Table 1

Regression-discontinuity (RD) results estimating smoking outcomes occurring immediately after the minimum age of tobacco sales in Canada, 2000–2014

Smoking outcomes	Absolute effect†			P value	Relative jump‡
Smoking outcomes	Effect (%)	Lower CI (%)	Upper CI (%)	P value	Jump (%)	Lower CI (%)	Upper CI (%)
Current-smokers: prevalence	2.70**	0.70	4.80	0.009	17.00	3.10	30.90
Daily smokers: prevalence	2.43**	0.74	4.12	0.005	24.59	5.47	43.70
Daily smokers: number of cigarettes smoked per day	−0.196	−1.187	0.796	0.70	−2	−10	7
Occasional smokers: prevalence	0.56	−0.56	1.68	0.32	9.58	−10.63	29.80
Occasional smokers: number of cigarettes smoked on smoking day	−0.087	−0.623	0.448	0.75	−3	−23	17
Occasional smokers: number of smoking days, past month	0.113	−1.439	1.665	0.89	1	−19	21
Average number of cigarettes smoked past month, per person (smokers and non-smokers)	8.45*	1.22	15.67	0.02	18	3	39

*P value<0.05.
**P value<0.01.
†The estimated absolute effect in the smoking outcome appearing immediately after the MATS. Positive values reflect increases.
‡The relative effect (‘jump’) in the smoking outcome appearing immediately after the MATS, calculated by dividing the RD model absolute effect by the RD model intercept (the estimated point where the pre-MATS regression line meets the MATS). Positive values reflect increases.
MATS, minimum age for tobacco sales; RD, regression-discontinuity.