Elsevier

Social Science & Medicine

Volume 67, Issue 6, September 2008, Pages 988-1001
Social Science & Medicine

Explaining gender differences in ill-health in South Korea: The roles of socio-structural, psychosocial, and behavioral factors

https://doi.org/10.1016/j.socscimed.2008.05.034Get rights and content

Abstract

This study examines and explains the gender disparity in health despite rapid modernization in South Korea where the social structure is still based on traditional gender relations. A nationally representative sample of 2897 men and 3286 women aged 25–64 from the 2001 Korean National Health and Nutrition Examination Survey was analyzed. Health indicators included self rated health and chronic disease. Age-adjusted prevalence was computed according to a gender and odds ratios (OR) derived from logistic regression. Percentage changes in OR by inclusion of determinant variables (socio-structural, psychosocial, and behavioral) into the base logistic regression model were used to estimate the contributions to the gender gap in two morbidity measures. Results showed a substantial female excess in ill-health in both measures, revealing an increasing disparity in the older age group. Group-specific age-adjusted prevalence of ill-health showed an inverse relationship to socioeconomic position. When adjusting for each determinant, employment status, education, and depression contributed the greatest to the gender gap. After adjusting for all suggested determinants, 78% for self rated health and 86% for chronic disease in excess OR could be explained. After stratifying for age, the full model provided a complete explanation for the female excess in chronic illness, but for self rated health a female excess was still evident for the younger age group. Socio-structural factors played a crucial role in accounting for female excess in ill-health. This result calls for greater attention to gender-based health inequality stemming from socio-structural determinants in South Korea. Cross-cultural validation studies are suggested for further discussion of the link between changing gender relations and the gender health gap in morbidity in diverse settings.

Introduction

Most studies on gender and health have been carried out in Western countries. However, this paper focuses on examining and explaining the gender disparity in ill-health in the Korean context. Recent studies have revealed more complex patterns of gender differences in morbidity (Arber and Cooper, 1999, Lahelma et al., 1999, Macintyre et al., 1996), suggesting growing interest in an “age-specific and condition-specific view” (Macintyre et al., 1996: 624) of gender differences in health. Despite this development, few studies have examined the association between gender and health in an East Asian Confucian context.

While gender relations are changing, women as a group still face structural inequality in most societies (Doyal, 1995, McDonough and Walters, 2001). The social effect of this inequality on women's physical and mental health has been widely studied. However, very few have assessed the factors contributing to gender inequality in the morbidity gap. Using the theoretical background of different exposure and different vulnerability, some Canadian studies (Denton et al., 2004, Denton and Walters, 1999) predicted the gender differential in social determinants, but analyzed women's and men's health separately. The lack of evidence may mainly be due to the complex interplay of biological, psychosocial, behavioral, and social determinants (Lahelma et al., 1999) or perhaps the inconsistency in the pattern of gender gap in morbidity (Macintyre et al., 1996). Some European studies (Arber and Cooper, 1999, Lahelma et al., 1999) have argued that the “convergent trends” in the (reducing) gender gap in morbidity have appeared due to the improved status of women in recent decades. However, gender differentials in morbidity may be differently displayed in other countries

South Korea, like other East Asian Confucian countries, is highly gender differentiated, leaving women marginalized in society. Using data from various sources, a recent study (Chun, Doyal, Payne, Cho, & Kim, 2006) suggests that higher levels of gender inequality in Korean society might be the major cause of female excess morbidity, despite growing wealth in the country. We hereby aim to build on the evidence that social factors can be strong mediators that produce gender inequality in health. Our research attempts to unravel the observed gender gap in ill-health, attributable to socio-structural, psychosocial, and behavioral factors, using a representative Korean data set. As noted, recent literature has highlighted a complex array of gender health differences. However, none of these studies have proposed an explanation of consistent gender gap in morbidity. This knowledge gap prompted our research questions.

As a country well known for its rapidly growing economy, the unique characteristics of South Korea are also reflected in its gender inequality. The outer façade exhibits modernized characteristics and social change, but both family and social life are still based on traditional gender relations.

In terms of gender equitable social development, Korea has made considerable progress in opening educational opportunities to women whose educational attainment has expanded remarkably since the 1960s. In fact, since the 1990s, there has been no gender difference in the proportion of secondary school graduates (KWDI, 2004). The percentage of women attaining tertiary education rose from 2.4% in 1975 to 18% in 2000, whereas men experienced a rise from 9.5% to 31%. The education level of Korean women is presently very high in international comparison (UNDP, 2006). However, the dramatic increase in women's education is not positively associated with economic participation (Hwang, 2003). Indeed, until the 1980s an educated female was inhibited from participating in the labor market, while less educated women have been working continuously (Kim, 1995, Lee and Jung, 1999). This is because college education was valued merely as a preliminary step to marriage for the middle class. Even until very recently, the largest and most prestigious female university in Korea had a prohibition against married students in order to help ensure women would complete their degree before marriage (The Chronicle of Higher Education, 2003). Women were expected to leave work to concentrate on family care after marriage; while “domestic ideology” dictated that men were the main “bread-winners.” Although the female economic participation rate has increased noticeably among college graduates from 36% in 1980 to 55% in 2000, this figure has remained below the male participation rate of 83% (KNSO, 2000).

