Decade-long research based on data in CKB showed that smoking is set to kill one in three of all the young men in China. In contrast, the overall number of female deaths caused by tobacco is decreasing, with less than 1% of deaths in women born since 1960 now attributable to tobacco. Through extensive analysis of a wide spectrum of diseases captured in CKB, smoking was found to be associated with increased risks of 56 diseases and 22 causes of death (Lancet Public Health 2022), including cancer (Cancer 2015), chronic respiratory and cardiovascular diseases (Lancet 2015), diabetes (Lancet Public Health 2018), and many other under-studied conditions. Unless there is widespread cessation, the annual number of tobacco deaths in China will rise from about 1 million in the 2010s to 2 million in the 2030s and 3 million in the 2050s.
China now consumes about 40% of the world’s cigarettes, almost exclusively by men.
The main increase in cigarette consumption took place after 1980 and appeared to have plateaued in recent decades, with a time-lag between rural and urban areas. Hence, as a delayed effect, the full health consequence of this large recent increase in cigarette consumption has yet to be seen. With detailed smoking information that was objectively validated by exhaled carbon monoxide (Int J Epidemiol 2013), CKB is ideally placed to monitor the long-term evolution of the tobacco epidemic over time, overall and separately in men versus women and in urban versus rural area. In CKB, two-thirds of men smoked at baseline, with little dependence of smoking prevalence on age. However, many had not smoked cigarettes throughout their adult life. Comparing men born before and since 1950, in the older generation, the age at which smoking started was later and, particularly in rural areas, lifelong exclusive cigarette use was less common than in the younger generation. Among women, overall only 3% smoked, with the prevalence strongly dependent on year of birth: among women born in the 1930s, 1940s, 1950s, and since the 1960s, the smoking prevalence was 10%, 5%, 2% and 1%, respectively (IJE 2011, Lancet 2015). This reflected a surprising tenfold intergenerational reduction in smoking uptake rates for reasons that are still not fully understood.
The research in CKB showed that the proportion of all male deaths at ages 40-79 that are attributed to smoking has doubled from about 10% in the early 1990s, to more than 20% in the 2010s.
The research in CKB showed that the proportion of all male deaths at ages 40-79 that are attributed to smoking has doubled from about 10% in the early 1990s, to more than 20% in the 2010s. (Lancet 2015) In urban areas this proportion is higher at more than 25%, and still rising. In rural areas it is currently lower (15%), but is set to rise even more steeply than in cities, due to the high prevalence and intensity of smoking and low rate of cessation in rural men. Conversely, among Chinese women the overall deaths caused by tobacco are decreasing, with less than 1% of deaths in women born since 1960 due to tobacco. Other studies, however, have shown a recent increase in smoking uptake by young women that could potentially reverse this downward trend.
The CKB also showed that despite the rising risk per smoker over time, the smoker versus non-smoker relative risks for most diseases e.g. lung cancer, ischaemic heart disease (IHD), stroke, and Chronic Obstructive Pulmonary Disease (COPD) in China are still much less extreme than in recent studies in US or UK. However, for many diseases including lung cancer, COPD and stroke, the absolute excess mortality in Chinese smokers is still substantial, as Chinese people have much higher non-smoker rates than Americans, especially in later middle age. In China, lung cancer rates in those who had never smoked fell steadily over time, probably due to a large reduction in household air-pollution, whereas the smokers’ lung cancer rates rose and will continue to do so, resulting in greater relative risks now than in the past (Lancet 2015; Lancet Public Health 2022). With a few exceptions, for most diseases female smokers have more modest relative risks than male smokers, in line with differences in exposure patterns and smoking intensity. For lung cancer and COPD, however, there was little difference in risk estimates between male and female smokers, providing support for the notion that women appeared to be more susceptible to lung diseases associated with smoking, due probably to potential protective role of oestrogen in these diseases and the anti-oestrogenic effects of tobacco smoke. Likewise, the smoking-associated risks of diabetes appeared greater among women than men for a given amount smoked. This may be explained in part by their greater proportion of body fat compared with men. Consistent with this finding, among men, the smoking-associated risk of diabetes appeared to be greater among individuals with higher adiposity (e.g. 30% versus 60% excess risk at BMI<25 and ≥30 kg/m2, respectively). Although on average, smokers tended to be leaner than non-smokers, heavy smokers were more likely to have higher abdominal adiposity than light or non-smokers, greatly increasing the risk of diabetes.
Encouragingly, an increasing proportion of smokers in China are choosing to stop smoking. Compared with an earlier cohort of 220,000 men in China, the proportion of smokers who had quit rose from 3% in 1991 to 9% in 2006 in CKB.
Encouragingly, an increasing proportion of smokers in China are choosing to stop smoking. Compared with an earlier cohort of 220,000 men in China, the proportion of smokers who had quit rose from 3% in 1991 to 9% in 2006 in CKB (IJE 2011; Lancet 2015). For smokers who stopped before developing any serious disease, after ten years of not smoking their risk of death was similar to that of people who had never smoked. Previous studies have suggested that giving up smoking may increase the risk of developing type 2 diabetes, at least in the short term. By carefully analysing information on the reasons why people stopped smoking, CKB demonstrated that this excess risk of diabetes among recent quitters was only seen in those who had stopped smoking because of illness, and not those who chose to stop smoking for other reasons. Among individuals who had chosen to give up smoking and had stopped for more than five years, there was a small excess risk of diabetes, which could be explained by moderate weight gain following smoking cessation (Lancet Public Health 2018). These findings add to existing evidence of the health benefits of giving up smoking, not only for prevention of cancer, respiratory and cardiovascular diseases, but also for prevention of diabetes. They should also help dispel common myths about tobacco and its use in China, including the widely held view that stopping smoking could inadvertently damage health by disrupting the established biological balance.
More recently, CKB has begun exploring molecular mechanisms linking smoking with risks of multiple diseases based on the new proteomics data generated using Olink and SomaScan assay platforms that together quantify plasma levels of about 10,000 proteins. The analyses found that smoking affects plasma levels of about 100 proteins that are involved in multiple pathways and functions. The study showed that weight changes associated with smoking or smoking cessation are partly mediated through a protein called growth differentiation factor 15 (GDF15). Smoking can increase GDF15 levels and elevations in GDF15 reduce appetite and food intake, leading to reduced weight among smokers and weight gain following smoking cessation.
Impact of research
Tobacco use is the largest preventable cause of premature death worldwide and is projected to claim one billion lives in the 21st century, with up to 80% of these deaths occurring in low- and middle-income countries (LMICs), particularly in China. The decade-long research in CKB has, and will continue to provide reliable nationwide evidence in China of the hazards of smoking and how this epidemic is likely evolve in the future. Such evidence has informed China’s public health policies, including the drafting of two government white papers on tobacco control. Drawing on decades of research in tobacco epidemiology, CKB researchers have written a commissioned article critically reviewing the evolution of the smoking epidemic, the existing epidemiological evidence on its health impact, and the major issues related to tobacco control in China (Lancet Public Health 2023).
The self-regulating national tobacco monopoly and its influence on policy, the country’s relatively low tobacco tax, weak package warnings and enforcement of other tobacco control strategies are all challenges to tobacco control in China. However, these challenges can also provide opportunities to discourage young women from starting to smoke and encourage cessation in men, to assist China’s long march towards better health.