Associations of socioeconomic and religious factors with health: a population-based, comparison study between China and Korea using the 2010 East Asian social survey

A large body of research has shown that socioeconomic factors have a significant impact on health within countries [1–3]. The pattern of health differentials within countries can be shaped by national health promotion policies, social security systems, delivery of healthcare services, and economic and political environments [4, 5]. Therefore, a comparative analysis of the association between socioeconomic factors and health in countries with different historical and economic backgrounds may provide useful evidence for addressing health differentials [5, 6].

Many previous studies have compared the association between socioeconomic factors and health among Western countries [5–10]. These studies have shown that such associations vary in strength across countries. Using data from the Collaborative Research on Ageing in Europe (COURAGE) survey conducted in Finland, Poland, and Spain, Freeman et al. [10] found a significant association between depression and socioeconomic status in all countries. Socioeconomic status was computed using the combined scores of the total number of years of education and the quintiles of the country-specific income level of the household. After adjusting for confounders, the odds of depression decreased significantly for every unit increase in the SES index for Finland, Poland, and Spain. Additionally, higher education significantly decreased the odds of depression in each country, but income did not [10].

However, very few studies have compared the association between socioeconomic factors and health in East Asian countries [11–14]. Among these, Hanibuchi et al. [11] explored the relationship between socioeconomic status and health in China, Japan, Korea, and Taiwan. However, their study had the notable limitation of relying on the 2006 East Asian Social Survey, which lacked information on disease status and health-related behaviors and therefore allowed only one control variable (age) in the analysis. Moreover, using cross-sectional surveys conducted in 2003, Nomura et al. [13] investigated the effects of socio-demographic factors on health in Korea and Taiwan. However, the sampling methods used in the two countries differed, thus preventing cross-country comparisons of the results. In addition, their research focused on the effects of social capital and demographic factors on health, rather than the effects of a broad range of socioeconomic factors.

Meanwhile, previous studies have demonstrated the protective effect of religion on the health status of individuals in Western countries [15–17]. Nicholson et al. [15] considered whether attendance at religious services was associated with better self-rated health in a range of European countries and reported an association between less frequent attendance of religious services and poor health in Europe, although they emphasized the importance of taking individual and contextual factors into account. However, the association between religion and health in Asian countries may be different from that in Western countries. For example, in China, religion is unpopular and church attendance is not common [18]. China has a population of about 1.3 billion, but only roughly 100 million of them (approximately 7.7%) are claimed to adhere to Buddhism, Taoism, Islam, Catholicism, or Protestant Christianity to at least some extent [19]. When the proportion of religious people is as low as in China, there are not enough religious people proportionately in the population to reflect the impact of religion on unhealthy behaviors [20]. Nevertheless, very few cross-national studies have examined the effect of religion on health in East Asian countries.

The aim of this study is to compare the associations of socioeconomic and religious factors with self-rated health in China and the Republic of Korea (hereafter, Korea). In our study, socioeconomic determinants were assessed based on education levels, employment, household income, and social class. Cross-national comparative research on health outcomes between China and Korea is of particular interest. China and Korea share many similarities in terms of Confucian culture, family structure, lifestyle, and geographical location [21], but the two countries also differ in important ways, namely, government structures, income distribution, social security systems, the delivery of healthcare services, social inequality, and religion [14]. Meanwhile, cross-national comparative research on health has not always been possible because of a lack of comparable data. To overcome these problems, we analyzed a dataset from the 2010 East Asian Social Survey (EASS), a cross-national survey consisting of nationally representative samples from China and Korea. To the best of our knowledge, no research to date has explicitly compared the associations between socioeconomic and religious factors and health in China and Korea.

Additional file 1: Table S1 shows the odds ratios of good self-rated health by socioeconomic, demographic, risk/health care, and social capital factors within China and Korea (base outcome is excellent self-rated health). Additional file 1: Table S2 shows the odds ratio of poor self-rated health by socioeconomic, demographic, risk/health care, social capital factors within China and Korea (base outcome is excellent self-rated health).

Even after adjustment, a higher level of education significantly decreased the likelihood of reporting poor health versus excellent health in both countries. In particular, there was a steeper education gradient in health in Korea. The odds ratio for poor health was 0.28 (95% CI, 0.14–0.59) for Koreans with a college education or higher and 0.49 (95% CI, 0.25–0.96) for those with senior high school education, in comparison to their counterparts with elementary school education or lower (Model 2, Table 4). However, education level showed no significant association with good health versus excellent health in either China or Korea (Model 2, Table 3).

