Table 2.
Adjusted analyses for physical activity, change in activity level additional breast cancer events, and death from all causes in a cohort of women with a history of breast cancer (n = 2,361)
| Physical activity (MET-h/week) | N | n | Additional breast cancer events | n | All-cause mortality | p (trend) | |||
|---|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | p | p (trend) | HR (95% CI) | p | |||||
| Baseline activity level | |||||||||
| Total PA (continuous) | 2,361 | 295 | 0.99 (0.99, 1.00) | .21 | 163 | 0.98 (0.96, 0.99) | .003 | ||
| Total PA (categorical) | |||||||||
| Quintile 1 [0–2.5] | 481 | 69 | Reference | .58 | 43 | Reference | .08 | ||
| Quintile 2 (2.5–7.5] | 485 | 65 | 0.91 (0.64, 1.28) | .58 | 43 | 1.01 (0.66, 1.55) | .97 | ||
| Quintile 3 (7.5–14.9] | 451 | 53 | 0.85 (0.59, 1.22) | .38 | 32 | 0.85 (0.53, 1.35) | .48 | ||
| Quintile 4 (14.9–24.7] | 472 | 60 | 0.97 (0.68, 1.39) | .87 | 27 | 0.75 (0.46, 1.23) | .26 | ||
| Quintile 5 (24.7–107] | 472 | 48 | 0.74 (0.50, 1.10) | .13 | 18 | 0.47 (0.26, 0.84) | .01 | ||
| Mod-vig PA (continuous) | 2,361 | 295 | 0.99 (0.98, 1.00) | .13 | 163 | 0.98 (0.96, 0.99) | .002 | ||
| Mod-vig PA (categorical) | |||||||||
| Quintile 1 [0–1.3] | 474 | 70 | Reference | .30 | 45 | Reference | .02 | ||
| Quintile 2 (1.3–6.3] | 473 | 64 | 0.97 (0.68, 1.36) | .84 | 38 | 1.00 (0.64, 1.55) | .99 | ||
| Quintile 3 (6.3–12.5] | 474 | 64 | 0.98 (0.69, 1.38) | .89 | 41 | 0.99 (0.64, 1.54) | .96 | ||
| Quintile 4 (12.5–22.9] | 470 | 53 | 0.91 (0.62, 1.31) | .60 | 24 | 0.71 (0.42, 1.19) | .19 | ||
| Quintile 5 (22.9–107] | 470 | 44 | 0.67 (0.45, 1.00) | .04 | 15 | 0.39 (0.21, 0.72) | .003 | ||
| Meeting PA guidelinea | |||||||||
| No | 1,186 | 161 | Reference | 102 | Reference | ||||
| Yes | 1,175 | 134 | 0.89 (0.70, 1.14) | .36 | 61 | 0.65 (0.47, 0.91) | .01 | ||
| Baseline-to-1-year change | |||||||||
| Meeting PA guideline | |||||||||
| No → No | 899 | 112 | Reference | 75 | Reference | ||||
| No → Yes | 287 | 49 | 1.44 (1.02, 2.03) | .04 | 27 | 1.21 (0.77, 1.90) | .42 | ||
| Yes → No | 213 | 31 | 1.22 (0.81, 1.83) | .34 | 17 | 1.04 (0.61, 1.77) | .89 | ||
| Yes → Yes | 962 | 103 | 0.93 (0.70, 1.24) | .62 | 44 | 0.60 (0.40, 0.88) | .01 | ||
| Δ Total PA (continuous) | 2,361 | 295 | 1.01 (1.00, 1.02) | .26 | 163 | 1.00 (0.98, 1.01) | .59 | ||
| Δ Total PA (categorical) | |||||||||
| Quintile 1 [−68.8, −5.5] | 474 | 52 | Reference | .60 | 23 | Reference | .70 | ||
| Quintile 2 (−5.5, −0.3] | 473 | 58 | 0.95 (0.64, 1.42) | .81 | 39 | 1.24 (0.71, 2.16) | .45 | ||
| Quintile 3 (−0.3, 2.3] | 474 | 65 | 1.24 (0.82, 1.87) | .31 | 40 | 1.23 (0.69, 2.20) | .49 | ||
| Quintile 4 (2.3, 7.5] | 470 | 57 | 1.10 (0.73, 1.65) | .66 | 35 | 1.17 (0.66, 2.09) | .60 | ||
| Quintile 5 (7.5, 92.3] | 470 | 63 | 1.19 (0.80, 1.77) | .39 | 26 | 0.89 (0.49, 1.64) | .71 | ||
| Δ Mod-vig PA (continuous) | 2,361 | 295 | 1.01 (1.00, 1.02) | .23 | 163 | 0.99 (0.98, 1.01) | .48 | ||
| Δ Mod-vig PA (categorical)b | |||||||||
| Quintile 1 [−63.8, −5.3] | 476 | 54 | Reference | .58 | 24 | Reference | .84 | ||
| Quintile 2 (−5.3, 0] | 805 | 99 | 0.95 (0.65, 1.40) | .76 | 69 | 1.17 (1.69, 1.97) | .56 | ||
| Quintile 3 (0, 1.3] | 147 | 18 | 1.00 (0.57, 1.75) | .99 | 9 | 0.90 (0.40, 2.02) | .81 | ||
| Quintile 4 (1.3, 7) | 563 | 55 | 0.99 (0.66, 1.49) | .97 | 32 | 1.06 (0.60, 1.89) | .84 | ||
| Quintile 5 [7, 92.3] | 470 | 69 | 1.23 (0.83, 1.80) | .30 | 29 | 0.92 (0.51, 1.66) | .78 | ||
All analyses adjusted for age at randomization, race, fruit and vegetable consumption, BMI at randomization, menopausal status, tumor type, tumor grade, tumor stage, anti-estrogen use, clinical site, time from diagnosis to randomization, hot flashes, and study group
aTo approximate guideline of 150 min/wk, we used a cutoff of 10 MET-h/week of moderate-to-vigorous PA
bQuintiles are uneven due to a large number of participants with change scores of exactly 0