(P11-044-23) Time-Varying Analysis of Multivitamin Use and Mortality Risk in Two Large, Prospective US Cohorts

Objectives: In the US, 1 out of every 3 adults use multivitamins (MV), and usage is higher among women and older Americans. The primary motivation for MV use is to enhance health and prevent disease; thus, mortality studies can inform public health guidance. In 2022, the US Preventative Services Task Force reviewed data on MV supplementation and all-cause mortality from randomized controlled trials and found insufficient evidence for determining benefits or harms owing, in part, to limited follow-up time and external validity. Observational evidence is mixed potentially owing to competing issues of confounding by healthy lifestyle and reverse causation whereby those diagnosed with chronic disease may start MV use. Our objective was to address limitations of prior MV use and mortality studies.

Methods: We used Cox proportional hazards regression models with time-varying covariates to estimate multivariable-adjusted hazard ratios (HR) and 95% confidence intervals (CI) in two large, prospective, and geographically diverse, US cohorts, the NIH-AARP Diet and Health Study (N=244,138; n=130,149 deaths) and the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial Study (N=50,251; n=20,807 deaths). Both had repeat assessments of MV use, detailed information on lifestyle, and >20 years of follow-up. We modeled risk estimates in the first and second half of follow-up due to a violation of the proportional hazards assumption.

Results: Daily MV users, as compared to nonusers, had fewer mortality risk factors. They were more likely to be never smokers, female, college educated, and normal BMI but were also more likely to report a history of cancer. After adjustment for potential confounders, MV use was not associated with lower mortality risk; summary HR estimates for daily MV users versus nonusers were 1.06 (95% CI=1.03-1.10) and 0.98 (0.93-1.04) during the first and second half of follow-up, respectively. Results were similar in age and sex-stratified analyses. A weak association between daily MV use and higher mortality risk was observed earlier in follow-up after excluding those with poor self-rated health or a history of chronic disease, arguing against reverse causation.

Conclusions: Overall, we found that MV use is not associated with a mortality benefit but may be associated with modest harm.

Funding Sources: NCI Intramural Research Program; ODS Scholars Award