Only in China were employed persons much less likely than their unemployed counterparts to report either poor health or good health versus excellent health. For employed persons, the odds ratios for good and for poor health were 0.67 (95% CI, 0.55–0.82) and 0.58 (95% CI, 0.45–0.75), respectively (Model 2, Tables 3 and 4).

There was also a steep income gradient in health with respect to both 1) good health versus excellent health and 2) poor health versus excellent health in China, of which the latter was much clearer than the former. For Chinese subjects in the highest income quartile, the odds ratio for poor health was 0.49 (95% CI: 0.33–0.73) relative to their counterparts in the lowest income quartile. Meanwhile, in Korea this association between income and poor health was not significant after adjustments (Model 2, Table 4).

In the fully adjusted models, the association of social class with either 1) good health versus excellent health or 2) poor health versus excellent health was significant only in China, where a positive relationship was shown between social class and health. Compared to people in the lowest social class, the odds ratio for good health was 0.42 (95% CI: 0.27–0.67) and that for poor health was 0.47 (95% CI: 0.27–0.84) for people in the highest social class (Model 2, Tables 3 and 4).

The association between religion and health was significant only in Korea. For Buddhists, the odds ratio for good health was 1.72 (95% CI: 1.16–2.55), and 1.96 (95% CI: 1.23–3.13) for poor health (Model 2, Tables 3 and 4), compared to Christians. In addition, for Koreans in the “other” religion category, the odds ratio for good health was 1.66 (95% CI: 1.19–2.32) compared to Christians (Model 2, Table 3).

This study examined the association between socioeconomic and religious factors and self-rated health in China and Korea. We found that some socioeconomic and religious factors matter more in one country than the other. Employment, income, and social class seem to have generally stronger effects on health in China, while education and religion tend to have larger effects on health in Korea. Using multinomial logistic regression, we found that the findings are more pronounced with respect to poor health versus excellent health than for good health versus excellent health, as shown by the fact that the odds ratios reported in Table 4 tend to have greater statistical significance than do those in Table 3.

Our results reveal that in both China and Korea, education has a significant association with the odds of poor health versus excellent health, rather than with the odds of good health versus excellent health; these are consistent with the results of prior research indicating that education is strongly associated with health [23–25]. However, this education gradient in health was steeper in Korea. Considering that higher levels of education may confer knowledge and cognitive assets that are health protective [1], public policies to enhance education could provide the Korean people with greater health benefits.

In China, unlike Korea, employed persons were much less likely to report poor health than their unemployed counterparts, which is in line with the findings of earlier research [26–29]. The reason for this might be that employed persons were in the labor market because they were healthy [30]. However, this cannot explain the lack of evidence of an employment gradient in health in Korea. Another reason might pertain to differences in the accessibility of health determinants according to employment status in China. While Korea has implemented a universal public health insurance system providing all basic benefits that covers the whole population, China does not have a universal health insurance system, so that even insured people have different degrees of coverage depending on employment status and job characteristics [31, 32].

Meanwhile, we observed a steeper income gradient in health in China, suggesting the health impact of serious income inequalities in that country, which are currently viewed as a major problem in Chinese society. According to 2010 data, China’s Gini coefficient (0.481) places it among the countries with the highest levels of inequality in the world, not only higher than those of developed western countries but also higher than the Asian average (0.351) [33]. We may thus assume that although the healthcare system has been reformed, it is still difficult for people in low-income groups to access much-needed healthcare services [34]. People in the high-income bracket not only have a greater chance of accessing health care [35], but can also afford costly commercial insurance. In addition, income may also represent purchasing power or access to tangible resources (e.g., better housing, working conditions, food and health care, and increased social support and community cohesion) at both the individual and community level, with concomitant implications for health status [1]. Therefore, it seems that healthcare policy reform in China is necessary so as to reduce health inequalities resulting from income inequalities.

In the adjusted models, we found significant associations between social class and health in China but not in Korea. This finding is consistent with previous studies that have shown that self-rated health may be sensitive to social class differentials in some countries [11, 29, 36–38]. One possible explanation for this is a less equitable distribution of health-improving resources across social class in China than in Korea.

