Table 3.

Factors modifying FEV₁ change (in mL) after work cessation based on a linear mixed effects model, with exposure modelled as cotton vs. silk.

	FEV1, mL [95% CI]
	All		Non-smokers
Cessation-years	14.2***	[12.0, 16.5]	9.4***	[7.1, 11.7]
Cessation-years2	0.3***	[0.1, 0.4]	0.3***	[0.2, 0.5]
Cotton*cessation-years	−2.7*	[−5.0, −0.5]	−1.7a	[−4.2, 0.7]
Current smoker*cessation-years	1.1	[−2.6, 4.9]	-	-
Former smoker*cessation-years	0.4	[−4.4, 5.2]	-	-
Male*cessation-years	−5.7**	[−9.1, −2.3]	−4.5**	[−7.8, −1.3]
Observations	4,702		3,068

^*p<0.05

^**p<0.01

^***p<0.001

Note: Quadratic interaction terms were significant only for the male*cessation-years² term, with β=0.5 [0.2, 0.8], p=0.0009 for all workers and β=0.4 [0.1, 0.7], p=0.02 for the analysis restricted to non-smokers.

^acotton*cessation-years interaction p=0.08 in analysis restricted to non-smokers

Model adjusted for age, height, gender, exposure (cotton vs. silk), smoking (never, current, former), cumulative pack-years, and interaction terms between cessation-years and occupational exposure, cessation-years and smoking, and cessation-years and gender. Loss-to follow-up accounted for using inverse probability of censoring weights. Note: as the modelled outcome was FEV₁ and not change in FEV₁, the interaction terms with cessation-years represent the associations between occupational exposure, smoking, gender, and change in FEV₁ after leaving work.