Table 4:

Association of baseline smoking status with longitudinal trajectory of muscle attenuation in analysis stratified by age. The first part shows the characteristics of each subgroup, while the second part shows regression coefficients of fully adjusted linear mix-effects models for each subgroup.

	Male, Age (years) Group		Female, Age (years) Group
	Aged 55–64	Aged 65–74	Aged 55–64	Aged 65–74
Cohort characteristics
No. participants	8,168	3,122	5,834	1,895
BMI (kg/m2), SD	28.4 (4.6)	27.9 (4.3)	27.7 (5.6)	27.3 (5.3)
Smoking status, %
Current	3,946 (48.3)	2,435 (44.9)	3,038 (52.1)	800 (42.2)
Former	4,222 (51.7)	2,987 (55.1)	2,796 (47.9)	1,095 (57.8)
Disease history, %
CVD	3,277 (40.1)	1,620 (51.9)	1,963 (33.7)	875 (46.2)
Diabetes	844 (10.3)	390 (12.5)	395 (6.8)	155 (8.2)
COPD	1,153 (14.1)	591 (18.9)	1,186 (20.3)	454 (24.0)
Muscle Attenuation
Year-0 (HU), SD	31.9 (5.5)	29.2 (5.5)	27.3 (5.6)	25.0 (5.4)
Change rate (HU/y), SD	−0.62 (1.45)	−0.63 (1.45)	−0.60 (1.43)	−0.57 (1.41)
Association of smoking status (current vs. former) with muscle attenuation trajectory
Main effect
Coefficient, SE	−0.903 (0.105)	−0.831 (0.180)	−0.603 (0.126)	−0.939 (0.227)
P Value	<.001	<.001	<.001	<.001
Change-rate
Coefficient, SE	−0.120 (0.033)	0.043 (0.055)	−0.039 (0.041)	−0.063 (0.070)
P Value	<.001	=.43	=.33	=.36
† Association of smoking quit time (y) with muscle attenuation trajectory
Main effect
Coefficient, SE	0.042 (0.015)	0.054 (0.022)	0.058 (0.019)	−0.019 (0.031)
P Value	=0.005	=.02	=.003	=.54
Change-rate
Coefficient, SE	−0.002 (0.005)	0.012 (0.007)	0.003 (0.006)	−0.006 (0.009)
P Value	=.62	=.09	=.68	=.54

^†Analyses with smoking quit time as primary predictor were performed among those who had quit smoking at baseline.

CVD=cardiovascular disease. COPD=chronic obstructive pulmonary diseases