Health Education Research Advance Access originally published online on July 31, 2006
Health Education Research 2006 21(5):731-739; doi:10.1093/her/cyl015
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Lay concepts of the relative importance of different influences on health; are there major socio-demographic variations?
MRC Social and Public Health Sciences Unit, 4 Lilybank Gardens, Glasgow G12 8RZ, UK
*Correspondence to: S. Macintyre. E-mail: sally{at}msoc.mrc.gla.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionDesign and methodsResultsDiscussionAcknowledgementsReferences |
|---|
There is an extensive literature within anthropology, sociology and psychology about lay concepts of determinants of health and illness. Many of these studies have used single sex or social class samples, often in narrow age bands, and many are qualitative in approach. We asked respondents in a health survey to say how important (on a five-point scale) they thought seven potential influences on health (habits, self-care, the environment, family relationships, one's constitution, money and luck) were. The first three were regarded as very important, the second three as less important and luck as least important. Responses were consistent with current public health and epidemiological knowledge; these respondents endorsed prevailing views about personal responsibility for health and about the role of the physical and social environment in influencing health. In mutually adjusted models, there were no significant gender differences, social class differences and neighbourhood differences in three out of seven influences, and age differences in four out of seven influences. Thus, socio-demographic differences were less marked than might be inferred from studies of specific social groups, indicating a need for caution in health education and health promotion practice against always assuming socio-demographic differences.
|
Introduction |
|---|
|
TopAbstractIntroductionDesign and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Understanding how lay people understand the determinants of health and illness is important for health professionals dealing with patients or their families, and for public health and health promotion efforts to prevent disease and promote recovery [1]. If members of the public do not share professionals' beliefs about key influences of health (for example, if they do not believe that smoking is health damaging [2]), they are less likely to be persuaded to adopt professional guidelines for strategies to manage health or prevent illness. As Macfarlane and Kelleher note: ‘Health professionals need to comprehend the ways in which people critically assess, modify, reject or accept formal messages about health’ [3, p. 1397].
It is thus not surprising that there is an extensive literature within anthropology, sociology and psychology about lay concepts, and causal theories about determinants of health and illness [2–12]. Some studies examine casual attributions for specific illnesses or conditions such as heart disease [13], lung cancer [14] or diabetes [15], while others examine ideas about health and illness in general [16]. Overall, such studies provide evidence of the complexity, sophistication and interconnectedness of ideas about health and illness [12, 16].
A number of studies suggest that beliefs about health and illness may be shaped by historical and local contexts [3, 8], whether respondents are thinking about health or behaviour in general or about their own [9, 13], and personal experience and observation [2, 17].
Many studies use either wholly qualitative methods (e.g. [2, 17, 18]) or partly quantitative and partly qualitative methods (e.g. the coding up of open-ended answers to questions on causes of health and illness in a large scale population survey [9], or Q-sort methodology [16]). Some studies have used single social class or single sex (usually female) samples, often in relatively narrow age bands (e.g. working-class grandmothers [8], 30- to 35-year old women with skilled manual backgrounds [19], 24 working-class respondents in Bethnal Green [2], older people in Ireland [3], 6- to 14-year old children's beliefs about health and illness in Brazil [20]). Studies with such specific and focused samples may be unable to address questions about socio-demographic differences, within a particular culture, in concepts of health and illness, because they cannot make systematic comparisons between men and women, different social classes, different age groups, etc.
The few quantitative or quasi-quantitative studies using a larger spread of participants have tended to suggest that there may be differences by social class, gender or age. Blaxter [9], for example, noted that in responses to open-ended survey questions about causes of disease in general, those in their 40s and 50s more frequently mentioned psychosocial stress, while the physical environment tended to be mentioned more by those with higher income and education and by younger people. Furnham [12] found that younger, unmarried people tended to stress the role of fate, and that more educated people tended to emphasize emotional and behavioural factors, in determining health. However, neither of these studies undertook multivariate analysis to examine the relative importance of these socio-demographic factors, or the importance of any of them while controlling for the others, in predicting responses to questions about health determinants. This is important because those born in earlier periods may, for example, be more likely to have been in lower social classes or to have received less education (because of historical shifts in the labour market and educational systems), and to be more likely to be women (because of women's greater longevity), than younger counterparts; and therefore, one might erroneously attribute differences to age which are actually to do with social class or gender. Similarly, in studies of specific or socially contrasting localities it is not always clear whether individual (or household) social class or local contextual features are being conceptualized as influencing illness attributions [2, 21]. Given the growing literature which finds that area of residence has an independent effect on health after taking individual socio-demographic characteristics into account [22], it is of interest to examine the patterning of health beliefs by area of residence.