The association between religion and health was significant only in Korea, where Buddhists and people of other religions were more likely than Christians to report poor health. Many studies show that religion and spirituality are linked to health outcomes [15–17]. However, within this body of literature, it is unclear why people experience specific health benefits from religion [15, 39]. One possible explanation is that Christians disapprove more strongly of behavior that is potentially damaging to health [16]. However, we found that the association between religion and health was not significant in China. It is plausible that Chinese religions such as Buddhism and Taoism, as foundations of traditional Chinese culture, differ from Western religions in their conceptualizations of a supernatural being and afterlife, rituals, and organization, as well as health-influencing lifestyles [18]. In a study on religion, superstition, and Chinese suicide attempters, Zhang and Xu found that suicide attempters with high religiosity had a higher degree of suicide intent than did those with low religiosity or no religion, unlike the findings of Western studies [40]. They assumed that these findings might be due to Chinese religions and traditional cultural values. In Chinese religions there is no single god for people to worship, and these religions offer no social support system or coping mechanism as the majority of adherents do not meet regularly. Unlike the mainstream religions in the West, Chinese religions often seem to be associated with superstition, referred to as religious superstition (zongjiao mixin). To some Chinese people, being religious is equivalent to being superstitious, and death is seen as a solution to all problems and the beginning of a new life. Therefore, it is possible that those prone to extreme actions or in difficult circumstances are likely to think about starting a new life by ending their present miserable one quickly [40]. Meanwhile, our finding that religion is related to health in Korea and not in China may be due to a lack of statistical power. However, even in the additional analysis where we categorized religion into four groups (Christian, Buddhist, no religion, and others) and conducted multinomial logistic regression analyses, we found that the association between religion and health remained significant only in Korea (results can be provided on request).

To the best of our knowledge, the current study is the first to examine the associations of socioeconomic and religious factors with self-rated health in China and Korea using national data. The dataset permits direct cross-national comparisons through a coordinated research setting.

Our study had several clear limitations. First, objective health may be a better measure than self-rated health. Subjective reports of health status may be confounded by other variables, such as neuroticism or psychological distress, and may not correlate with the underlying pathology [41]. In addition, there is some uncertainty about what is being measured when using self-rated health as a health outcome [42]. However, measures of self-rated health have proven to be both reliable and valid health indicators across a wide range of age groups [43, 44] and predictors of objective health statuses [42–46], showing them to be closely associated with physicians’ diagnoses [47] and people’s future healthcare use use [48]. Second, it is important to bear in mind that the use of cross-sectional data precludes any definitive causal conclusions about the relationship between socioeconomic and religious factors with self-rated health. Third, the 2010 EASS had relatively low response rates. The response rates of 73.0% in China and 63.0% in Korea might have entailed non-response bias. We conducted an additional analysis comparing the distribution of individuals for each age category between the sample used in the present study and the sample of the national census for each country. According to these results, the difference in the proportion of individuals for each age category was less than 3 percentage points for each country, except for differences of 5.5 percentage points for individuals aged ≤29 in China and of 3.7 percentage points for individuals aged 30–39 in Korea (Additional file 1: Table S3). Fourth, there is a possibility that we could not completely control for health care system factors using only the two variables of health insurance and unmet medical need. With respect to health insurance, we only examined whether the participant had health insurance. Korea has a universal system of health care, a single insurer (the National Health Insurance Service), while China does not. Therefore, we could not control for the type of insurance or coverage depth (the scope and percentage of expenses reimbursed) in China. Fifth, health ratings may be influenced by cultural factors. For example, as stated in the 2016 OECD Health Data, Korea scored among the lowest of the member countries, as only 32.5% of the Korean population gave a good evaluation of their health in the self-rated health survey. In contrast, 88.1% of Americans rated their health as good, the highest rate among OECD members [49]. As has been noted elsewhere, when evaluating their health, respondents tend to draw upon different health aspects, including both physical and psychological well-being, and health behaviors [50]. However, previous research has shown that East Asians share similar patterns of perception and expression of, and response to, health [51]. Finally, we used only one variable pertaining to religion, religious affiliation. Because the EASS 2010 dataset does not provide additional information, we could not include other measures of religiosity such as the frequency of attendance at religious services, which has been used in previous studies [15, 16]. Further studies need to include such religious factors [15, 16].

The present study found that socioeconomic and religious factors were associated with self-rated health in China and Korea, but that the patterns of the association differed between the two countries. The differences in health seem to be associated with employment, household income, and social class in China, whereas education and religion seem to be good predictors of health differentials in Korea. These inter-country differences may reflect systemic and cultural differences between China and Korea that impact social and economic inequalities, ultimately leading to differences in health. This study illuminates differences in health determinants between the countries of interest and may inform future health improvement in either country. Ultimately, there is a need for more sophisticated, cross-national studies to be conducted in more countries and to better facilitate those in China and Korea.