Paradox Lost? Interpreting Gender Differences in Life and Health Expectancy

Marc Luy , Vienna Institute of Demography
Paola Di Giulio, Vienna Institute of Demography, Wittgenstein Centre (IIASA, OeAW, University of Vienna)
Vanessa di Lego, Vienna Institute of Demography
Yuka Minagawa, Sophia University

Starting in the 1920s, a common wisdom about a paradox in gender differences in health and mortality emerged, best exemplified by Lorber and Moore’s (2002) well-known paraphrase "women get sicker, but men die quicker". We hypothesise that this gender paradox can be explained by two factors: the 'mortality effect' (ME) resulting from women's higher life expectancy and 'different item functioning' (DIF), i.e., gender differences in health reporting behaviour. To test this hypothesis, we estimate unhealthy life years (ULY) at age 50 for the three health domains (HD) of self-perceived health, activity limitation, and chronic morbidity using the Sullivan method (1971). Data on gender- and age-specific prevalence of the three HD conditions is taken from the Survey of Health, Age and Retirement in Europe (SHARE). Life tables are from the Human Mortality Database. Gender differences in ULY are then decomposed into the effects due to differences in mortality (ME) and health (health effect, HE) with the method proposed by Nusselder and Looman (2004). Finally, the HE is adjusted for DIF using the approach of Luy et al. (2023) on the basis of anchoring vignettes from SHARE. Data for Germany shows that the gender gap in ULY can indeed be fully explained by ME and DIF for all three HDs. We will extend this analysis to seven other European populations (Belgium, France, Greece, Italy, the Netherlands, Spain, and Sweden) to test whether and to what extent these findings can be generalised.

See extended abstract

 Presented in Session 120. Flash session Gender Differences in Health, Wellbeing and Morbidity