Elsevier

Social Science & Medicine

Volume 51, Issue 12, 15 December 2000, Pages 1741-1754
Social Science & Medicine

Health behaviours and health: evidence that the relationship is not conditional on income adequacy

https://doi.org/10.1016/S0277-9536(00)00107-6Get rights and content

Abstract

This study used Canadian data to examine whether the relationships between two health behaviours (physical activity and smoking) and two measures of health (self-perceived health status and number of chronic health conditions) are conditional on income adequacy. Studies that have investigated the manner in which socioeconomic circumstances, such as income adequacy, and health behaviours interact to influence health are few in number and characterized by inconsistent findings. In addition, there is a complete absence of published Canadian research that has explored these relationships. I investigated the relationship between health behaviours and health by income adequacy with a secondary analysis of data from the first cycle of the National Population Health Survey (NPHS), conducted by Statistics Canada in 1994-95. The sample consisted of 11,941 NPHS respondents 20–64 years of age who did not have an illness or disability that prevented them from being employed. As a whole, findings from a series of hierarchical multiple regression analyses did not provide adequate evidence to conclude that the effects of physical activity and smoking on self-perceived health status and chronic health conditions are conditional on income adequacy. Instead, findings showed that the health behaviours each had a similar degree of influence on the self-perceived health status and number of chronic health conditions of respondents at all income adequacy levels. Moreover, the magnitude of the relationships between the health behaviours and health measures was very small. By enhancing knowledge about both the nature and magnitude of the relationships among Canadians’ income adequacy, health behaviours, and health, this study makes a significant contribution to the small body of research that has explored the possibility that the relationship between health behaviours and health varies by socioeconomic circumstances. I conclude the paper with a discussion of the implications that the findings have for public health policies and programs.

Introduction

This study investigated the interaction effect of income adequacy and health behaviours upon the health of Canadians. Its impetus is rooted in some of the findings from a previous study in which I examined the relationships among poverty status,1 health behaviours, and health. As part of the previous study, I explored the relationships that both health promoting and health damaging behaviours have with the health of people living in working poor families and those living in families receiving social assistance. Findings from multivariate analyses indicated that, for the most part, there was no statistical association between the health of study participants and either health promoting behaviours (exercise, home dental care, sleep, preventative dental care, and preventative medical care) or health damaging behaviours (smoking and alcohol consumption) (Williamson, 1995).

These findings differed substantially from what has become conventional wisdom about the role that individual health behaviours play in the determination of health. Results from numerous studies conducted during the past two decades have shown consistently that health is positively associated with behaviours such as physical and recreational activities, optimal body weight for height, and preventative medical examinations (Adams, 1993, Belloc, 1973, Belloc and Breslow, 1972, Blair, 1993, Blair et al., 1992, Kooiker and Christiansen, 1995; Mackenbach, Van Den Bos, Joung, Van De Mheen & Stranks, 1994; Segovia et al., 1989, Stephens, 1986, Stephens, 1993). As well, results have shown that health is negatively associated with behaviours such as smoking (Adams, 1993, Belloc, 1973, Belloc and Breslow, 1972, Kooiker and Christiansen, 1995, Mackenbach et al., 1994, Pederson, 1993, Stephens, 1986). The studies from which these findings were obtained utilized an assortment of health indicators, research designs, and statistical analyses. Indeed, it is curious that the findings from my study about poverty status, health behaviours, and health did not coincide with other findings about the influence that health behaviours have on health. Why is it that health promoting and health damaging behaviours did not play a role in the health of people living in two groups of poor families?

One explanation is suggested by the findings from a large study by Blaxter (1990), in which data from the 1984/85 British Health and Lifestyle Survey were used to explore the relationships between health behaviours and health by socioeconomic circumstances. Blaxter’s (1990) principle discovery was that both health promoting and health damaging behaviours had the greatest impact on the health of respondents with the most favourable socioeconomic circumstances, and relatively less influence on the health of respondents with less favourable circumstances. These results suggest that the findings from my study about poverty status, health behaviours, and health may not have coincided with other research results about the influence that health behaviours have on health because health behaviours have less of an influence on the health of people living in poverty2 than on the health of those who have adequate incomes. In other words, the relationship between individual health behaviours and health may be conditional on income adequacy. This explanation suggests that since all of the participants in my study were living in poverty, the effect that their health behaviours had on their health may have been too small to be demonstrated by the statistical analyses that I employed.

However, the results from other studies that have explored the manner in which health is influenced by the interaction of socioeconomic circumstances and health behaviours (Davey Smith and Shipley, 1991, Kooiker and Christiansen, 1995, Marmot, 1986, Marmot et al., 1984) have not been consistent with the findings from Blaxter’s (1990) study. Davey Smith and Shipley, 1991, Marmot, 1986, and Marmot et al. (1984) analysed data from the Whitehall Study and found that age standardized 10-year mortality rates from all causes and from both coronary heart disease and lung cancer among civil servants who smoked were inversely associated with socioeconomic circumstances within the British civil service hierarchy. That is, mortality rates of civil servants who were current smokers increased as occupational status declined (Davey Smith & Shipley, 1991; Marmot, 1996; Marmot et al., 1984). Even though these results and those by Blaxter (1990) all showed that the relationships between health behaviours and health vary by socioeconomic circumstances, findings about the manner in which health damaging behaviours interact with socioeconomic circumstances to influence health differ. Findings from the analysis of the Whitehall data suggest that the negative impact that smoking has on health is more pronounced at lower socioeconomic levels than at higher socioeconomic levels (Davey Smith and Shipley, 1991, Marmot, 1986, Marmot et al., 1984). In contrast, Blaxter’s (1990) results imply that the degree of influence that health damaging behaviours have on health is positively associated with socioeconomic circumstances.