In this paper, we explore lay perceptions of the importance of certain factors highlighted in previous research as being prominent in lay views about determinants of health in general (e.g. luck, environment, lifestyle, etc), and systematically examine whether these perceptions vary by age, gender, social class and neighbourhood.
|
Design and methods |
|---|
|
TopAbstractIntroductionDesign and methodsResultsDiscussionAcknowledgementsReferences |
|---|
This analysis uses data from the locality component of the longitudinal The West of Scotland Twenty-07 Study: ‘Health in the Community’. The objective of the study is to explore the social processes which produce or maintain differences in health by key social positions, in particular by gender, age, social class, family composition and area of residence. The ‘Twenty-07 Study’ began in 1987 and will follow three cohorts (born in 1932, 1952 and 1972), using home-based interviews and postal questionnaires, until 2007. The individuals in the cohorts were aged 15, 35 and 55 years at first contact in 1987–88. Since we were targeting these specific age cohorts, we used an enhanced electoral register, containing age, as a sampling frame.
Respondents were asked a wide range of questions on paid and unpaid work, housing, income, family composition, stress, life events, leisure, health-promoting and health-damaging behaviours, beliefs and values and many other material, cultural and psychological factors, along with measures of physical and mental health and well-being. (For further details of the design and methods of the study see [23].)
The locality component of the study involved selecting two areas of Glasgow City with contrasting socio-residential characteristics—North West (NW) and South West (SW)—for relatively intensive study. The NW is relatively advantaged with better health indices, while the SW is relatively disadvantaged with worse health indices. Neither of the localities is at the extreme of health or social advantage in Glasgow. (For further details of the locality component of this study see [24].) Data from the respondents in the locality sample have been gathered on four occasions so far, with face to face interviews in 1987, 1992, 2000–03, and a postal survey in 1997. We have already reported a number of differences in health, health behaviours and health beliefs between these localities, having controlled for social class, age and sex [25–28]. Each sweep of data collection was approved by the University of Glasgow's Ethical Committee for non-clinical research involving human subjects.
This analysis uses data from the 1997 localities postal survey, when individuals from the cohorts were 25, 45 and 65 years old. The overall response to this postal survey was 816 (68.4% of the eligible sample; 74.8% in the NW and 61.3% in the SW) and the achieved sample size still living in the localities (which is what we use in this analysis) was 658 (159 aged 25 years, 224 aged 45 years, 275 aged 65 years). As is usual with longitudinal studies, there was some selective attrition by gender and social class; in 1997 57% respondents were female, compared with 53% in 1987; and 20% were in social classes IV and V compared with 23% in 1987. There was some item non-response so the effective sample size for some analyses is lower. Only one person per family (i.e. no spouses, siblings or children) was sampled. A total of 99.15% respondents were white British. We grouped occupational social class into three categories: the first group comprises Classes I, II and III non-manual (white-collar workers such as bank managers, accountants, social workers and secretaries); the second group consists of Class III manual (skilled blue-collar workers such as plumbers, carpenters, telephone engineers) and the third group is combined from Classes IV and V manual (semi-skilled or unskilled blue-collar workers such as care assistants, janitors, cleaners), using the registrar general's classification of occupations [29].
