To the Editor: Gu and colleagues (Jan. 8 issue)1 report thatin 2005, an estimated 673,000 deaths in China were attributableto smoking. The study is of timely importance. But it did notinclude some important variables in the analysis. First, theeffects of family income were not considered. In China, socialdeprivation is a major risk factor for ill health,2 and datafrom a survey about household income and cigarette consumption3and from a study involving low-income employees4 showed thatsmoking was associated with relatively high income. Withoutthis adjustment, the association of mortality with smoking mayhave been attenuated. Second, the analysis did not include passivesmoking. Nonsmokers may have been exposed to passive smoking,leading to the higher mortality in the reference group. Third,the number of deaths in rural China, where 70% of the populationlives, may have been underestimated. It is interesting thatthe authors observed a lower relative risk of death associatedwith smoking in rural areas than in urban areas. In rural areas,persons who never smoked would be poorer and consume less nutritiousfoods than their counterparts who smoked and thus would havea higher mortality, reducing the association.
Ruoling Chen, M.D. University College London London WC1E 6BT, United Kingdom ruoling.chen{at}ucl.ac.uk
References
Gu D, Kelly TN, Wu X, et al. Mortality attributable to smoking in China. N Engl J Med 2009;360:150-159. [Free Full Text]
Pei X, Rodriguez E. Provincial income inequality and self-reported health status in China during 1991-7. J Epidemiol Community Health 2006;60:1065-1069. [Free Full Text]
Hu TW, Mao Z, Liu Y, de Beyer J, Ong M. Smoking, standard of living, and poverty in China. Tob Control 2005;14:247-250. [Free Full Text]
Hesketh T, Lu L, Jun YX, Mei WH. Smoking, cessation and expenditure in low income Chinese: cross sectional survey. BMC Public Health 2007;7:29-29. [CrossRef][Medline]
The authors reply: In response to Chen: as we discussed in ourarticle, one limitation of our study is that we were unableto adjust for some important potential confounding factors.Data on family income were not collected in our study. However,we collected data on and adjusted for levels of education andwork-related physical activity, two important indexes of socioeconomicstatus that were highly related to family income.1 Furthermore,relative risk, population attributable risk, and mortality werecalculated separately for rural and urban residents; this shouldhave eliminated the confounding effects of rural–urbandifferences in family income. We adjusted for the body-massindex, a measure of malnutrition in our study. In addition,only deaths from cardiovascular disease, cancer, and chronicrespiratory disease were associated with cigarette smoking andincluded in the estimation of smoking-related deaths. Thereis no evidence that these diseases are caused by malnutrition.We agree that passive smoking has been associated with deathfrom coronary heart disease and lung cancer,2,3 and we notedthat our study might have underestimated the deaths from thesediseases that were attributable to smoking.
Dongfeng Gu, M.D., Ph.D. Chinese Academy of Medical Sciences Beijing 100037, China
Tanika Kelly, Ph.D., M.P.H. Jiang He, M.D., Ph.D. Tulane University School of Public Health and Tropical Medicine New Orleans, LA 70112 jhe{at}tulane.edu
References
Jing Y. Analysis of population structure in rural areas of China. Chin J Popul Sci 1998;10:17-30. [Medline]
He J, Vupputuri S, Allen K, Prerost MR, Hughes J, Whelton PK. Passive smoking and the risk of coronary heart disease -- a meta-analysis of epidemiologic studies. N Engl J Med 1999;340:920-926. [Free Full Text]
Department of Health and Human Services. The health consequences of involuntary exposure to tobacco smoke: a report of the Surgeon General — executive summary. Washington, DC: Government Printing Office, 2006.
Previous
