Search

This paper examines the links between attitudes towards cigarette smoking and the social environments of communities involved in the U.S. National Cancer Institute's Community Intervention Trial for Smoking Cessation (COMMIT). Our objective is to identify sources of social–geographic variation in smoking attitudes and norms which can hinder or enhance public health efforts to reduce tobacco use. The analysis had two stages: (1) place (measured as region and community) was identified as an important main effect accounting for individual variation in smoking attitudes independent of smoking status and personal characteristics; (2) case studies of COMMIT sites in North Carolina, Iowa, Washington, New Jersey and New Mexico were conducted to reveal features of the local milieux which could account for variations in smoking attitudes. Some of the place characteristics that we suggest are linked to local attitudes include economic reliance on the tobacco industry, libertarian political orientations, socio-economic conditions, legislative context and ethnic composition. Given the effects of regional and community attributes on individual attitudes towards smoking, we conclude that public health efforts to control smoking should continue to be targeted beyond individual smokers to the broader social environment.

A parallel case study design was used to investigate psychosocial effects in populations exposed to solid waste facilities. Psychosocial effects were defined as a complex of distress, dysfunction and disability, manifested in a range of psychological, social and behavioural outcomes, as a consequence of actual or perceived environmental contamination. This paper presents the results of logistic regression analyses designed to identify determinants of psychosocial effects of exposure. The data come from an epidemiologic survey of residents (N = 696) living within a prescribed radius from each of three solid waste facilities in southern Ontario. The analytical model has three main components: external variables (e.g. individual and exposure-related variables); mediating variables (e.g. social network membership and involvement, general health status measures); and outcome variables (e.g. concern, effects and actions). Results for a series of site specific analyses show that outcome measures can be successfully explained by a combination of external and mediating factors. In general, variables from each of the three main components enter the concern-related models while the action models are clearly dominated by social network variables. Analyses using data from all three sites indicate the explanatory power of site-related characteristics. However, given the number and diversity of variables in the models, there is no support for a simple cause and effect relationship. The implication is that strategies aimed to address and alleviate psychosocial effects need to be specific to the characteristics of the populations in particular settings.