In the postal survey, to obtain information on perceptions of factors affecting people's health, respondents were asked: ‘How important do you think the following influences are on people’s health .... Luck, The constitution you're born with, Environment (housing, climate, pollution), Habits (smoking, drinking or what you eat), Looking after yourself (taking vitamins, keeping warm, getting enough sleep), How much money you have, and Family relationships'. These factors were based on findings from previous research, in particular that undertaken by Blaxter [9] (who identified individual behaviour, the external environment, standard of living, family and social relationships and heredity/personal/familial susceptibilities as common themes in thinking about the causes of health/illness) and by Furnham [12] and Calnan [30] (who have examined fatalism). We split individual behaviour into ‘more modern’ behavioural aspects (habits; involving diet, smoking and exercise) on the one hand, and more ‘old fashioned’ aspects (involving ‘looking after oneself’) elicited in Blaxter's study of grandmothers in Aberdeen [8, 18]. Respondents were presented with a choice of five response categories—‘very important’, ‘quite important’, ‘neither important nor unimportant’, ‘not very important’ or ‘not at all important’.
Response categories were coded from 0 (not at all important) to 4 (very important). Total mean scores were compared between males and females, social class groups, cohorts and localities, and analysis of variance statistics were calculated. In addition, general linear modelling (GLM) was used to compare means while adjusting for all variables. We tested for two-way interactions between variables, and included only significant interactions. Missing values were filtered out of the dataset before undertaking the analysis, leaving 592 respondents. A conventional level of significance of P < 0.05 is used.
Prior to the GLM analysis, we undertook logistic regression on dichotomized dependent variables (‘very important’ versus ‘quite/neither nor/not very/not all important’) to determine the probability of the different sex/class/age/locality groups perceiving the factors as having a very important influence on health. However, we felt that creating a continuous score and carrying out GLM, rather than using dichotomous categorical variables and regression, more fully captured the complete distribution of responses so this is what we present here. It should be noted that results did not differ between the logistic regression (results not shown) and GLM.
|
Results |
|---|
|
TopAbstractIntroductionDesign and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Table I shows the distribution of ratings of the importance of these potential influences on health. Habits were perceived by 75.3% of respondents as very important, while 64.5% believed self-care, 61.2% the environment and 53.9% family to be very important. Just over 50% (51.6%) perceived constitution to be quite important and 60.5% perceived money to be quite important. Just over a quarter (28.1%) of respondents regarded luck as quite important, while another third regarded it as neither important nor unimportant.
|
Figure 1 displays the mean scores (and 95% confidence intervals) for each potential influence on health. Habits were deemed the most important, with self-care, the environment and family relationships also scoring highly and having tight confidence intervals. Constitution and money showed lower scores, while luck was regarded as the least important.
|
Table II shows unadjusted and mutually adjusted mean scores by the socio-demographic factors. The magnitude differed little between the adjusted and unadjusted models, and no socio-demographic associations gained or lost statistical significance after adjustment.
|
Social class and age were associated with thinking habits important (lower social classes and older people thinking habits ‘less’ important). For self-care, there was a marginally non-significant association with gender, women thinking this ‘more’ important than men. There were no significant socio-demographic differences for the environment. Lower social classes, older people and those in the poorer locality were significantly more likely to think family relationships were important. Older people and those in the more affluent locality thought constitution more important. Lower social classes and older people attributed significantly more importance to money, and people in the more affluent areas attributed more importance to luck.
|
Discussion |
|---|
|
TopAbstractIntroductionDesign and methodsResultsDiscussionAcknowledgementsReferences |
|---|
These adults in the West of Scotland responded to these questions in ways consistent with current public health and epidemiological knowledge [31, 32], attributing most importance to habits (‘smoking, drinking or what you eat’), self-care (‘taking vitamins, keeping warm, getting enough sleep’), the environment (‘housing, climate, pollution’) and family relationships. ‘Money’ and ‘the constitution you're born with’ were attributed with less importance, and luck with least importance. This suggests that these respondents endorsed, at least in answer to a questionnaire about health matters, prevailing views about personal responsibility for health and about the role of the physical (‘housing’ etc) and social (‘family relationships’) environment in influencing health. It is impossible however to tell whether respondents replied in this way because they knew this is what they ‘meant’ to say, or whether they actually believed in the importance of these factors.
