Skip to main content
. 2022 Jan 15;14:11. doi: 10.1186/s13098-022-00784-2

Table 4.

Multiplicative and additive interactions between CSI and demographic and lifestyle characteristics

Stratified factors Discovery stage: TNCHS study a Replication stage: CHNS study b
MetS ( +)/All CSI categories Ptrendc Pinteraction d RERI
(95%Cl) e
MetS ( +)/All CSI categories P trend c P interaction d RERI
(95%Cl) e
Tertile 1 Tertile 2 Tertile 3 Tertile 1 Tertile 2 Tertile 3
OR (95%Cl) OR (95%Cl) OR (95%Cl) OR (95%Cl) OR (95%Cl) OR (95%Cl)
Sex 0.245

0.31

(− 0.19 to 0.81)

0.312

− 0.07

(− 0.69 to 0.54)

 Female 1932/8149 1.19 (0.73–1.96) 1.78 (0.98–3.23) 1.33 (0.67–2.66) 0.050 755/4589 1.03 (0.53–2.01) 1.42 (0.67–3.03) 0.93 (0.45–1.93) 0.738
 Male 1614/7073 1.22 (0.95–1.56) 1.82 (1.46–2.28) 2.15 (1.72–2.69)  < 0.001 804/3976 1.05 (0.80–1.36) 1.26 (1.00–1.60) 1.17 (0.91–1.50) 0.085
Age 0.015

− 0.35

(− 0.86 to 0.16)

0.001

− 1.29

(− 2.13 to − 0.45)

 < 70 years 1988/9544 1.25 (0.96–1.63) 1.87 (1.47–2.37) 2.31 (1.81–2.95)  < 0.001 1313/7591 0.97 (0.76–1.24) 1.38 (1.09–1.76) 1.33 (1.03–1.71) 0.004
 ≥ 70 years 1558/5678 1.04 (0.71–1.52) 1.46 (0.99–2.15) 1.41 (0.95–2.08) 0.020 246/974 0.71 (0.29–1.75) 0.61 (0.27–1.37) 1.07 (0.62–1.84) 0.813
Exercise status 0.542

− 0.10

(− 0.29 to 0.49)

0.930

0.004

(− 0.32 to 0.31)

 Exercise/like 2334/9267 1.04 (0.80–1.35) 1.81 (1.42–2.30) 1.97 (1.54–2.53)  < 0.001 462/2608 1.06 (0.72–1.57) 1.28 (0.87–1.90) 0.73 (0.47–1.15) 0.701
 No exercise /dislike 1212/5955 1.68 (1.13–2.50) 1.66 (1.12–2.47) 2.08 (1.43–3.01)  < 0.001 1097/5957 1.44 (1.06–1.95) 1.31 (0.99–1.73) 1.11 (0.84–1.45) 0.180
Alcohol consumption 0.054

− 0.38

(− 0.83 to 0.07)

0.597

0.16

(− 0.15 to 0.46)

 Never consumed alcohol 3204/14,057 1.23 (0.92–1.65) 1.89 (1.43–2.50) 2.48 (1.86–3.30)  < 0.001 1038/5799 1.15 (0.78–1.69) 1.35 (0.95–1.91) 0.88 (0.62–1.24) 0.839
 Past/current consumer 342/1165 1.02 (0.69–1.52) 1.37 (0.94–1.99) 1.44 (1.00–2.08) 0.038 521/2766 1.23 (0.89–1.70) 1.36 (1.00–1.86) 1.31 (0.94–1.82) 0.047

CSI comprehensive smoking index, OR: odds ratio, CI confidence interval, BMI body mass index, RERI relative excess risk due to interaction

aAdjusted for age, physical activity, BMI, dietary habits

bAdjusted for sex, age, BMI, province, region

cLinear trend tests were performed by entering the median value of each group of cigarette smoking variables as continuous variables in the models

dP values for the multiplicative interactions between CSI and stratified factors

eIn the calculation of RERI, smoking status was modeled as a dichotomous variable (never vs. ever)