Table 4 Associations between the components of individual level job strain at baseline and non-smoking at follow-up in light and moderate/heavy baseline smokers: adjusted odds ratios (ORs) (95% CIs) from multilevel logistic regression models (n = 4928) in the Finnish Public Sector Study, 2000–2005

	OR, (95% CI), adjusted for sex, age, marital status, employer, heavy drinking, body mass index, physical activity, trait anxiety and socioeconomic position
	Light smokers (<10 cigarettes/day in baseline) (n = 2102)	Moderate/heavy smokers (> = 10 cigarettes/day in baseline) (n = 2826)
Individual job strain:
1 (high)	1.00	1.00
2	1.01 (0.76 to 1.34)	1.00 (0.74 to 1.35)
3	0.98 (0.74 to 1.31)	0.86 (0.63 to 1.19)
4 (low)	1.28 (0.96 to 1.71)	1.08 (0.78 to 1.49)
p Value for linear trend*	0.033	0.709
Individual job control:
1 (low)	1.00	1.00
2	0.93 (0.69 to 1.25)	0.86 (0.64 to 1.16)
3	0.94 (0.70 to 1.27)	0.81 (0.60 to 1.11)
4 (high)	1.28 (0.94 to 1.73)	0.85 (0.62 to 1.18)
p Value for linear trend*	0.127	0.375
Individual job demands:
1 (high)	1.00	1.00
2	1.10 (0.84 to 1.44)	1.07 (0.78 to 1.46)
3	1.05 (0.80 to 1.37)	1.01 (0.74 to 1.38)
4 (low)	1.16 (0.87 to 1.54)	1.19 (0.87 to 1.64)
p Value for linear trend*	0.114	0.280

*Linear trends with continuous variables.