Although there is now greater participation in the workforce by women, occupational segregation persists. While most managerial/professional jobs are dominated by men, female-dominated jobs are mostly in limited areas such as service/sales work (74% in 2003; KNSO, 2003). Correspondingly, the male to female wage gap still remains great (ratio of estimated F/M income is 0.46; UNDP, 2006). According to one study, the average wage of male workers increases as they get older until the ages of 45–49, while female workers receive their highest wages at the ages of 30–34, with their wages gradually decreasing thereafter (Yoo, 2003). However, recent Korean data show that both of these trends are changing (KWDI, 2004).

Although Confucianism lost its position as state ideology long ago, contemporary Confucianism persists as a cultural force. It considers the male as a positive being (yang) and the female as a negative counterpart (ying). The Confucian influence helps to keep Korean women subordinate socially and economically. Sung (2003) noted that Korean traditions, especially in families, remain strong despite many social and economic changes. Patriarchal gender relations have been used to secure the wife's subordinate position. Women's domestic work was treated as inferior; for men to do it has been regarded as shameful, defaming masculinity. Even in dual income families only wives are expected to do housework. This attitude started to slowly change (only in women) around the 1990s. According to a recent survey (KNSO, 2005), 88% of working women still reported doing all, or almost all, household chores.

The extent of the gender disparity, despite the high level of economic development in Korean society, may be of great interest. According to an international comparison (UNDP, 2006), the Gender Empowerment Measure (GEM; indicator representing the level of economic autonomy and political participation in society) shows a low status for Korean women, 53rd out of 80 countries, as opposed to a high 27th rank out of 140 countries according to the Gender Development Index (GDI; a composite index of life expectancy, knowledge, and a decent standard of living). A further indicator from international gender gap monitoring (Zahidi, 2006), the Gender Gap Index (composed of economic status, political empowerment, educational attainment, and health and well-being), ranked South Korea 92nd out of 115 countries.

So, while the social status of men has been consistently high, the socioeconomic status of women is “lagging behind” economic growth in Korea. For instance, some have asserted that the gender wage gap in Western countries has narrowed, not by increases for women but by decreases for men (O'Neill & Polacheck, 1993). However, this has not happened in Korea. While economic development has moved Korea forward socially and politically in terms of gender equitable opportunities, this has not transferred to equality in employment and family structure. While few studies have empirically tested the relationship, this is very likely to have an impact on women's health in South Korea.

Gender differences usually mean inequality and discrimination, with a deleterious effect on the lives of women as a subordinate group (Doyal, 1995, Krieger, 2000). Generally speaking, women as a group have a lower social status, fewer opportunities for higher education, work, and poorer financial resources than their male counterparts. Gendered structural inequalities are also usually marked by unequal access for men and women to material and other resources (Sen, George, & Östlin, 2002). In a society such as South Korea with a high level of gender discrimination, cultural norms and values exacerbate the phenomenon. This can be linked to access to health related information and the allocation of resources (i.e., food, housing, rest, health care). The power imbalance also translates into differential levels of autonomy and control in employment. Ross and Bird (1994) demonstrated that gender stratification in work and the subjective experience of this inequality disadvantage women. Additionally, women are more likely to encounter undesirable life events and probably concurrent stress and strain (Mcleod & Kessler, 1990), which could account for higher female morbidity. While issues of higher female susceptibility to symptoms are not disregarded in the gender and health literature, there is substantial evidence to refute the “reporting artifact” explanation (Gijsbers van Wijk et al., 1995, Macintyre et al., 1999).

We have argued that socially produced gender inequality could be a strong casual factor in women's health in South Korea with its distinctive patriarchal system, namely the gender imbalance in economic power and political authority. From a Korean context, the purpose of the study is (1) to assess the patterns of gender difference in ill-health according to social factors; and (2) to examine the extent to which social determinants can explain away the differences with specific attention to socio-structural factors. These objectives are explained in the age groups 25–44 and 45–64, with the assumption that women in different cohorts would have been exposed to a different level of socio-structural gender inequality in a rapidly changing Korean society. Specifically, this study sets out to address the following issues: how are social determinants related with the health of Korean men and women? Is the gender gap different by generation? How far can social structural factors explain the gender gap in ill-health?

