Skip to main content
. Author manuscript; available in PMC: 2020 May 1.
Published in final edited form as: Cancer Prev Res (Phila). 2019 Mar 25;12(5):315–326. doi: 10.1158/1940-6207.CAPR-18-0452

Table 8.

Association between physical activity at 10-year follow-up and risk of subsequent colorectal cancer in the Multiethnic Cohort, 2003–2013

Hours/day Men (n=34,089) Women (n = 41,921) P for
heterogeneityd
Cases HR (95% CI)a Casesb HR (95% CI)c Cases HR (95% CI)a Casesb HR (95% CI)c
Recreational activity
 <0.21 158 1.00 (ref) 135 1.00 (ref) 171 1.00 (ref) 154 1.00 (ref)
 0.21‒<0.46 71 0.61 (0.46–0.80) 57 0.59 (0.43–0.81) 96 0.99 (0.77–1.28) 84 1.05 (0.80–1.37)
 0.46‒<0.82 98 0.75 (0.58–0.97) 86 0.79 (0.60–1.05) 80 0.95 (0.73–1.24) 73 1.09 (0.82–1.45)
 0.82‒<1.32 63 0.71 (0.53–0.96) 58 0.81 (0.59–1.12) 40 0.87 (0.62–1.24) 35 0.97 (0.66–1.42)
 ≥1.32 63 0.62 (0.46–0.84) 55 0.67 (0.48–0.92) 34 0.95 (0.65–1.37) 29 1.06 (0.70–1.60)
 P for trend 0.01 0.09 0.55 0.88 0.38
Work-related activity
 <0.21 134 1.00 (ref) 119 1.00 (ref) 79 1.00 (ref) 71 1.00 (ref)
 0.21‒<0.46 119 0.85 (0.67–1.09) 102 0.83 (0.64–1.09) 144 1.16 (0.88–1.52) 129 1.18 (0.88–1.57)
 0.46‒<0.82 89 0.75 (0.57–0.98) 74 0.71 (0.53–0.95) 95 1.11 (0.83–1.50) 83 1.13 (0.82–1.55)
 0.82‒<1.43 55 0.71 (0.52–0.97) 47 0.69 (0.49–0.97) 50 0.98 (0.69–1.40) 45 1.04 (0.71–1.51)
 ≥1.43 56 0.74 (0.54–1.02) 49 0.71 (0.50–1.00) 53 1.11 (0.78–1.58) 47 1.16 (0.80–1.69)
 P for trend 0.07 0.06 0.99 0.78 0.18
Moderate/vigorous activity
(MET-hours)
 <1.86 115 1.00 (ref) 100 1.00 (ref) 96 1.00 (ref) 86 1.00 (ref)
 1.86‒<3.34 74 0.67 (0.50–0.90) 63 0.66 (0.48–0.90) 105 0.94 (0.71–1.24) 95 1.00 (0.74–1.34)
 3.34‒<5.71 76 0.65 (0.49–0.88) 62 0.61 (0.44–0.85) 75 0.92 (0.68–1.25) 64 0.96 (0.69–1.33)
 5.71‒<10.3 98 0.66 (0.51–0.87) 87 0.69 (0.51–0.92) 79 0.90 (0.67–1.22) 72 1.02 (0.74–1.41)
 ≥10.29 90 0.61 (0.46–0.81) 79 0.62 (0.45–0.84) 66 0.94 (0.68–1.30) 58 1.05 (0.74–1.49)
 P for trend 0.01 0.04 0.76 0.69 0.19
a

Adjusted by Cox proportional hazards regression for age at 10-year follow-up and race/ethnicity.

b

Excluding participants with missing information on covariates at 10-year follow-up.

c

Further adjusted for family history of colorectal cancer, history of colorectal polyp, body mass index, pack-years of cigarette smoking, multivitamin use, non-steroidal anti-inflammatory drug use, menopausal hormone therapy use for women only, and intake of alcohol, total energy, red meat, dietary fiber, calcium, folate, and vitamin D.

d

Tests for heterogeneity between men and women were performed based on the multivariate-adjusted models.