Short ReportReliability of perceived health by sex and age
Introduction
A single question on perceived general health is an often used item in health surveys. Perceived health seems to summarize a broad range of health related information available to the individual (Tissue, 1972; Maddox and Douglas, 1973; Jylhä et al., 1986; Johnson and Wolinsky, 1993). These include medically confirmed information on the individual's health status and diagnosed chronic conditions as well as functional limitations and disability. Perceived health may also be influenced by more subtle knowledge of family history of chronic disease, behavioral and life style characteristics relevant for health as well as cognitive and affective psychosocial characteristics of individuals and personality.
In health research the usefulness of perceived health is partly based on its relevance as an instrument in monitoring population health. Furthermore, perceived health has been found to predict health care use (Roos and Shapiro, 1981; Hulka and Wheat, 1985; Krakau, 1991; Fylkesnes, 1993) and mortality (Mossey and Shapiro, 1982; Kaplan and Camacho, 1983; Idler and Kasl, 1991; Idler, 1992; Wolinsky and Johnson, 1992; McCallum et al., 1994; Appels et al., 1996; Kaplan et al., 1996). These strong associations indicate that perceived health is a valid measure of health (i.e. perceived health does measure health), but it does not necessarily give a clear indication of the reliability of perceived health. To be valid, a measure has to be reliable, but the extent of reliability can still vary.
Thus, before more detailed evaluation of the possible reasons for the relationship between perceived health, health services use and mortality can be undertaken the reliability of the measure of perceived health should be better known. Very few studies have addressed this issue although it is crucial for empirical research (Lundberg and Manderbacka, 1996). The aim of this study is to assess the test–retest reliability of perceived health by sex and age, the two most central unmodifiable demographic dimensions of all study populations.
Section snippets
Population sample
The study analyses data from the comprehensive Mini-Finland Health Survey which has been carried out on a nationally representative sample of 8000 adults, 3637 men and 4363 women, aged 30 and over. The subjects were invited to attend a personal health interview and a two phase health examination in 1978–1980 (Aromaa et al., 1989). The invitation to the health examination included a self-administered questionnaire. Altogether 7703 persons (96%) were interviewed by public health nurses in their
Results
On the five category scale the distribution of perceived health differed somewhat between the personal interview and the questionnaire (Table 1). Among men a lower proportion of the three intermediate categories and a higher proportion of the categories good and poor was observed in the personal interview than in the self-administered questionnaire. Also among women a lower proportion reported fairly good and intermediate health in the interview.
In Table 2, the two measures of perceived health
Discussion
The population studied in the Mini-Finland Health Survey was a large nationally representative sample and participation rates were very high except in the oldest age group. The two measurements of perceived health were obtained at two separate occasions with an interval of about 1–6 weeks and can be considered technically independent. The large sample enabled us to study reliability in different age and sex groups, which has rarely been done.
We measured reliability in terms of agreement and κ
Acknowledgements
The work was supported by the Academy of Finland (grants 1018386 and 37631) and the Gyllenberg Foundation. The authors gratefully acknowledge the work of the public health nurses and the whole field team of the Mini-Finland Health Survey. We also acknowledge the invaluable comments from Kristiina Manderbacka on earlier drafts of the manuscript.
References (23)
- et al.
Self-rated health and mortality in a Lithuanian and a Dutch population
Social Science & Medicine
(1996) - et al.
Reliability and validity of interview data on chronic diseases: the Mini-Finland Health Survey
Journal of Clinical Epidemiology
(1993) - Aromaa, A., Heliövaara, M., Impivaara, O., Knekt, P., Maatela, J., Joukamaa, M., Klaukka, T., Lehtinen, V., Melkas, T.,...
- Fleiss, J.L., 1981. Statistical Methods for Rates and Proportions. John Wiley & Sons, New...
Determinants of health care utilization: visits and referrals
Scandinavian Journal of Social Medicine
(1993)- et al.
Patterns of utilization: the patient perspective
Medical Care
(1985) - et al.
Health perceptions and survival: do global evaluations of health status really predict mortality?
Journal of Gerontology
(1991) - Idler, E.L., 1992. Self-assessed health and mortality: a review of studies. In: Maes, S., Leventhal, H., Johnston, M....
- et al.
The structure of health status among older adults: disease, disability, functional limitation and perceived health
Journal of Health and Social Behaviour
(1993) - et al.
Self-rated health and associated factors among men of different ages
Journal of Gerontology
(1986)