Absolute and Comparative Cancer Risk Perceptions Among Smokers in Two Cities in China

Abstract

Introduction:

Knowledge about health effects of smoking motivates quit attempts and sustained abstinence among smokers and also predicts greater acceptance of tobacco control efforts such as cigarette taxes and public smoking bans. We examined whether smokers in China, the world’s largest consumer of cigarettes, recognized their heightened personal risk of cancer relative to nonsmokers.

Methods:

A sample of Chinese people (N = 2,517; 555 current smokers) from 2 cities (Beijing and Hefei) estimated their personal risk of developing cancer, both in absolute terms (overall likelihood) and in comparative terms (relative to similarly aged people).

Results:

Controlling for demographics, smokers judged themselves to be at significantly lower risk of cancer than did nonsmokers on the comparative measure. No significant difference emerged between smokers and nonsmokers in absolute estimates.

Conclusions:

Smokers in China did not recognize their heightened personal risk of cancer, possibly reflecting ineffective warning labels on cigarette packs, a positive affective climate associated with smoking in China, and beliefs that downplay personal vulnerability among smokers (e.g., I don’t smoke enough to increase my cancer risk; I smoke high-quality cigarettes that won’t cause cancer).

INTRODUCTION

China is home to the world’s largest population of cigarette smokers (Li, Hsia, & Yang, 2011). It suffers more than 600,000 annual deaths from tobacco use (Gu et al., 2009; Peto, Chen, & Boreham, 2009), with as many as 400,000 annual smoking-attributable deaths due to cancer (Wang et al., 2010). Despite China’s growing tobacco epidemic and critical role in global tobacco control efforts (Lv et al., 2011; Peto et al., 2009), relatively little is known about Chinese people’s awareness of the health risks of smoking. A national survey conducted in 1996 found low levels of awareness, with only 40% of respondents aware of the link between smoking and lung cancer and 4% aware of the link between smoking and heart disease (Yang et al., 1999). However, a more recent study led to somewhat more optimistic conclusions. The International Tobacco Control (ITC) China Survey, conducted in six large Chinese cities, suggested that respondents largely acknowledged the association between smoking and lung cancer but were less aware of the other health effects of smoking (Yang, Hammond, Driezen, Fong, & Jiang, 2010). Specifically, more than 90% of nonsmokers and 68% of smokers agreed that smoking causes lung cancer, while only a third of nonsmokers and 16% of smokers agreed that smoking causes stroke.

These findings, although informative about Chinese people’s general awareness of the health risks of smoking, provide limited insight into whether Chinese smokers see themselves as personally vulnerable to health effects of smoking. Even if people recognize a risk in the abstract, they may tend to downplay their personal vulnerability (Weinstein, 1980). Also, prior studies are limited because they explicitly asked about links between smoking and health outcomes, which may overestimate knowledge by prompting recognition and agreement (Yang et al., 2010, p. i22). Asking people to estimate their personal risk of health outcomes without explicitly referring to smoking may thus provide new insights into risk beliefs.

Present Study

This study examined judgments of personal risk of cancer among Chinese smokers and nonsmokers. Because smoking is a risk factor for many cancers such as those of the lung, esophagus, larynx, mouth, throat, kidney, bladder, pancreas, stomach, and cervix (U.S. Department of Health and Human Services, 2004, 2010; Wang et al., 2010), perceived cancer risk should be elevated in smokers compared to nonsmokers if people are informed about the risks (Ayanian & Cleary, 1999; Hahn & Renner, 1998; Honda & Neugut, 2004; Strecher, Kreuter, & Kobrin, 1995; Weinstein, 1998; Weinstein, Marcus, & Moser, 2005). Indeed, smokers in countries with strong tobacco control policies tend to perceive a greater personal risk of cancer than do nonsmokers (Honda & Neugut, 2004; Lee, 1989; Segerstrom, McArthy, Caskey, Gross, & Jarvik, 1993; Strecher et al., 1995; Weinstein et al., 2005).

METHODS

Participants and Data Collection

Participants (N = 2,568) were recruited from four localities: urban and rural areas of Beijing (Xicheng and Huairou Districts, respectively), and urban and rural areas of Hefei, Anhui Province (Wuhu and Huaibei Cities, respectively). Within each locality, three subdistricts were selected to represent low, middle, and high economic development. In each subdistrict, two residential neighborhoods were selected, from which 110 households were sampled. One member from each household aged more than 18 was selected for interview based on proximity of birth date to interview date. Trained staff from the Chinese Center for Health Education visited sampled households in-person. Participants with sufficient literacy skills completed a printed questionnaire by themselves. For those with reading difficulty, the staff read the questionnaire aloud and filled in participants’ answers. The present analyses excluded 20 participants with a prior cancer diagnosis and 31 participants with missing data on one or more demographics.

