Table 3. Unadjusted and Multivariable-Adjusted Associations of Every 1-Minute Reduction in Treadmill Exercise Duration Between Baseline and Year 7 Follow-up.
| Event | Association per Minute of Exercise Duration Reduction From Baseline to Year 7 | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Model 1a | Model 2b | Model 3c | |||||||
| No. of Participants | Effect Size, (95% CI) | P Value | No. of Participants | Effect Size (95% CI) | P Value | No. of Participants | Effect Size (95% CI) | P Value | |
| All-cause mortality, HRd | 2472 | 1.19 (1.06 to 1.33) | .002 | 2448 | 1.22 (1.08 to 1.38) | .002 | 2472 | 1.21 (1.07 to 1.37) | .002 |
| Incident CVD, HRd | 2472 | 1.24 (1.10 to 1.40) | <.001 | 2448 | 1.18 (1.02 to 1.37) | .03 | 2472 | 1.20 (1.06 to 1.37) | .006 |
| Presence of CAC, ORe | |||||||||
| Year 15 | 1803 | 0.98 (0.90 to 1.08) | .69 | 1792 | 0.98 (0.86 to 1.10) | .67 | NA | NA | NA |
| Year 20 | 1794 | 0.99 (0.93 to 1.06) | .82 | 1782 | 0.94 (0.86 to 1.03) | .19 | NA | NA | NA |
| Year 25 | 1789 | 1.01 (0.95 to 1.07) | .87 | 1779 | 0.97 (0.90 to 1.05) | .47 | NA | NA | NA |
| β Value for CAC at year 25 (logarithm-transformed, where CAC >0)f | 480 | −0.03 (−0.11 to 0.06) | .56 | 476 | −0.06 (−0.16 to 0.05) | .28 | NA | NA | NA |
| LV hypertrophy, ORe,g | 1796 | 1.18 (1.07 to 1.29) | <.001 | 1783 | 1.01 (0.90 to 1.14) | .83 | NA | NA | NA |
| β Value for LVmass index (year 25)f,h | 1796 | 0.97 (0.68 to 1.25) | <.001 | 1783 | 0.11 (−0.19 to 0.41) | .49 | NA | NA | NA |
| β Value for global longitudinal strain (year 25)f | 1690 | 0.20 (0.13 to 0.27) | <.001 | 1678 | 0.15 (0.08 to 0.23) | <.001 | NA | NA | NA |
Abbreviations: CAC, coronary artery calcification; CVD, cardiovascular disease; HR, hazard ratio; LV, left ventricular; NA, not applicable; OR, odds ratio.
Adjusted for baseline treadmill exercise duration.
Adjusted for baseline treadmill exercise duration, age, race, sex, parental history of myocardial infarction (MI), body mass index (BMI), systolic blood pressure, diabetes mellitus, smoking exposure, high-density lipoprotein (HDL) and total cholesterol levels, and 7-year changes in BMI, systolic blood pressure, smoking exposure, HDL and total cholesterol levels, and incident diabetes mellitus at year 7.
Adjusted for baseline treadmill exercise duration, age, race, sex, parental history of MI, and time-dependent BMI, systolic blood pressure, diabetes mellitus, smoking exposure, and HDL and total cholesterol levels.
Modeling framework was Cox proportional hazards regression.
Modeling framework was logistic regression.
Modeling framework was general linear models.
Defined as year 25 LV mass index >2 SDs above the race-sex–specific year 5 echocardiographic LV mass, indexed by height raised to the 2.7th power.
Calculated as LV mass divided by height raised to the 2.7th power.