Effects of husbands’ and wives’ education on each other's mortality

Abstract

Education is an important predictor of one's own cardiovascular disease (CVD) and overall mortality. Little is known, however, regarding the effect of other individuals, specifically a spouse, on these risks. In the present study, we examine the contribution of a spouse's educational attainment and the effect of educational discrepancy between spouses on CVD and overall mortality. Data were taken from the Israel Longitudinal Mortality Study, which linked a 20% sample of the 1983 census to mortality records through 1992. The study cohort comprised 37,618 married couples aged 45–69 years. During the 9.5-year follow-up 6058 men and 2568 women died. Overall and CVD mortality hazard ratios were calculated using Cox proportional hazard regression models. We found that the educational attainment of both spouses were significant predictors of one's own overall mortality. For CVD mortality, however, a wife's educational attainment was a stronger predictor of her husband's risk of dying than his own educational level, while for women a husband's education had little affect. Educational discrepancy between partners did not affect overall mortality and had a varied effect on CVD mortality by sex. Specifically, highly educated women had an almost two-fold increased risk of CVD mortality when married to less educated husbands, while lesser-educated women were not affected by their spouses’ educational attainment. Spouses’ education adds valuable information when assessing mortality differentials among married persons, and socioeconomic characteristics of one's immediate family are important influences on one's health.

Introduction

Mortality differentials associated with educational achievement have been widely documented among individuals living in industrialized countries (Borrell, Regidor, Arias, Navarro, Puigpinos, Dominguez et al., 1999; Elo & Preston, 1996; Krokstad, Kunst, & Westin, 2002; Kunst & Mackenbach, 1994; Mackenbach, Kunst, Groenhof, Borgan, Costa, Faggiano et al., 1999; Manor, Eisenbach, Israeli, & Friedlander, 2000; Manor, Eisenbach, Friedlander, & Kark, 2004). For overall and cardiovascular disease (CVD) mortality, increasing years of education confer a protective effect (Kaplan & Keil, 1993; Kawachi, Marshall, & Pearce, 1991; Mackenbach, Cavelaars, Kunst, & Groenhof, 2000; Manor et al., 2004; Phelan, Link, Diez-Roux, Kawachi, & Levin, 2004; Turrell & Mathers, 2001). This effect is primarily related to occupational and economic conditions (Elo & Preston, 1996; Lantz et al., 1998), health behaviors (Lantz et al., 1998; Lynch, Kaplan & Salonen, 1997), and psychosocial circumstances (Lynch et al., 1997; Phelan et al., 2004). Additionally, the effects of education on mortality vary by age (Borrell et al., 1999; Elo & Preston, 1996; Manor, Eisenbach, Friedlander, & Kark (2004), Manor, Eisenbach, Israeli, & Friedlander (2000)), gender (Elo & Preston, 1996; Mackenbach et al., 1999; Manor et al., 2000), race (Elo & Preston, 1996; Howard, Anderson, Russell, Howard, & Burke, 2000), and marital status (Goldman, Korenman, & Weinstein, 1995).

Marriage represents an important dimension of social relationships (Kiecolt-Glaser & Newton, 2001; Rogers, 1996; Berkman & Syme, 1979). It benefits individuals by providing support and social networks that affect morale, self-worth, and life satisfaction. Furthermore, marriage positively influences health behaviors such as diet, exercise, and utilization of medical and social services (Joung, van der Meer & Mackenbach, 1995; Umberson, 1992; Wyke & Ford, 1992).

Significantly less is known about the contribution of a partner's risk characteristics, such as their socioeconomic status (SES), to one's own health (Bosma, Appels, Sturmans, Grabauskas, & Gostautas, 1995; Carmelli, Swan, & Rosenman, 1985; Eaker, Haynes, & Feinleib, 1983; Egeland, Tverdal, Meyer, & Selmer, 2002; Haynes, Eaker, & Feinleib, 1983; McDonough, Williams, House, & Duncan, 1999; Medalie & Goldbourt, 1976; Monden, van Lenthe, de Graaf, & Kraaykamp, 2003; Stolzenberg, 2001; Strogatz, Siscovick, Weiss, & Rennert, 1988; Suarez & Barrett-Connor, 1984). Numerous studies have looked at the wife's influence on her husband's wellbeing (Bosma et al., 1995; Carmelli et al., 1985; Eaker et al., 1983; Egeland et al., 2002; Haynes et al., 1983; Medalie & Goldbourt, 1976; Strogatz et al., 1988; Suarez & Barrett-Connor, 1984), while there is very limited evidence about the husband's influence on his wife's health (Monden et al., 2003; Stolzenberg, 2001; McDonough et al., 1999). Conflicting results have been reported regarding the effect of an educated wife on her husband's health (Bosma et al., 1995; Egeland et al., 2002; Haynes et al., 1983; Strogatz et al., 1988). In a recent study that examined the effects of both partner's education on health, educated husbands and wives contributed in a positive direction to their spouses self-assessed health and smoking behavior but not excessive alcohol consumption (Monden et al., 2003).

Notwithstanding the effects of a spouse's education on morbidity and mortality, scarce data are available regarding the health risks associated with an educational gap between spouses. Researchers hypothesized that the effects of educational status discrepancy, whereby perceived or actual inconsistent rankings between two individuals produce psychological stress and tension, result in adverse health (Hurst & Guldin, 1981; Vernon & Buffler, 1988).

In the present study, mortality differentials for married couples were examined using data from the Israel Longitudinal Mortality Study (ILMS), 1983–1992. Specific goals were (a) to assess the contribution of a spouse's educational attainment on one's overall and CVD mortality and (b) to determine if educational discrepancy between spouses affects mortality.