The impact of housing on health: an ecologic study from the Canadian Arctic.

Objectives and method: An ecologic study was conducted to investigate the association between housing and health in 49 predominantly Native communities in the Northwest Territories (NWT) in the Canadian Arctic, making use of data from a housing survey and data relating to physical and social health routinely reported to various service delivery agencies.

Results: The rate of health centre visits, as a measure of morbidity, correlated with most housing and socioeconomic status (SES) indicators. Communities with worse housing and SES were found to have a higher rate of health centre visits. Using factor analysis, composite housing and SES indicators were constructed. Housing was poorly correlated with any of the outcome variables, whereas SES was strongly correlated with health centre visits and moderately correlated with the occurrence of fires. Communities with worse SES were more likely to have a higher rate of health centre visits but the opposite effect was observed with fires. When both housing and SES indices were included in a multiple regression model, SES was the stronger factor in predicting the rate of health centre visits and fires. No significant association exists between these variables and the frequency of sentenced admissions.

Conclusions: It is possible that improvement in housing quality in the NWT may have eliminated much disparity and perhaps removed most obvious illness-generating housing features. Improved housing through subsidies, however, does not alter other components of SES such as education, employment and income. The importance of SES in health and social well-being is well known and is demonstrated in the NWT as well. This study also shows that ecologic analyses can provide relatively quick and useful data for public health policy.