Role of Baseline Nutritional Status on the Effect of a Multivitamin on Cancer in the Physicians' Health Study II

Abstract

The Physicians' Health Study (PHS) II is the only long‐term randomized trial of 14,641 men aged 蠅50 y testing a daily multivitamin, reporting a significant 8% reduction in total cancer. An important consideration is the understudied, modifying role of baseline nutritional status. We examined whether baseline nutritional status modified the effect of randomized multivitamin use on total cancer during a median follow‐up of 11.2 y in PHS II.13,316 men completed a food frequency questionnaire, from which we categorized selected nutrient, fruit, vegetable, whole grain, and dairy intake into tertiles and estimated hazard ratios (HRs) and 95% CIs of total cancer comparing active vs placebo multivitamin in intention‐to‐treat analyses. We also tested the interaction between the randomized multivitamin and tertiles of each dietary factor, and the trend in HRs across tertiles of each factor. Significant interactions were found for vitamin B12, total carotenoid, ω‐3 fatty acid, fruit, and vegetable intake with multivitamin treatment on cancer. Potentially stronger cancer reductions were found in men taking a multivitamin and consuming lower amounts of vitamin B12, total carotenoids, and vegetables, in contrast to greater fruit intake. We also considered quartiles and quintiles of each dietary factor, and other cancer endpoints including prostate, non‐prostate, and colorectal cancer for which results were inconsistent and power more limited. In conclusion, selected dietary characteristics may influence whether a multivitamin reduced cancer. Additional studies will identify more specific dietary factors and patterns more or less likely to reduce cancer with long‐term multivitamin use in men. Men (Total Cancer Events) Group Active Placebo Hazard Ratioa (95% CI) P, interaction Vitamin B12, μg/d 0.019 <4.76 μg/d 2,110 (309) 2,108 (361) 0.83 (0.71‐0.96) 4.76 ‐ <7.94 μg/d 2,107 (407) 2,112 (392) 1.06 (0.92‐1.22) ≥7.94 μg/d 2,120 (395) 2,098 (415) 0.93 (0.81‐1.07) P, trend in HRs = 0.78 Folate, μg/d 0.61 <427.5 μg/d 2,138 (379) 2,080 (388) 0.93 (0.81‐1.07) 427.5 ‐ <651.0 μg/d 2,102 (383) 2,117 (401) 0.97 (0.84‐1.12) ≥651.0 μg/d 2,097 (349) 2,121 (379) 0.92 (0.80‐1.07) P, trend in HRs = 0.86 Vitamin D, IU/d 0.82 <126.1 IU/d 2,102 (351) 2,116 (372) 0.94 (0.82‐1.09) 126.1 – <217.4 IU/d 2,085 (366) 2,134 (390) 0.95 (0.82‐1.09) ≥217.4 IU/d 2,150 (394) 2,068 (406) 0.92 (0.80‐1.06) P, trend in HRs = 0.83 Total carotenoids, IU/d 0.008 <4430 IU/d 2,134 (347) 2,084 (382) 0.88 (0.76‐1.02) 4430 ‐ <7147 IU/d 2,113 (396) 2,106 (359) 1.10 (0.95‐1.27) ≥7147 IU/d 2,090 (368) 2,128 (427) 0.86 (0.75‐0.99) P, trend in HRs = 0.48 Omega‐3 fatty acids, g/d 0.01 <0.70 g/d 2,150 (360) 2,068 (357) 0.95 (0.82‐1.10) 0.70 ‐ <0.89 g/d 2,067 (364) 2,152 (453) 0.83 (0.72‐0.95) ≥0.89 g/d 2,120 (387) 2,098 (358) 1.08 (0.94‐1.25) P, trend in HRs = 0.16 Fruits, servings/d 0.033 <1.43 servings/d 2,223 (397) 2,209 (377) 1.05 (0.91‐1.21) 1.43 ‐ <2.57 servings/d 2,196 (409) 2,246 (442) 0.93 (0.82‐1.07) ≥2.57 servings/d 2,239 (357) 2,197 (407) 0.85 (0.73‐0.97) P, trend in HRs = 0.036 Vegetables, servings/d 0.033 <1.59 servings/d 2,196 (343) 2,240 (399) 0.86 (0.75‐0.995) 1.59 ‐ <2.61 servings/d 2,239 (430) 2,199 (398) 1.05 (0.92‐1.21) ≥2.61 servings/d 2,223 (390) 2,212 (428) 0.90 (0.79‐1.04) P, trend in HRs = 0.85 Whole grains, g/d 0.16 <15.3 g/d 2,140 (351) 2,078 (381) 0.89 (0.77‐1.03) 15.3 ‐ <32.7 g/d 2,094 (385) 2,125 (371) 1.02 (0.89‐1.18) ≥32.7 g/d 2,103 (375) 2,115 (416) 0.91 (0.79‐1.05) P, trend in HRs = 0.99