The questionnaire was modeled after the Health Information National Trends Survey (HINTS) of the U.S. National Cancer Institute. The questionnaire primarily assessed trust in and use of cancer-related information and beliefs and attitudes about cancer prevention, detection, and treatment. Detailed information about the number of sampled respondents who refused to participate was not recorded. However, the refusal rate was estimated to be below 5%, with very few people declining to participate.

Measures

Smoking Status

Participants reported whether they had smoked at least 100 cigarettes in their lifetime and whether they currently smoked every day, some days, or not at all. Participants who reported that they had not smoked 100 cigarettes in their lifetime were coded as never-smokers, those who reported that they smoked 100 cigarettes but currently did not smoke were coded as former smokers, and those who reported that they smoked 100 cigarettes and currently smoked everyday or some days were coded as current smokers.

Perceived Cancer Risk

Perceptions of absolute cancer risk were elicited by asking participants to judge their likelihood of getting cancer in their lifetime on a scale from 1 (extremely unlikely) to 5 (extremely likely). Perceptions of comparative risk were elicited by asking participants to rate their likelihood of getting cancer compared to other similarly aged people, using a scale from 1 (extremely unlikely) to 5 (extremely likely).

Demographics

Demographic characteristics included gender, age, education level, annual household income, and residence.

RESULTS

Table 1 shows the demographic characteristics of participants. Consistent with the low rates of smoking cessation in China (Qian et al., 2010; Yang et al., 1999), only 45 participants (1.8%) were classified as former smokers. Thus, former and never-smokers were collapsed into a single category (nonsmokers) for comparison with current smokers. The rate of current smoking (22%) was comparable to that observed in other studies (Qian et al., 2010).

Table 1.

Perceived Absolute and Comparative Cancer Risk, by Smoking Status and Demographics

Characteristic	N	Comparative risk		Absolute risk
		B	p value	B	p value
Gender
Female	1,305	–		–
Male	1,212	−0.00	.963	0.00	.913
Age, years		0.01	.631	0.00	.906
18–24	170
25–34	579
35–44	654
45–54	584
55–76	530
Annual household income (RMB)		0.03	.025	0.03	.097
<20,000	427
20,000 to <40,000	719
40,000 to <60,000	508
60,000 to <80,000	330
80,000 or more	533
Education		0.00	.981	0.01	.455
Primary school or less	273
Middle school graduate	711
High school graduate	666
College: 2-year degree	472
College: 4-year degree	395
Residence
Urban Beijing	637	–		–
Rural Beijing	661	−0.25	<.001	−0.25	<.001
Urban Hefei	625	−0.19	<.001	−0.20	<.001
Rural Hefei	594	0.05	.378	0.02	.791
Smoking status
Nonsmoker	1,962	–		–
Current smoker	555	−0.09	.035	−0.08	.091

Note. Unstandardized B coefficients and p values are from linear regressions predicting comparative and absolute risk perceptions for developing cancer. R² was .03 for both models. RMB (renminbi) refers to the official currency of the People’s Republic of China.

Perceptions of comparative cancer risk were lower among smokers (M = 2.41, SD = 0.83) than nonsmokers (M = 2.48, SD = 0.79), a difference that did not reach significance by t test (t(855.31) = 1.87, p = .061, equal variances not assumed) but emerged as significant in a linear regression controlling for demographics (Table 1). Perceptions of absolute risk were in the same direction (M _SMOKERS = 2.45, SD = 0.86; M _NONSMOKERS = 2.51, SD = 0.80), but this difference did not achieve significance by t test (t(848.03) = 1.40, p = .163, equal variances not assumed) or regression (Table 1). Participants higher in income judged themselves to be at greater comparative risk. Participants from rural Beijing and urban Hefei judged themselves to be at lower comparative and absolute risk than did participants from urban Beijing.

DISCUSSION

Smokers in China tended to judge their comparative risk of cancer as significantly lower than did nonsmokers, while judgments of absolute risk were in the same direction but did not achieve significance. These findings contrast with studies in countries with developed tobacco control policies (e.g., Australia and the United States) in which smokers tend to rate their personal risk of cancer as higher than do nonsmokers (Honda & Neugut, 2004; Lee, 1989; Strecher et al., 1995; Weinstein et al., 2005). These analyses, although limited to residents of two cities, suggest that scientific knowledge about the links between smoking and cancer have yet to affect smokers’ personal assessments of risk in China.

