Gender differences in old age mortality: Roles of health behavior and baseline health status

https://doi.org/10.1016/S0895-4356(03)00060-XGet rights and content

Abstract

This research aims to further current understanding of gender differences in old age mortality. In particular, it assesses the relative importance of health behavior and baseline health conditions in predicting the risk of dying, and how their effects differ between elderly men and women. Data for this research came from a prospective study of a national sample of 2,200 older adults in Japan from 1987 to 1999. Hazard rate models were employed to ascertain the interaction effects involving gender and health behavior (i.e., smoking and drinking) and baseline health status. Gender differences in old age mortality in the Japanese are quite pronounced throughout all of our models. In addition, interaction effects of gender and smoking, functional limitation, and cognitive impairment, indicate that females in Japan suffer more from these risk factors than do their male counterparts. Failure to adjust for population heterogeneity may lead to a significant underestimation of female advantage in survival. The inclusion of health behavior and health status measures only offsets a limited proportion of this gender differential. The increased mortality risk due to smoking, functional limitation, and cognitive impairment among elderly Japanese women suggests that narrowing of gender gap in mortality may be due to not only changes in the levels of these risk factors but also their differential effects on men and women.

Introduction

A number of hypotheses have been offered to explain women's advantage in survival by emphasizing the importance of biologic, behavioral, and sociocultural factors [1], [2]. According to the biologic explanation, genetic and hormonal differences between the sexes are pivotal in understanding the gender gap in health [3], whereas from a behavioral perspective, differences in the engagement of risk and health behaviors as well as the exposure of stress are critical. Finally, the levels of social development and relative status of women are considered to be important in accounting for the differences in mortality between the genders. Because of the observed gender variations over time, across countries and cultures, and in terms of socioeconomic status, biologic factors cannot be relied upon exclusively in accounting for sex differences in mortality [1], [4], [5].

As extensive as current research has been, it can be improved in several respects. First, gender differences in mortality have been generally attributed to differentials in the exposure of various risk factors. Yet, it is quite possible that the magnitude and direction of these effects may differ significantly among men and women. Do smoking and drinking influence the risk of dying the same way for men and women? Do baseline health conditions have the same effects in predicting the survival of men and women? Accordingly, women's relatively favorable survival may be attributed not only to lower levels of exposure to the same risk factors but to different responses to these risk factors or to a different set of strengths or vulnerabilities altogether [1].

Current understanding of gender differences in the effects of major risk factors of mortality is quite limited. We are able to locate only a handful of studies related to this issue. For instance, based on data from the NHANES I Follow-up Survey from 1971–1975 to 1982–1984, smoking and drinking both lead to greater risks of dying among older men (65–74 years old) than older women [6]. In addition, Lapane and associates [7] reported that elderly men with Alzheimer Disease (AD) in nursing homes had an increased risk of mortality relative to women, adjusted for age and race. Finally, according to a recent study in Spain, women who suffered from their first acute myocardial infarction (AMI) experienced more lethal and severe AMI and a greater risk of dying at 28 days and 6 months [8]. All three studies were based on data from Western developed nations with two of them based on samples of patients only.

Second, evidence pertaining to gender differences in health in non-Western societies is sparse [9]. In fact, there is little empirical research suggesting that explanations for gender differences in health observed in the United States can be extended to other societies. Using data from the United States, Verbrugge [9], [10] suggested that women have high rates of acute illnesses and of most nonfatal chronic conditions, whereas men experience higher prevalence rates of the leading fatal conditions. However, based on two large community surveys in Britain, few gender differences in health have been observed [11]. As another example, many variables, which according to the U.S. research literature are important components of the explanation for gender differences in morbidity, turn out not to be so in a sample of over 2,000 married adults in Thailand [12].

