Skip to main content
. 2020 Feb 5;9(4):e014168. doi: 10.1161/JAHA.119.014168

Table 2.

Changes in BMI, Physical Activity, and Alcohol Consumption Between the Survey Waves in Initiators and Prevalent Users of Preventive Medication Versus Noninitiators

Lifestyle Factor/User Group Outcome
Mean (SDa) Mean Change (P Valuea) β Value (95% CI)b
Baseline Second Survey Unadjusted Differencec Adjusted Differenced
BMI, kg/m2
Initiators (n=8425) 26.3 (4.2) 26.8 (4.5) 0.51 (<0.001) 0.13 (0.08–0.17) 0.19 (0.16–0.22)
Prevalent users (n=25 348) 27.0 (4.6) 27.3 (4.7) 0.31 (<0.001) −0.04 (−0.07 to 0.01) 0.08 (0.06–0.10)
Noninitiators (n=44 091) 24.9 (3.7) 25.2 (3.9) 0.37 (<0.001) Reference Reference
Physical activity (MET h/d)
Initiators (n=8658) 4.15 (3.77) 3.97 (3.60) −0.18 (<0.001) −0.13 (−0.22 to −0.04) −0.09 (−0.16 to −0.02)
Prevalent users (n=26 368) 3.78 (3.44) 3.69 (3.38) −0.09 (<0.001) −0.12 (−0.17 to −0.06) −0.08 (−0.13 to −0.04)
Noninitiators (n=45 312) 4.71 (4.00) 4.70 (4.03) −0.01 (0.47) Reference Reference
Alcohol consumption, g/wk
Initiators (n=8721) 69.0 (108) 66.4 (110) −2.64 (0.011) −2.57 (−4.75 to −0.38) −1.85 (−3.67 to −0.14)
Prevalent users (n=26 546) 66.7 (109) 63.9 (111) −2.38 (<0.001) −1.73 (−3.13 to 0.34) −0.72 (−1.92 to 0.49)
Noninitiators (n=45 488) 61.1 (93.2) 60.6 (95.6) −0.44 (0.26) Reference Reference
No. of cigarettes/daye
Initiators (n=876) 13.1 (7.05) 12.5 (6.83) −0.60 (<0.001) −0.37 (−0.63 to −0.11) −0.34 (−0.60 to −0.08)
Prevalent users (n=2251) 12.3 (6.65) 12.1 (6.71) −0.23 (0.016) −0.02 (−0.20 to 0.16) 0.05 (−0.13 to 0.24)
Noninitiators (n=4532) 11.9 (6.60) 11.7 (6.45) −0.19 (0.002) Reference Reference

Only participant‐observations who had data available on all baseline covariates were included in adjusted models (n=74 291 or 95.4% of those with data on body mass index at both surveys, n=75 613 or 94.1% of those with data on physical activity at both surveys, n=75 781 or 93.8% of those with data on alcohol consumption at both surveys, n=7339 or 95.8% of those reporting number of cigarettes smoked at both surveys). BMI indicates body mass index; MET, metabolic equivalent of task.

a

SDs and P values reported here do not take into account that each individual may have contributed to >1 data set.

b

Derived from generalized estimating equations.

c

Adjusted for baseline year and time interval between the survey waves.

d

Adjusted for baseline year; time interval; age; sex; marital, occupational, and employment status; diabetes mellitus; smoking status; and other lifestyle factors.

e

Among those who were current smokers at both waves.