Importantly, even smokers who recognize the link between smoking and cancer may find ways to downplay their personal risk (Ayanian & Cleary, 1999; Honda & Neugut, 2004; Oakes, Chapman, Borland, Balmford, & Trotter, 2004; Peretti-Watel et al., 2007; Segerstrom et al., 1993; Slovic, 2000; Strecher et al., 1995; Weinstein, 1998; Weinstein et al., 2005; Williams & Clarke, 1997). Indeed, a recent focus group study of Chinese smokers uncovered examples of self-exempting beliefs such as the notion that health depends primarily on genes rather than smoking, that only low-quality cigarettes are harmful, and that risks can be avoided through moderate use (Ma et al., 2008). In the absence of strong public health campaigns emphasizing the dangers of smoking, such beliefs may encourage smoking.

Positive associations with smoking in China (formed through tobacco marketing) may also contribute to dampened risk perceptions among smokers, as psychological research has found that inducing positive affect toward a product tends to reduce its perceived harmfulness (Slovic, Finucane, Peters, & MacGregor, 2004). Ultimately, communicating the harmfulness of cigarette smoking may require eliminating associations between smoking and happiness, cherished holidays, gift-giving, respect, and virility in China (Chu, Jiang, & Glantz, 2011; Li & Yong, 2009; Zhang, Chan, Fong, Malone, & Lam, 2012).

Implications

Risk perceptions are crucial in motivating quit attempts and sustained abstinence, and they may also predict greater acceptance of tobacco control efforts such as cigarette taxes and public smoking bans (Borrelli, Hayes, Dunsiger, & Fava, 2010; Dillard, McCaul, & Klein, 2006; Gibbons, McGovern, & Lando, 1991; McCoy et al., 1992; Weinstein et al., 2005; Yang et al., 2010). This study points to an urgent need for educational efforts in China that connect smoking to cancer risk. Although coverage of health risks of smoking in Chinese media such as newspapers has increased in recent years (Gao, Chapman, Sun, Fu, & Zheng, 2012), it remains less frequent than coverage of other health risks (Peng & Tang, 2010), is less frequent than in other countries such as the United States and Australia (Gao et al., 2012), and is particularly lacking in the extent to which it tends to elicit feelings of personal vulnerability to the health effects of smoking (He, Shen, Yin, Xu, & Lan, 2013). Also, while China officially entered into the WHO Framework Convention on Tobacco Control (FCTC) in 2006, recent studies have found large gaps in China’s implementation of several of the FCTC’s tobacco control measures (Lv et al., 2011), with a recent assessment awarding China only 2 out of 16 points for its implementation of public smoking bans and 0 out of 2 points for elimination of tobacco advertising and promotion (World Health Organization, 2012). These findings, combined with the present results showing that smokers did not perceive themselves to be at increased risk of cancer, emphasize the need for more consistent and widespread warnings about the dangers of tobacco in China.

The present results also highlight the tendency of individual smokers to exempt themselves from known risks of smoking. Thus, communication efforts must emphasize the risks of smoking any amount and for any duration (Schane, Ling, & Glantz, 2010) and highlight the benefits of smoking cessation in all smokers (Kenfield, Stampfer, Rosner, & Colditz, 2008). Warning labels on cigarette packages are a potential starting place for implementing these strategies, as current warning labels in China consist of general statements such as “Smoking is harmful to your health” (Qin et al., 2011; Wan et al., 2012). Continued research is needed to uncover factors that may interfere with Chinese smokers’ recognition of their heightened cancer risk. For example, they may have beliefs that normalize the dangers of smoking because of the pervasiveness of risk (Oakes et al., 2004)—for instance, given the prevalence of air pollution in a particular region (Chen et al., 2013).

Limitations

This study elicited perceptions of general cancer risk rather than focusing on individual cancer types, such as lung cancer. While there are benefits to this survey strategy (e.g., smoking increases the risk of many types of cancer, so it is important to know whether people understand the risks), it raises the question of how participants would have rated their personal risk of lung cancer specifically. Also, this study was not nationally representative, which is crucial because China is a diverse country with many regions differing in economic development and administrative policies (e.g., autonomous regions and prefectures, mountain villages). Research is needed to examine smoking knowledge and cancer risk perceptions across subpopulations, particularly when local characteristics may impede health knowledge or necessitate unique communication strategies.

CONCLUSIONS

Smokers in two Chinese cities did not recognize their heightened risk of cancer, suggesting that public health knowledge and self-exempting beliefs concerning cancer may be ripe targets for interventions to reduce smoking-related morbidity and mortality. Communication efforts should associate smoking with cancer, cast smoking cessation as a primary method of preventing cancer, and anticipate the tendency for smokers to downplay their personal risk.

FUNDING

This work was funded by the Chinese Center for Health Education, Ministry of Health, China.

DECLARATION OF INTERESTS

None declared.