According to Nathanson [1], [2], the gender differential in life expectancy is predominately contributed by deaths of individuals over 65. As a preliminary specification, the explanatory problem posed by the gender gap might be narrowed to the problem of explaining gender differences in survival in old age. For instance, mortality among persons 50 years of age or over accounts for 87% of all deaths in the United States in 1995 and 93% in 1998 in Japan [13]. Furthermore, as gender differences in health are particularly evident in chronic diseases and disability [10], it makes sense to focus on the stage of life course where chronic health problem are most likely to arise. Many chronic conditions are linked to lifestyle factors, such as smoking and drinking, and the health disadvantages of undesirable lifestyle choices are cumulative and often take years to manifest [14]. Consequently, focusing on those who have reached later life should provide the best vantage point for assessing gender differences in mortality. The present research aims to contribute to current understanding regarding gender differences in mortality by analyzing the data from a 12-year prospective study of a national sample of 2,200 individuals 60 years of age and over in Japan.

Given that Japanese elderly and society differ significantly from their counterparts in the United States and other Western nations, Japan provides an interesting context for further research on gender differences in old age mortality, particularly in contrast with the United States. Both nations are similar in economic development and are experiencing rapid population aging. The Japanese have the highest life expectancy at birth (77 years for males and 83.6 years for females in 1995) in the world, and can live about 4 years longer than Americans (72.7 years for males and 79.4 years for females in 1995) [15]. What is so remarkable is that the high life expectancy in Japan was only accomplished during the last 4 decades [16]. In terms of the major causes of death, Japan has significantly lower rates of death due to heart disease, lung cancer, and breast cancer than the United States. On the other hand, death rates due to stomach cancer and stroke are higher in Japan than the United States [13].

Among developed countries Japan is a persistent outlier in terms of women's status [17]. In Japan the male–female wage gap is greater in that salaries among full-time female employees were 63% of their male counterparts [18]. Although high school seems to have become the minimum acceptable level of education for both sexes in Japan, men are far more likely to go on to 4-year universities than do women. The gap has narrowed considerably, but even in 1997, 26% of women and 43% of men advanced from high school to universities [18].

Japan also differs from other developed nations in terms of the health behaviors of its population and health care system. In 1999, 54% of adult male and 15% of adult female in Japan were smokers [13]. The rate of smoking for Japanese male is higher than most of the developed nations, while the rate for Japanese female is lower than their counterparts in other developed nations. On the other hand, the per capita consumption of pure alcohol among adults (15 years of age or over) was 7.9L in 1996, which was lower than those in France (13.7L), Germany (11.7L), Britain (9.4L), and the United States (8.9L) [19]. Finally, health care delivery also differs significantly in Japan. The Japanese spend more on prescription drugs than any Western nations [20]. More importantly, Japanese medicine is more pluralistic in that Western biomedicine is well integrated with traditional Chinese medicine [21].

Although there are numerous studies of old age mortality in Japan, very little attention has been directed to the interaction between gender and the effects of health behavior and health status on mortality [22], [23], [24], [25]. Frequently, key risk factors are not included. In the following, a conceptual model is proposed to examine gender differences in the linkages between old age mortality and the factors such as health behavior and health status.

Section snippets

Model specifications and hypotheses

To analyze the interaction effects involving gender, health behavior, and baseline health status on the risk of dying, a conceptual model is proposed. Specifically, factors associated with mortality are differentiated into four domains including: (a) demographic variables (age, gender, and community size), (b) socioeconomic status (education, income, and home ownership), (c) social relationships (marital status, employment, household size), and (d) health related factors such as health behavior

Sample and data

Data came from a prospective cohort study of older adults in Japan. With a two-stage stratified probability sampling design, 2,200 Japanese aged 60 and over were interviewed in November 1987. These individuals were followed up once every 3 years thereafter (i.e., 1990, 1993, 1996, and 1999). The response rate for the baseline survey was 69%, reflecting 2,200 of the 3,185 elderly Japanese who were alive and noninstitutionalized at contact. A comparison of this sample with the 1985 Japanese

Descriptive gender differences

Table 1 presents the descriptive statistics by gender. Between 1987 and 1999, mortality among men was 46%, which was significantly higher than that among women (28%). With a few exceptions, there are significant gender differences in all risk factors at baseline. For instance, in contrast to their male counterparts, elderly Japanese women have lower educational attainment, and are less likely to be married and employed. In terms of health behavior, 61% of elderly Japanese men are current users

Discussion

Interpreting gender differences in old age mortality without adjusting for population heterogeneity is likely to lead to a substantial underestimation, because women suffer from significant disadvantages in education, martial status, and employment (Model 2, Table 2). This is consistent with the observation that the disadvantageous social and economic position of women in U.S. society keeps the sex mortality gap from being even larger [26]. Our finding extends the generalizability of the