Section snippets

Design and study population

The data were derived from the 2001 Korean National Health and Nutrition Examination Survey (KNHANES), a cross-sectional survey. A stratified multistage sampling design was applied to the South Korean population, according to geographic areas, age and gender groups. Trained interviewers conducted surveys in households and administered a structured questionnaire to obtain the demographic, socioeconomic, behavioral and health characteristics of each respondent. The overall response rate was 88.5%

Results

Table 1 shows age and ailment-specific gender differences in ill-health among Korean adults. Except for men aged 35–39, female higher morbidity was observed in both self rated health and chronic diseases measures, with the gender gap being much wider in older adults (see Fig. 1 and Table 1). Stratified by disease categories, musculoskeletal and mental/behavioral diseases showed significant gender differences. However, without musculoskeletal and mental/behavioral diseases, the overall gender

Discussion

This study assessed the patterns and determinants of gender-based health inequality, based on a large representative sample of South Koreans. Contrary to several recent studies of Western countries (Lahelma et al., 1999, Macintyre et al., 1996), our results showed significant female excess morbidity in self rated poor health and chronic diseases among Korean adults aged 25–64, according to both the relative measure (OR) and the absolute measure (age-adjusted prevalence). We also acknowledge the

Conclusion

This paper adds to the gender and health literature by examining and unraveling the contributing factors of female excess morbidity, which is assumed to be based on gender-based structural inequality against an otherwise dynamic Korean society. The results suggest that policy priorities should be given to gender equitable opportunities in social production in order to reduce avoidable risks in the morbidity of women. Future longitudinal studies need to confirm the assumption we tested that

Acknowledgements

This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MOEHRD) (KRF-2005-078-BS0004). The authors thank Adam Turner and the anonymous reviewers for their fruitful comments.

References (55)

  • P. McDonough et al.

    Gender and health: reassessing patterns and explanations

    Social Science & Medicine

    (2001)
  • S. Macintyre et al.

    Do women ‘over-report’ morbidity? Men's and women's responses to structured prompting on a standard question on long standing illness

    Social Science & Medicine

    (1999)
  • S. Macintyre et al.

    Gender differences in health: are things really as simple as they seem?

    Social Science & Medicine

    (1996)
  • L. Strazdins et al.

    Women, work, and musculoskeletal health

    Social Science & Medicine

    (2004)
  • D. Umberson

    Gender, marital status and the social control of health behavior

    Social Science & Medicine

    (1992)
  • S. Arber et al.

    Gender and inequalities in health across the life course

  • J. Chang

    Economic crisis and women's work

    (2001)
  • H. Chun et al.

    Understanding women health and social change: the case of South Korea

    International Journal of Health Services

    (2006)
  • L. Doyal

    What makes women sick: Gender and the political economy of health

    (1995)
  • S.K. Hwang

    Job characteristics and occupational sex segregation: View from women's work in South Korea

    (2003)
  • H.S. Kang et al.

    The lives of daughters-in-law who care for parents with dementia

    Taehan Kanho Hakhoe Chi

    (1999)
  • Y.H. Khang et al.

    Health inequalities in Korea: age- and sex-specific educational differences in the 10 leading causes of death

    International Journal of Epidemiology

    (2004)
  • KIHASA/MOHW

    The 2001 Korean National Health and Nutrition Survey

    (2002)
  • Y.O. Kim

    The unstable transition of female employment and related policy tasks

    (1995)
  • KNSO

    Annual survey on economically active population

    (2000)
  • KNSO

    Annual report on vital registration: Birth and death

    (2003)
  • KNSO

    2004 Time Use Survey

    (2005)
  • Cited by (52)

    • Risk factors associated with initiation of a biologic disease modifying anti-rheumatic drug in patients with rheumatoid arthritis: A nested case-control study on 34,925 patients

      2021, Joint Bone Spine
      Citation Excerpt :

      Our data indicated that male genders and younger patients were more likely to initiate bDMARDs. While the association between old age and less use of bDMARDs has been consistent across studies including ours [8,9], the effect of gender on bDMARD initiation is complex in terms of not only RA activity but also socio-economic restraints [21,22]. While no particular association between bDMARD initiation and gender was previously found [8,9], socio-economic inequalities in Korea favoring men over women might be one of explanations to our finding [22].

    • Gendercide and femineglect

      2015, Early Human Development
      Citation Excerpt :

      The same findings are evident in several Chinese studies where the issue is exacerbated by the ‘one-family, one-child policy’ [5]. Societal paediatric gender bias extends to similar cultures, including Pakistan [32], Bangladesh [40], Nepal [37] and South Korea [7]. Sporadic reports of gender bias have also arisen in Egypt [51], Tunisia [33,52], Peru [26,53] and Chile [47,54].

    • Health inequalities in Japan: The role of material, psychosocial, social relational and behavioural factors

      2014, Social Science and Medicine
      Citation Excerpt :

      The population attributable risk fraction (‘population proportional attributable risk’ in Kirkwood and Sterne (2003)) due to having a social class not at the highest level or an income below the median income was calculated using the standard formula. Although the comparison of estimates derived from different models using logistic regression is a frequently used technique (Aldabe et al., 2011; Chun, Khang, Kim, & Cho, 2008; Huijts, Eikemo, & Skalicka, 2010; Laaksonen et al., 2005), it is based on the assumption that the residual variation about each logistic regression model is the same across different models since the estimated effects depend on both the regression coefficients and residual variation (Mood, 2010). Sensitivity analyses fitting linear probability models using ordinary least squares regression were conducted to examine this assumption.

    View all citing articles on Scopus
    View full text