Sociodemographics, self-rated health, and mortality in the US
Introduction
Self-report health measures play an increasingly important role in clinical practice, research and policy. The utility of self-rated measures derives from their validity, reflected, in part, by their relationship to clinical conditions, indicators of morbidity, and mortality. There is some evidence of their ability to predict important clinical outcomes beyond that of objective indicators (Bergner & Rothman, 1987; Idler & Benyamini, 1997). Generic health status measures also permit assessment of health across a range of illnesses, thereby providing measurements that can reflect population health in both clinical and community settings. These attributes afford managed care organizations, employers, public health professionals and regulators the potential to monitor the outcome and quality of a broad range of clinical care and public health programs.
A number of studies from both the US and elsewhere report on aspects of sociodemographic influences on health status, and their relationships with mortality. Outside of the US, a number of nationally representative studies have pointed to the lower health status of women, and those who are socioeconomically disadvantaged (Kind, Dolan, Gudex, & Williams, 1998; Denton & Walters, 1999; Humphries & van Doorslaer, 2000). Most of these studies, however, do not fully examine adjusted socioeconomic effects, so the contribution of specific socioeconomic indicators to health status is largely unknown. In the US, little work has been done in characterizing health status disparities using nationally representative samples and Stewart and Napoles-Springer (2000) have called attention to the need for better understanding about how these measures perform in diverse social groups (Stewart & Napoles-Springer, 2000). A number of studies using regional or selected US samples have reported on the lower health status of minorities, those with lower socioeconomic status (SES), women, and older persons (Burdine, Felix, Abel, Wiltraut, & Musselman, 2000; Andresen & Brownson, 2000; Cunningham, Hays, Burton, & Kington, 2000). National surveys, such as the National Health Interview Surveys, routinely report the associations of responses to the single general self-rated health question “In general, how would you rate your health?” with sociodemographic descriptors and mortality (Narrow, Rae, Moscicki, Locke, & Regier, 1990; Mendoza et al., 1991; Wolinsky & Stump, 1996; Ferraro & Farmer, 1996; Idler & Benyamini (1996a), Ren & Amick (1996b); Idler & Benyamini, 1997; Liao et al., 1998; McGee, Liao, Cao, & Cooper, 1999; Chandola & Jenkinson, 2000). The usefulness of this information is compromised by the narrow range of possible responses (five), and the limited adjustment provided for other sociodemographic factors.
Despite their growing use, there is little information about how sociodemographic factors affect how persons understand and prioritize components of their perceived health; there is, though, evidence that age, gender, and race affect how people respond to the single general self-rated health question (Idler, 1993; Krause & Jay, 1994; Jylha, Guralnik, Ferrucci, Jokela, & Heikkinen, 1998). In addition, there has been little systematic reporting in the US on how these measures behave across sociodemographic sub-populations or the extent to which they demonstrate predictive validity for outcomes in different socioeconomic and racial/ethnic groups. To address these limitations, we report on the relationships among four measures of self-reported health (health perceptions, physical functioning, role functioning, and mental health) and sociodemographic factors for a representative sample of the US population. To explore the predictive validity of these measures in different sociodemographic groups, we also examine the adjusted relationships of the self-rated health measures with subsequent mortality. A simple model underlies these analyses. Sociodemographics factors (age, sex, race/ethnicity, education, and income) are viewed as primarily affecting each of the measured domains of self-reported health. Together, sociodemographic factors and each of the domains of self-reported health are viewed as affecting subsequent mortality.
Section snippets
Methods
Data from the Household Survey component of the National Medical Expenditure (NMES)(Edwards & Berlin, 1989) were used in this analysis. The Household Survey consisted of a 1 year, cross-sectional survey of nearly 35,000 individuals from approximately 14,000 households representing the 1987 US civilian, non-institutionalized population. The survey used a stratified, multistage area probability design with over-sampling of minorities, poor persons, disabled persons, and the elderly. Details of
Results
The unadjusted relationships of each of the self-reported health scales with the sociodemographic variables are shown in Table 1. Lower self-reported health, except for mental health, is associated with increasing age. Women report, lower health status than men on all subscales. Compared with whites, blacks report, lower health status, whereas Latinos reported higher physical function. More education and income were both associated with higher levels of self-reported health across domains.
The
Discussion
Lower SES and being black are associated with lower reported health status and increased mortality. Women report lower health status but exhibit lower mortality, and Latinos report increased health status and experience lower mortality. Increasing age is associated with declines in self-rated health, except mental health which increases slightly with age; mental health, however, is the only health status measure that shows no relationship with mortality after adjustment for other health status
References (64)
- et al.
A comparison of the relationships of education and income with mortalityThe national longitudinal mortality study
Social Science & Medicine
(1999) Health and the social status of blacks in the United States
Annals of Epidemiology
(1993)- et al.
Gender differences in structural and behavioral determinants of healthAn analysis of the social production of health
Social Science & Medicine
(1999) - et al.
Race, socioeconomic status, and cause-specific mortality
Annals of Epidemiology
(2000) - et al.
Income-related health inequality in Canada
Social Science & Medicine
(2000) - et al.
Sex and depression in the National Comorbidity Survey. I: Lifetime prevalence, chronicity and recurrence
Journal of Affective Disorders
(1993) - et al.
Gender differences in healthAre things really as simple as they seem?
Social Science & Medicine
(1996) - 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) - et al.
Black–white mortality differences by family income
Lancet
(1992) - Agency for Health Care Policy and Research. (1997). National Medical Expenditure Survey, 1987: Data From the Household...