Marital Status and Health Among the Elderly

From AcaWiki
Jump to: navigation, search

Citation: Goldman, et al. (1995) Marital Status and Health Among the Elderly.
Internet Archive Scholar (search for fulltext): Marital Status and Health Among the Elderly
Tagged: uw-madison (RSS), wisconsin (RSS), sociology (RSS), demography (RSS), prelim (RSS), qual (RSS), WisconsinDemographyPrelimAugust2009 (RSS), marriage (RSS), health (RSS), elderly (RSS), marriage protection (RSS), marriage selection (RSS)

Summary

Notes: Introduction: Previous research dating back to the mid-nineteenth century indicates that health status and mortality differentials exist between the married and unmarried. There are two main explanations for these differentials: (1) marriage protection-marriage confers benefits which are protective of health and (2) marital selection-healthy people are more likely than unhealthy/disabled people to marry. However, existing studies have not attempted to ascertain whether marital status continues to exert any influence on health and mortality at the older ages. . . The primary goal of this investigation is to determine whether these influences of marital status [e.g., higher SES and more social networks] on health continue to operate at the oldest ages (p. 1717). Methods: Data were taken from the National Health Interview Study: Longitudinal Study of Aging, 70 Years and Over (LSOA), which began with a sample of about 7,500 in 1984 and conducted follow-up interviews every other year until 1990. These data were linked to the Medicare hospitalization records and the National Death Index. Baseline health status was assessed via measures on pre-existing medical conditions, functional limitations and subjective self-reports. Two measures of health status are considered as outcome variables: disability status in the 1990 interview and mortality during the period between the baseline survey (1984) and the 1990 interview (p. 1720). Social support was assessed using a variety of measures (e.g., living arrangements, contact with friends and participation in social activities). Direct measures of wealth are not in the LSOA, but economic well being was measured by proxies such as education, home ownership and private health insurance. Logit models were fit separately for men and women due to theoretical expectations of gender differences (e.g., marital status is not believed to have an equivalent effect on the health of men and women). Findings: Bivariate relations show some differences in baseline health by marital group, although the differences are sometimes unexpected. For instance, widowed men and women are the most likely to report their health status as 'very good' or excellent' (p. 1722). (The authors speculate that this may be a function of selection-institutionalized individuals are not present in this sample of older adults). By contrast, the social environment and economic status of married persons appears to be considerably better than those of their unmarried counterparts. The logit models show that baseline disability status, self-assessed health status and various medical conditions are significantly associated with the odds of dying for men and women, with the expected 'dose-response' relationship (p. 1724). Also, while widowers have about a 25% increased risk of mortality, widows do not exhibit this strong association. Moreover, the odds of dying for divorced and single men and women do not differ significantly from those of the married (p. 1725). Economic factors appear to have little association with mortality for men, but for women the effects are relatively large. Importantly, all aspects of the social environment are significantly related to the probability of dying for both genders. In terms of disability risk, widowers face odds 1.8 times greater than married men. Widows do not face this large increase in disability risk, which leads the authors to suggest that marriage benefits husbands to a greater degree than it benefits wives in various ways that are ultimately related to health (p. 1726). Strangely, never married women and divorced men are less likely than their married counterparts to report a disability at follow-up. Economic factors appear to play an important role in disability, while social factors are less relevant. Conclusions: Our results suggest that marital status is indeed associated with health and survival outcomes at the oldest ages. With regard to mortality, the effects appear to be modest and are only significant among widowed males. Consistent with the mortality models, we also find that widowed men have much higher risks of being disabled than do married men, while the corresponding differential for women is considerably smaller (p. 1727).