Declines in income-related mortality disparities in Canada: A world first for tobacco control?

Introduction. Large, growing socioeconomic mortality disparities appear to exist everywhere except Canada where such mortality disparities fell from 1986 to 1996. The duration and possible causes of Canada's recent reductions in mortality disparities are not well quantified. So StatCan mortality rates by urban area income quintile in 1986, 1991, 1996, and 2001 were located.

Methods. Relationships between mortality and cumulative tobacco smoke damage (smoke load), income, and year were assessed by stepwise linear regression using lung cancer rates as a smoke load proxy. Regressions were run with and without likely misclassified 1986 deaths.

Results. Smoke load, as quantified by lung cancer rates, was the dominant predictor of all other mortality rates (slope 6.86 (95% confidence interval 5.98-7.74 for 1986-2001) in males and all other excluding breast cancer in males and females. After adjustment for smoke load, income quintile in both periods and year in 1991-2001 were not statistically significantly related to total mortality excluding lung cancer in males. Smoke load variation alone could explain over 95% of the variance in total mortality across the year-income-gender strata in 2001.

Discussion. Canada reduced male socioeconomic smoke load disparities and their associated all cause mortality disparities from 1986-2001 and female disparities in 2001. That may be a valuable and uniquely early accomplishment of tobacco control in Canada.