A great deal of research has explored the extent to which health and the determinants of health are influenced by the characteristics of individuals and/or their environments. (1) There is agreement that both play a part, but investigations have focused on understanding the relative contribution of individual and environmental effects and the most influential elements of the environment. (2)
Similarly, in terms of lifestyle behaviours, many studies have explored the promoting and hindering influences of individual characteristics and environments. (3) Associated with this knowledge is the fact that social and physical environments may vary between neighbourhoods within a single city. Reports indicate that disadvantaged neighbourhoods, for example, are associated with increased barriers to health-promoting behaviours. (4)
This literature and these challenges provide the context of the present SSHRC-funded study, "Deconstructing the Determinants of Health at the Local Level," which investigates neighbourhood-level variations in health and lifestyle behaviours by comparing adults (18 yrs +) from contrasting neighbourhoods in Hamilton, Ontario. We also examine the contribution of individual- and neighbourhood-level characteristics to lack of regular exercise, smoking, overweight Body Mass Index (BMI), reduced wellbeing, and poor self-rated health.
The present research is part of a larger study--unique to Canada--that examines broad determinants of health in Hamilton neighbourhoods (see http://www.mcmaster.ca/mieh/research/deconstructing.html for more about "Deconstructing the Determinants of Health at the Local Level"). Determinants of health have largely been studied at the national and international level. By examining relationships in these local populations, the present project extends our understanding of the determinants of health.
II Research Steps
Using 1996 census data, four unique neighbourhoods were identified:
* Chedoke-Kirkendall: high income, high education, low unemployment, many recent immigrants
* Central Downtown: low income, low education, high unemployment, many recent immigrants
* Northeast Industrial: low income, low education, high unemployment, few recent immigrants
* Southwest Mountain: high income, high education, low unemployment, few recent immigrants. (5)
A telephone survey was conducted with 300 individuals in each neighbourhood plus a comparison group of 300 within the "Old City" (i.e., the city boundary prior to amalgamation in 2000). The survey asked about health status (self-rated, chronic conditions), and wellbeing (GHQ-20 item). (6) Questions were also asked about smoking, exercise frequency, BMI, social networks, community involvement (e.g., volunteer, sport clubs), demographics, perceptions of and satisfaction with the neighbourhood and the state of the home (e.g., disrepair). The response rate was 60% and data collection was completed in April 2002. In addition to survey responses about the local environment, we integrated data based on observations of neighbourhood surroundings (e.g., number and location of playing fields, fast food outlets) utilising some of the techniques described by Handy et al. (7)
The survey instrument is available by request.
~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *
The mean age was 47.9 years (with a standard deviation of 17.1) and 46% of the sample of 1,500 participants was male.
Overall, neighbourhood variations showed that the Mountain and Chedoke neighbourhoods fared better in terms of health status and determinants. In particular, the Downtown and Industrial participants reported fair or poor health, chronic conditions, poor coping, and emotional distress more frequently than the Chedoke and Mountain participants. In addition, the Downtown and Industrial participants reported that they lived alone, had no community involvement, and were dissatisfied with their neighbourhood more frequently than the remaining neighbourhoods. A detailed table of results is available on the project website.
With respect to physical surroundings, the Industrial neighbourhood had the highest combined density of fast food outlets, cigarette outlets, liquor stores, etc., and the Mountain the lowest. A relatively high density of these features was associated with reduced self-rated health, overweight BMI, and smoking.