Luck, both in the general sense and in terms of ‘the constitution you're born with’, was not stated to be particularly important. ‘Fatalistic’ attributions were thus relatively uncommon. Interestingly, given earlier literature suggesting that poorer health and health behaviours among lower class groups might be because of fatalistic attitudes [2, 7, 33], residents in the SW, the more socially disadvantaged locality, were significantly less likely to regard luck as important than residents in the NW.
In the light of so many studies of lay concepts focusing on women, it is interesting that in our study there were no significant sex differences. This suggests the need for caution in interpreting single sex studies from which implicit comparisons might be drawn with the opposite sex on the assumption that there are always gender differences.
Similarly, there were rather few socio-economic differences, whether measured by occupational social class or area of residence, and observed either before or after adjustment for other socio-demographic factors. For both measures, the magnitude of the differences was not great, and only three of the seven suggested influences showed a significant difference.
Differences between the cohorts suggest generational shifts in understandings of health, with younger cohorts giving answers more consistent with current public health thinking; the 65 year olds were, for example, less likely to think habits were important, and more likely to think family relationships, constitution and money were important. This group had high levels of morbidity and their responses may have been shaped by direct observation and experience of health and illness, while those of younger respondents may have been influenced more by contemporary health promotion messages.
This study is subject to several limitations. First, it is based on a relatively small sample size, of three specific age groups, at a particular time and place (Glasgow, in 1997). The extent to which our findings are generalizable to other age groups, times and places is an empirical question the answer to which can only be determined by studies undertaken on other samples.
Second, this analysis of responses to a fixed choice question in a postal survey does not allow us to consider lay views of the relative importance of different influences on health in any depth. However, it does allow us to compare both the relative importance attributed to these potential influences on health and the importance attributed to them by groups differing in sex, social class, age cohort and area of residence.
We have presented a systematic attempt to compare the importance attributed to different types of influences on health, and to compare responses given by different socio-demographic groups, controlling for other social variables. The sample size for this analysis is not large. Nevertheless, we have been able to demonstrate differences in the importance attributed to different factors, with habits, self-care and environment being seen as most important and luck as least important.
The emphasis in many previous studies on particular age, sex and socio-demographic groups may inadvertently have led to the assumption that there are marked and substantial differences between social groups in concepts of health and its determinants. Our findings, particularly in relation to sex and socio-economic status, suggest the need for caution in health education and health promotion practice against always assuming socio-demographic differences.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionDesign and methodsResultsDiscussionAcknowledgementsReferences |
|---|
All three authors, and the West of Scotland twenty-07 study, are funded by the UK Medical Research Council. We are grateful to all the participants in the study and to survey staff who assisted with the postal survey. We are also grateful to Geoff Der for statistical advice.
|
References |
|---|
|
TopAbstractIntroductionDesign and methodsResultsDiscussionAcknowledgementsReferences |
|---|
1. Fitzpatrick R. Lay concepts of illness. In Fitzpatrick R, Hinton J, Stanton N (Eds.). The Experience of Illness.London: Tavistock 1984 pp. 11–31.
2. Cornwell J. Hard-Earned Lives: Accounts of Health and Illness from East London.London: Tavistock 1984.
3. Macfarlane A and Kelleher D. Concepts of illness causation and attitudes to health care among older people in the Republic of Ireland. Soc Sci Med 2002 54:1389–400.[CrossRef][ISI][Medline]
4. Kleinman A. What kind of model for an anthropology of medical systems. Am Anthropol 1978 80:661–74.[CrossRef]
5. Kleinman A. Patients and Healing in the Context of Culture: An Exploration of the Borderline between Anthropology, Medicine and Psychiatry.Berkeley, CA: University of California Press 1980.
6. Helman CG. Feed a cold and starve a fever. Folk models of infection in an English suburban community, and their relation to medical treatment. Cult Med Psychiatry 1978 2:107–37.[CrossRef][Medline]
7. Pill R and Stott N. Relationship between health locus of control and belief in the relevance of lifestyle to health. Patient Couns Health Educ 1981 3:95–9.[ISI][Medline]
8. Blaxter M. The causes of disease: women talking. Soc Sci Med 1983 17:56–69.
9. Blaxter M. Health and Lifestyles.London: Tavistock/Routledge 1990.
10. Blaxter M. Whose fault is it? People's own conceptions of the reasons for health inequalities. Soc Sci Med 1997 44:747–56.[CrossRef][ISI][Medline]
11. Pitts M and Phillips K. The Psychology of Health.London: Routledge 1991.
12. Furnham A. Explaining health and illness: lay beliefs on the nature of health. Pers Individ Dif 1994 17:455–66.[CrossRef]
13. French DP, Senior V, Weinman J, et al. Causal attributions for heart disease: a systematic review. Psychol Health 2001 16:77–98.
