Table 3

Adjusted* prevalence ratios and 95% CIs for the association between tobacco retailer counts and proximity in the residential neighbourhood and activity space,† and smoking status among 1994 participants from the ISIS study (Montreal, Canada, 2011–2012)

Residential neighbourhoodActivity space
PR95% CIPR95% CI
Tobacco retailer counts
LowRefRef
Medium1.180.93 to 1.501.221.01 to 1.47
High1.531.23 to 1.911.461.26 to 1.70
 QICu§20792082
p for trend0.000.00
Tobacco retailer proximity
LowRefRef
Medium1.180.95 to 1.471.321.03 to 1.68
High1.150.95 to 1.391.421.09 to 1.86
 QICu20942087
 p for trend0.170.01
  • Bold results are statistically significant at the 0.05 level.

  • *All models are adjusted for individual-level variables age, sex, occupational status and educational attainment.

  • †For 85 participants (4.3%), the activity space corresponded to the residential neighbourhood because they had answered as having not conducted any of the inquired activities (n=53) or they only conducted activities at home (n=32).

  • ‡Categories for residential counts are low: 0–1 retailer; medium: 2–5 retailers; high: 6 retailers or more. For mean activity space counts, categories are: low: 0–4 retailers; medium: 4–7.9 retailers; high: 8 retailers or more.

  • §Goodness-of-fit statistic.

  • ¶Categories for residential and mean activity space proximity are low: more than 350 m; medium: 150–350 m; high: less than 150 m.

  • ISIS, Interdisciplinary Study on Inequalities in Smoking; PR, prevalence ratio; QICu, Quasi-likelihood under the independent model criterion.