Taking all the data into account, we isolated the variables associated with each health outcome (the contributing variables are listed in order of importance):
* Fair/poor self-rated health: poor coping with daily demands, no close friends, use of health services, income less than $30,000, illness event in the past 12 months
* Reduced wellbeing (GHQ score greater than or equal to 4, indicating recent emotional distress): poor coping with daily demands, dissatisfied with neighbourhood, money worries, illness event in past 12 months, not born in Canada, financial difficulties in past 12 months, home needs repairs, no community involvement, smoking, use of health services, not in workforce
* Lack of regular exercise: did not complete high school, no community involvement, not born in Canada, financial problems in past 12 months, female, one or more chronic conditions (professional occupations were more likely to exercise)
* Overweight/obese BMI: machine/manufacturing occupation, Industrial neighbourhood, wanting to move, one or more chronic conditions, money worries, older-age (females, smokers, caregivers, and those renting were less likely overweight)
* Smoker: did not complete high school, no community involvement, machine/manufacturing occupation, financial problems in past 12 months, renting (being overweight, home needing repairs, and having pets were associated with not smoking)
~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *
IV Lessons learned
Based on our results, we conclude that reduced health status and unhealthy lifestyle behaviours are associated with disadvantaged neighbourhoods in Hamilton.
This supports our suggestion regarding the larger study--that data collected at the national level does not always provide a picture of what may occur within a city. Local-level variations best identify the barriers to and facilitators of healthy lifestyles and improved health status faced by local populations.
Based on the present analysis, having friends and the ability to cope may be vital for better self-rated health, and neighbourhood and local features were not important. However, trying to link local features to everyday life is difficult. For a neighbourhood analysis, we need to specify boundaries. We realise, however, that people may not necessarily purchase fast food nor exercise solely in their neighbourhood. Despite this, we did construct a reliable and valid indicator of unhealthy features.
We observed that individual characteristics and social and physical environments are of varying importance for the lifestyle and wellbeing measures.
One concern is the "multiple jeopardy" experienced by some people in certain locations. Not only are the residents less affluent, they face increased barriers due to where they live. Although they report neighbourhood dissatisfaction and experience poorer health or lifestyles (after controlling for age, income, and gender), many cannot afford to move.
~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *
V The Way Forward
Ultimately we are looking for ways to promote positive health for all. To help explain the contributions of social, environmental, and individual aspects to healthy behaviours, we have interviewed ten people in each of the four neighbourhoods, focusing on factors that promote or hinder healthy lifestyles. What are understood as the main facilitators and barriers? How much do people want to change? Do they believe they should or need to? This data forms part of a master's thesis and a summary of the findings will be available on the project website in the near future.
We are also currently analysing a follow-up survey in the four neighbourhoods, asking about social contacts, safety, use of parks, and neighbourhood ties. Incorporating the interview and follow-up findings will enrich our explanations. Gaining a better understanding of the relative influences, specifically the relevance of individual aspects, social and physical environments to health and lifestyle is essential in order to enhance health. Some of the follow-up results will be available in June and will be presented at the 2004 CPHA conference.
(1) For example, see Macintyre, S., Ellaway, A., & Cummins, S. (2002). Place effects on health: how can we conceptualise, operationalise and measure them? Social Science and Medicine, 55, 125-139.
(2) For example, see Shaw, M., Dorling, D., & Mitchell, R. (2002). Health, place and society. Prentice Hall: Harlow, England.
(3) For example, see Adrian, M., Ferguson, B., & Her, M. (1996). Does allowing the sale of wine in Quebec grocery stores increase consumption? Journal of Studies on Alcohol, 57, 434-448.
(4) For example, Reidpath, D., Burns, C., Garrard, J., Mahoney, M., & Townsend, M. (2002). An ecological study of the relationship between social and environmental determinants of obesity. Health and Place, 8, 141-145.
(5) For example, see Luginaah, I., Jerrett, M., Elliott, S., Eyles, J., et al. (2001). Health profiles of Hamilton: Spatial characterization of neighbourhoods for health investigations. GeoJournal, 53, 135-147.
(6) With the GHQ-20, participants receive a score based on 20 items that each ask how usual it is to have felt a certain way in the past 2 weeks, for example, "reasonably happy," or "nervous and tense." For the GHQ 20-item version, a score of 4 or more indicates emotional distress.
(7) Handy, S., Boarnet, M., Ewing, R., & Killingsworth, R. (2002). How the built environment affects physical activity: Views from urban planning. American Journal of Preventive Medicine, 23 (2S), 64-73.