Also, it is noteworthy that Kooiker and Christiansen (1995), who used large data sets from the Netherlands and from Denmark to explore inequalities in health by focusing on the interaction of socioeconomic circumstances and health behaviours, were unable to replicate Blaxter’s (1990) findings. Contrary to results by Blaxter, 1990, Kooiker and Christiansen, 1995 found that the relationship between health behaviours and health did not differ by socioeconomic circumstances. Health behaviours had a similar influence on the health of both socioeconomically deprived and privileged respondents.

In summary, studies that have explored the manner in which socioeconomic circumstances and health behaviours interact to influence health are few in number and are characterized by inconsistent findings. In addition, there is a complete absence of published Canadian research that has investigated whether the relationships between health behaviours and health are conditional on socioeconomic circumstances. Thus, in an attempt to advance knowledge about the relationships among socioeconomic circumstances, health behaviours, and health as well as to build upon my previous research about poverty status, health behaviours, and health, the purpose of the current study was to use Canadian data to explore the interaction of income adequacy and health behaviours upon health. The specific objectives of this study were to: (1) determine whether relationships between health behaviours and health are conditional on income adequacy; and (2) determine the manner in which the relationships between health behaviours and health differ by income adequacy.

Section snippets

Methods

To investigate the relationships between health behaviours and health by income I conducted a secondary analysis of data from the first cycle of the National Population Health Survey (NPHS). The NPHS, which is carried out by Statistics Canada, provides the most current data about the health behaviours and health of Canadians (Statistics Canada, 1995). Complex sampling techniques that included stratification and multiple stages of selection were used to obtain a sample of 17,626 Canadians who

Results

Results from model 2 of each of the hierarchical multiple regression analyses that were used to explore the relationships between the health behaviours and health by income adequacy are summarized in Table 4, Table 5. Table 4, Table 5 indicate that there was not a linear interaction effect between the two health behaviours and income adequacy upon the health measures. A linear interaction effect would be demonstrated by a progressive increase or decrease, across income adequacy levels, in: (1)

Discussion and conclusions

With only a handful of studies that are characterized by contradictory findings, it is evident that researchers are still in the early stages of exploring the possibility that the relationships between health behaviours and health vary by socioeconomic circumstances. The study reported here enhances knowledge about both the nature and magnitude of the relationships among Canadians’ income adequacy, health behaviours, and health — and so is significant.

As a whole, the findings from this study do

Acknowledgements

Financial support for this study by an operating grant (#81-62755) from the Small Faculties Research Grant Program, University of Alberta, is gratefully acknowledged. I would also like to thank Judee Onyskiw and Dorothy Forbes for their assistance with the data analysis for this study. And, I appreciate the valuable feedback that Jerry Kachur, Anne George, and two anonymous reviewers provided about earlier drafts of this paper.

References (54)

  • A. Antonovsky

    Unravelling the mystery of health. How people manage stress and stay well

    (1987)
  • Awid, Y., Israeli, N. (1997). Poverty and income inequality: An international comparison, 1980s and 1990s. (On line:...
  • N.B. Belloc et al.

    Relationships of physical health status and health practices

    Preventive Medicine

    (1972)
  • S.N. Blair

    C.H. McCloy research lecture: Physical activity, physical fitness and health

    Research Quarterly for Exercise and Sport

    (1993)
  • S.N. Blair et al.

    How much physical activity is good for health?

    Annual Review of Public Health

    (1992)
  • M. Blaxter

    Health and lifestyles

    (1990)
  • S.J. Bondy et al.

    Low-risk drinking guidelines: The scientific evidence

    Canadian Journal of Public Health

    (1999)
  • Bradshaw, J., Chen, J. (1998). Poverty in the UK: A comparison with nineteen other countries. (On line:...
  • J. Cairney et al.

    Socioeconomic position, lifestyle and health among Canadians aged 18 to 64: A multi-condition approach

    Canadian Journal of Public Health

    (1998)
  • G. Davey Smith et al.

    Confounding occupation and smoking: Its magnitude and consequences

    Social Science and Medicine

    (1991)
  • Working hard, living lean: A qualitative study of working low-income families in Edmonton

    (1991)
  • Statistical report on the health of Canadians

    (1999)
  • J.S. Feinstein

    The relationship between socioeconomic status and health: A review of the literature

    The Milbank Quarterly

    (1993)
  • K.F. Ferraro

    Self-ratings of helath among the old and old-old

    Journal of Health and Social Behaviour

    (1980)
  • The policy on health and well-being

    (1992)
  • of Newfoundland Government et al.

    Strategic social plan

    (1998)
  • J.S. House et al.

    Age, socioeconomic status and health

    The Milbank Quarterly

    (1990)
  • Cited by (0)

    This manuscript is an expanded version of a poster presentation at the Annual Conference of the Canadian Public Health Association, Montreal, Quebec, June 1998

    View full text