Table 3

The effects of fire safe cigarette laws by smoker prevalence (in quartiles)

Quartile 1Quartile 2Quartile 3Quartile 4
Smoker prevalence (1999–2011)
 Mean (%)16.919.721.324.1
 Range (%)12.4–18.919.0–20.520.6–21.821.9–27.5
OutcomePanel A: state and year effects only
 All-cause fire mortality0.95 (0.90 to 1.01)0.95 (0.86 to 1.06)0.97 (0.84 to 1.12)0.93 (0.80 to 1.08)
 All-cause fire mortality (residential)0.98 (0.89 to 1.08)1.04 (0.80 to 1.34)0.98 (0.69 to 1.40)0.91 (0.77 to 1.08)
 Cigarette-related fires1.02 (0.71 to 1.47)1.13 (0.98 to 1.30)1.07 (0.70 to 1.62)0.86 (0.71 to 1.05)
 Cigarette-related fire mortality0.68 (0.48 to 0.96)0.82 (0.52 to 1.27)0.50 (0.22 to 1.19)0.55 (0.30 to 1.01)
Panel B: state effects, year effects and state-specific trends
 All-cause fire mortality0.95 (0.90 to 1.01)0.95 (0.86 to 1.06)0.97 (0.84 to 1.12)0.90 (0.79 to 1.03)
 All-cause fire mortality (residential)0.97 (0.87 to 1.07)1.01 (0.78 to 1.30)0.96 (0.70 to 1.31)0.92 (0.78 to 1.08)
 Cigarette-related fires1.07 (0.76 to 1.52)1.08 (0.98 to 1.19)1.13 (0.82 to 1.54)0.95 (0.73 to 1.22)
 Cigarette-related fire mortality0.67 (0.45 to 1.00)0.95 (0.60 to 1.50)0.51 (0.15 to 1.68)0.60 (0.31 to 1.15)
  • Poisson regression models with clustered-robust SEs were used to estimate the effects while accounting for correlated errors within states. The dependent variable is offset by population size (per state-year) in all models. The coefficients are presented as IRR with 95% CIs in parentheses.