14. Linn MW, Linn BS, Stein SR. Beliefs about the causes of cancer in cancer patients. Soc Sci Med 1982 16:835–9.[CrossRef][ISI][Medline]
15. Sissons-Joshi M. Lay explanations of the causes of diabetes in India and the UK. In Markova I and Farr M (Eds.). Representations of Health, Illness and Handicap.London UK: Harwood Academic 1995.
16. Stainton Rogers W. Explaining Health and Illness: An Exploration of Diversity.Hemel Hempstead, UK: Wheatsheaf 1991.
17. Davison C, Smith GD, Frankel S. Lay epidemiology and the prevention paradox. Sociol Health Illn 1991 13:1–19.[Medline]
18. Blaxter M and Paterson E. Mothers and Daughters; a Three Generational Study of Health Attitudes and Behaviour.London: Heinemann 1982.
19. Pill R and Stott N. Concepts of illness causation and responsibility: some preliminary data from a sample of working class mothers. Soc Sci Med 1982 16:43–52.[CrossRef][ISI][Medline]
20. Boruchovitch E and Mednick BR. Causal attributions in Brazilian children's reasoning about health and illness. Rev Saúde Pública (J Public Health) 2000 34:484–90.
21. Popay J, Bennett S, Thomas C, et al. Beyond ‘beer, fags, egg and chips’? Exploring lay understandings of social inequalities in health. Sociol Health Illn 2003 25:1–23.[CrossRef][ISI][Medline]
22. Pickett K and Pearl M. Multilevel analyses of neighbourhood socio-economic context and health outcomes: a critical review. J Epidemiol Community Health 2001 55:111–22.
23. Macintyre S, Annandale E, Ecob R, et al. The West of Scotland twenty-07 study; health in the community. In Martin C and McQueen D (Eds.). Readings for a New Public Health.Edinburgh: Edinburgh University Press 1989 pp. 56–74.
24. Macintyre S, Maciver S, Sooman A. Area, class and health; should we be focusing on places or people? J Soc Policy 1993 22:213–34.[ISI]
25. Ellaway A and Macintyre S. Does where you live predict health related behaviours? A case study in Glasgow. Health Bull 1996 54:443–6.
26. Ellaway A, Anderson A, Macintyre S. Does area of residence affect body size and shape? Int J Obes 1997 21:304–8.[CrossRef][ISI][Medline]
27. Macintyre S, McKay L, Ellaway A. Are rich people or poor people more likely to be ill? Lay perceptions, by social class and neighbourhood, of inequalities in health. Soc Sci Med 2004 60:313–7.[CrossRef]
28. Macintyre S, McKay L, Ellaway A. Who is more likely to experience common disorders: men, women, or both equally? Lay perceptions in the West of Scotland. Int J Epidemiol 2005 34:461–6.
29. Office of Population Censuses and Surveys. Standard Occupational Classification.London: HMSO 1991.
30. Calnan M. Health and Illness: The Lay Perspective.London: Tavistock 1987.
31. Department of Health. Saving Lives: Our Healthier Nation.London: The Stationery Office 1999.
32. Department of Health. Choosing Health; Making Healthy Choices Easier.London: The Stationery Office 2004.
33. Pill R and Stott N. Preventive procedures and practices among working class women: new data and fresh insights. Soc Sci Med 1985 21:975–83.[CrossRef][ISI][Medline]
Received on April 19, 2005; accepted on March 30, 2006
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M Stafford, J Nazroo, J M Popay, and M Whitehead Tackling inequalities in health: evaluating the New Deal for Communities initiative J. Epidemiol. Community Health, April 1, 2008; 62(4): 298 - 304. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||