Table 2.

Association between smoking behavior among youth and adults and measures of outdoor tobacco advert density and proximity in Semarang, Indonesia.

	Youth Sample				Adult Sample
	Density		Proximity		Density		Proximity
	OR (SE)		OR (SE)		OR (SE)		OR (SE)
	[1]		[2]		[3]		[4]
(a) Overall	n = 400				n = 492
Density tertiles
Low	Ref				Ref
Medium	1.93 **	(0.52)			1.25	(0.31)
High	2.16 **	(0.59)			1.01	(0.24)
Proximity
At least one within 200 m			0.97	(0.22)			0.95	(0.20)
(b) Junior high schools	n = 260
Density tertiles
Low	Ref					NA
Medium	1.76	(0.54)
High	1.93 **	(0.64)
Proximity
At least one within 200 m			0.68	(0.18)
(c) Senior high schools	n = 140
Density tertiles
Low	Ref					NA
Medium	2.83	(1.58)
High	2.78 **	(1.38)
Proximity
At least one within 200 m			2.80 **	(1.23)
(d) Poorer areas	n = 180				n = 223
Density tertiles
Low	Ref				Ref
Medium	2.64 **	(0.89)			1.66	(0.55)
High	5.16 **	(3.00)			0.96	(0.46)
Proximity
At least one within 200 m			2.03 **	(0.64)			1.70	(0.53)
(e) Richer areas	n = 220				n = 269
Density tertiles
Low	Ref				Ref
Medium	0.99	(0.60)			0.79	(0.30)
High	1.01	(0.37)			0.94	(0.29)
Proximity
At least one within 200 m			0.19 **	(0.08)			0.55	(0.18)

Note: OR = Odds Ratio, SE = Standard Errors, n = Sample, Ref = Reference group. Odds ratios were obtained from logit regressions of smoking status on density/proximity, controlling for age (in STATA 15.1). There were 400 students interviewed from 20 high schools; 492 adults interviewed near those schools. The measures of density and proximity were the same for youth and adults. Density was measured by the total number of adverts within 400 m of each school. For neighborhood, poorer/richer areas are with higher/lower subdistrict-level poverty rates. The density was 1.43 times higher at schools in richer (mean = 11.27, SD = 8.29) areas than those in poorer areas (mean = 7.89, SD = 7.14). ** = significant at 5% level.