urban and rural areas in all States and Union Territories with the exception of Andaman-Nicobar and Lakshadweep. At this time extensive information about the lifestyles of over two lakh individuals residing in these households was collected. In 2011-12 these same households were surveyed again. We were able to re-interview about 83 per cent of the original households. At the time of the re-interview information on current location of the individuals from the original household was obtained including whether they are still alive. Thus we have access to a prospective data set which contains both information on smoking tobacco products and whether the individual has died in the seven years between the two interviews. The results unambiguously show that even after we take into account individuals’ age gender education and household wealth those who are reported to be daily smokers are more likely to die.
In the initial interview 26 per cent men and 1.6 per cent women above the age of 15 smoked. These statistics are very similar to those observed in the Global Adult Tobacco Use Study by Professor Ram and his colleagues at International Institute of Population Sciences conducted on behalf of the Ministry of Health and Family Welfare in 2009-10. The GATS study also noted that 24.3 per cent of men and 2.9 per cent of women aged 15 and above smoked.
Categorising smokers
Since smoking is often underestimated for younger people when smokers tend to hide their habits from the older family members we focus on individuals who are 30 and above. In this age group nearly 36 per cent men in the IHDS sample smoke; 29 per cent smoke daily. Among women only 2.5 per cent smoke. Although men from all walks of life smoke smoking is disproportionately concentrated among Dalit Adivasi and Muslim men. Among this group about 45 per cent men smoke; 37 per cent smoke daily. Lack of education also plays a role. About 46 per cent of illiterate men smoke while only 16 per cent of the college graduates smoke. Moreover smoking is concentrated among the lowest income group. Nearly 46 per cent of the men in bottom fifth smoke compared to only 20 per cent in the upper fifth. Death rates are higher for daily smokers than for non-smokers or occasional smokers. About 11.3 per cent of men aged 30 and above and who smoke daily died in the seven years following our initial survey; only 10.2 per cent of the non-smokers and occasional smokers died. However as we noted above smokers come from lower socio-economic strata. Hence it is difficult to know if these characteristics rather than smoking may be the cause of higher death rates among smokers. So we compare like with like and control for education marital status age caste/religious background urban/rural residence state of residence and whether the individual was employed at the first interview. We also control for household wealth. This does not change the relationship observed above.
Even after taking into account all these differences we find that smokers have a higher death rate than non-smokers. Among men daily smokers are 1.14 times as likely to die between the two interviews as the non-smokers and occasional smokers. Lest this seem like a small difference the improvement in survival by giving up smoking would be more than by difference between illiterate and those with eight years of education or between men living in Uttar Pradesh and Karnataka. For women smoking is even more harmful but given the very small number of women who smoke this impact is not significant.
Delaying anti-smoking campaigns will take a heavy toll on the Indian population particularly poorer and less educated men. What is disturbing about the current debate is the message that it sends to current and potential smokers. Whether the pictorial warning covers 20 or 30 per cent of the package is less important than the implication that Indian population’s risk profile is somehow different from that of non Indians and that until a study has been conducted in India we should not believe that smoking increases health risks in India. Nonetheless the results we show above should lay to rest the argument that Indians are not somehow immune to health consequences of smoking that beset non-Indian populations.
Debate in other countries
This debate is reminiscent of similar battles fought in other countries. In the U.S. for decades cigarette companies tried to throw a smoke screen over research results that unambiguously showed that smoking caused cancer and increased mortality. In France even in the late 20th century the legislature argued that controlling public smoking was pitting non-smokers’ rights against smokers’ rights. Financial interests played an important role in the French debate too. Constance Nathanson notes that by 1990 French tobacco market had been captured by multinationals leading a smoking proponent to grumble that anti-smoking advertising would selectively weaken French tobacco industry and “there will no less smoking or drinking in sweet France but smoking and drinking will be less French and more American.” These delays in anti-smoking legislations have led to slower decline in smoking in France than in other high income countries; as World Atlas of Smoking shows today 34 per cent French men smoke compared to 23 per cent in neighbouring Switzerland.
Let us not give mixed messages to our young men and increasingly young women; smoking is not harmless smoking is not cool. Smoking kills even in India.
Published in: The Hindu May 4 2015