Acknowledgements

This research was supported by grant R37-AG154124 (Jersey Liang, PI) from the National Institute on Aging. Additional support was provided through a grant from the Japanese Ministry of Health, Labor and Welfare Longevity Foundation and by the Tokyo Metropolitan Institute of Gerontology. Rod Little contributed valuable insights regarding multiple imputation of missing data. Ken Warner, Carol Weisman, Paula Lantz, and Mary Haan provided useful suggestions concerning the interpretations of the

References (51)

  • K.L. Lapane et al.

    Gender differences in predictors of mortality in nursing home residents with AD

    Neurology

    (2001)
  • J. Marrugat et al.

    Mortality differences between men and women following first myocardial infarction

    JAMA

    (1998)
  • L. Verbrugge

    Gender and health: an update on hypotheses and evidence

    J Health Soc Behav

    (1985)
  • L. Verbrugge

    The twain meet: empirical explanation of sex differences in health and mortality

    J Health Soc Behav

    (1989)
  • T.D. Fuller et al.

    Gender and health: some Asian evidence

    J Health Soc Behav

    (1993)
  • Japan Health and Welfare Statistics Association

    Journal of health and welfare statistics 2000

    (2000)
  • C. Ross et al.

    Education, age, and the cumulative advantage in health

    J Health Soc Behav

    (1996)
  • OECD

    OECD health data 98: a comparative analysis of twenty-nine countries

    (1998)
  • R.G. Evans et al.

    Why are some people healthy and others not? The determinants of health of populations

    (1994)
  • M.C. Brinton

    Gender stratification in contemporary urban Japan

    Am Sociol Rev

    (1989)
  • Ministry of Foreign Affairs of Japan. Part one: situation of Japanese women....
  • World Health Organization

    Global status report on alcohol

    (1999)
  • L.A. Graig

    Made in Japan: universal health insurance

  • M. Lock

    Health, illness, and medical care in Japan: cultural and social dimensions

    (1987)
  • Y. Honma et al.

    Active life expectancy, life expectancy and ADL in Japanese elderly

    Jpn J Public Health

    (1998)
  • Cited by (39)

    • 1H-NMR metabolomics-based surrogates to impute common clinical risk factors and endpoints

      2022, eBioMedicine
      Citation Excerpt :

      While eight surrogates contributed to the male model, including ‘lipid medication’, ‘high age’, ‘high hsCRP’, and ‘low hdlchol’, only three surrogates contributed to the mortality prediction in females: ‘high hsCRP’, ‘high triglycerides’, and ‘low hemoglobin’. These findings are in line with previous reports that different risk factors seem to predict survival up to the highest ages for the different sexes.47–49 However, we add a more extensive discussion upon the selection of ‘high age’ only in men in Supplementary Material 1 (document S3).

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

      2014, Social Science and Medicine
      Citation Excerpt :

      In Japan, although there are many studies of factors linking SEP and health, to the best of our knowledge, none of these studies have explicitly tested all four dimensions of mechanisms simultaneously or calculated the extent of the attenuation in health inequalities by factors included. Health inequalities have most extensively examined in relation to education (Aida et al., 2011; Fujino, Iso, et al., 2005; Fujino, Tamakoshi, et al., 2005; Fujisawa, Hamano, & Takegawa, 2009; Hamano et al., 2010; Hirokawa, Tsutusmi, & Kayaba, 2006; Honjo, Tsutsumi, & Kayaba, 2010; Ichida et al., 2009; Ito et al., 2008; Iwasaki et al., 2002; Liang et al., 2002; Liang, Bennett, Sugisawa, Kobayashi, & Fukaya, 2003; Nishi et al., 2012; Wang et al., 2005) and, to a lesser extent, income (Liang et al., 2002, 2003; Oshio & Kobayash, 2009; Wang et al., 2005). Educational and income inequalities in health have been most consistently found in all-cause mortality and subjective health status, and these associations are attenuated somewhat by health behaviours, biomarkers, occupational factors and stress.

    View all citing articles on Scopus
    View full text