Table 4.

HRs and 95% CIs for invasive breast cancer associated with number of WCRF/AICR recommendations met at baseline

VITAL cohort (N = 30,797)Breast cancer cases (N = 899)Age-adjustedCovariate-adjusteda
Recommendations metN (%)N (%)HR (95% CI)HR (95% CI)
None571 (1.9)28 (3.1)1.001.00
15,025 (16.3)166 (18.5)0.67 (0.45–1.00)0.68 (0.46–1.02)
210,324 (33.5)300 (33.4)0.57 (0.39–0.84)0.56 (0.38–0.83)
38,616 (28.0)239 (26.6)0.54 (0.36–0.80)0.53 (0.35–0.78)
44,405 (14.3)125 (13.9)0.55 (0.36–0.83)0.52 (0.34–0.79)
5–61,856 (6.0)41 (4.6)0.42 (0.26–0.68)0.40 (0.25–0.65)
Per recommendation0.91 (0.86–0.96)0.89 (0.84–0.95)
Ptrendb0.001<0.001

NOTE: Age is used as the timeline in the Cox proportional hazards models. All models included 30,797 observations and 899 incident breast cancers.

  • aAdjusted for age (as the timeline in the Cox model), education, race, mammography, family history of breast cancer, age at menarche, age at first birth, age at menopause, and years of estrogen plus progestin hormone therapy use using categories in Table 2 and daily energy intake (kcal).

  • bHRs, 95% CIs, and P values were calculated using a two-sided test for linear trend modeling categories as a continuous variable (0–6).