It’s a familiar sight in any pharmacy or grocery store: a wall of multivitamins in countless brands and formulations, marketed for a variety of health benefits. Approximately one third of all American adults report taking multivitamins, making them the most commonly used dietary supplements in the US.1 Their popularity presumably stems from a widespread belief that they promote general health and help to prevent chronic disease, which in turn may extend lifespan. Yet a recent study has called into question these alleged benefits of multivitamins.2 The study results, which were picked up by various mainstream news outlets, have left many consumers wondering if they are wasting their money – or worse, if their use of multivitamins might even be slightly harmful for health. But can we really put any stock into these recent findings?
About the study
Researchers Loftfield et al. analyzed data from three US-based prospective cohort studies – the National Institutes of Health–AARP Diet and Health Study (NIH-AARP) cohort; the PLCO Cancer Screening Trial cohort; and the Agricultural Health Study (AHS) cohort – in order to investigate the relationship between long-term multivitamin (MV) use and mortality. The investigators specifically selected participants from these cohorts who were in relatively good health at baseline (e.g., free of any history of cancer, diabetes, or cardiovascular events), and after applying these and other exclusion criteria, they were left with a total participant pool of 390,124 adults: 327,732 from the NIH-AARP cohort; 42,732 from the PLCO cohort; and 19,660 from the AHS cohort.
MV use and demographic data were assessed through questionnaires at baseline and various follow-up points, with Loftfield et al. stratifying MV use across three categories: non-users, non-daily users, or daily users of MV. Participants were followed until their death (as determined from the National Death Index), loss to follow-up, or the study’s end. Follow-up time was divided into two phases, with follow-up period 1 (FP1) representing the first 12 years and follow-up period 2 (FP2) representing the last 15 years of follow-up.
In total, there were 7,861,485 person-years of follow-up, over which a total of 164,762 participants died. Over this time, daily MV users were found to have a 4% higher mortality risk than non-users (FP1: HR=1.04; 95% CI: 1.02-1.07; FP2: HR=1.04; 95% CI: 0.99-1.08) after adjusting for sex, age, race & ethnicity, education, marital status, BMI, cigarette smoking, daily alcohol intake, daily coffee intake, healthy eating index quartile (HEI-2015), family history of cancer, and individual supplement use.
In addition to all-cause mortality, the authors also analyzed associations between MV use and cause-specific mortality from cancer (49,836 deaths), heart disease (35,060 deaths), and cerebrovascular diseases (9,275 deaths). Daily MV use was reported to be associated with a 6% increase in risk of deaths from heart disease, though this was exclusively observed in FP1 (HR=1.06; 95% CI: 1.01-1.11). No other significant differences in risk between daily or non-daily MV use and non-use were observed for deaths from any specific causes included in the analysis.
Based on all of these findings, the authors concluded that use of multivitamins confers no benefit to lifespan and instead may increase mortality risk relative to non-use.
A questionable result with a large potential for bias
Before you start throwing your multivitamins in the trash, let’s first take a closer look at these results. The authors report a 4% higher mortality risk with daily MV use relative to non-use at both the first and second follow-up points, but this was only statistically significant at FP1, and as we’ve seen, no sustained, significant differences were observed in mortality from any specific causes. Further, this miniscule difference was overshadowed by a larger difference between non-daily MV users versus non-users (HR=1.09; 95% CI: 1.05-1.13) at the first follow-up, which disappeared completely by FP2 (HR=0.95; 95% CI: 0.88-1.02). In all, this reveals that the so-called “increase in risk” with MV use reflects an inconsistent association with no apparent dose-dependency – raising the likelihood that this result was due to random chance or bias, rather than to a true positive association.
But even if we determine that the 4% increase in risk is unreliable, can we still conclude based on these data that multivitamins offer no benefit?
Unfortunately not. One of the great challenges with research on dietary supplements is the potential for either “healthy user bias” (in which individuals who are more health-conscious and are engaging in several healthy behaviors may be more likely to take MVs) or “sick user bias” (in which those with an existing disease may be more likely to take MVs due to perceived health benefits). Loftfield et al. acknowledged this concern but claimed to have addressed it effectively by collecting data on – and adjusting for – variables such as BMI, smoking status, alcohol intake, and diet. However, observational studies such as this can never account for every variable related to health. For instance, body composition and use of other medications were not assessed. Thus, the risk of substantial influence from healthy or sick user bias cannot entirely be excluded.
The authors also chose to focus on participants who were relatively healthy at baseline, excluding those with a history of cancer, diabetes, heart disease or stroke, or late-stage kidney disease. This, too, might have been intended to minimize the effect of sick user bias, but it leaves us with a cohort of unusually healthy older adults, and the effect of this selection may differ between groups. Because the median age of participants in this study was 61.5 years, participants had potentially already gone through over half a lifetime of either taking or not taking MVs, which could have caused health separation between groups before baseline in such a way that would make one group more susceptible to this “healthy participant” selection bias than the other. For instance, if MVs do have a positive effect on general health and disease prevention, then those who have never taken MVs would be more likely to develop diabetes/cancer/etc. by age 60 than those who take MVs every day. This study’s exclusion criteria would then mean that, of all non-users, only those who didn’t develop these diseases by the study start – i.e., the very healthiest among this group – would be included in the analysis, whereas the sample of MV users included in the study would be more representative of all MV users.
Were some benefits missed?
An additional consequence of this study’s exclusion criteria is that we miss any opportunity to determine how the use of MVs might be associated with mortality in the context of poor baseline health or of specific disease states. Some evidence indicates that certain supplements may have more potent or qualitatively unique benefits when used in the context of specific health concerns as compared to regular use in a state of wellness. (For example, omega-3 fatty acids have been shown to improve recovery of neurological function after stroke.3) Likewise, it’s also possible that MVs would provide substantial benefits only when an individual is deficient in certain vitamins and minerals, a common occurrence which often goes unnoticed.
Additionally, risk for some causes of death may be more impacted by supplement use than others. Cause-specific mortality in this study was only evaluated for deaths due to heart disease, cancer, or cerebrovascular disease, but other common causes of mortality arguably have much stronger links to micronutrition (e.g., calcium and vitamin D impact bone health, which in turn impacts risk of falls and frailty-related accidents, a major cause of mortality in older adults). Had separate analyses been conducted specifically in individuals with diseases thought to be improved through various supplements, perhaps MV use would have been found to reduce overall mortality risk or mortality due to specific causes.
Bottom line
The authors of this study claim that their results show that the use of multivitamins offers no benefit for reducing mortality risk, and in fact, that daily use of multivitamins may confer a 4% higher mortality risk than non-use. However, questionable data and various biases make these results unreliable, and ultimately, this study tells us nothing about how multivitamin use may or may not be associated with mortality in a general population.
Even if MVs don’t offer any mortality benefit (which we can’t conclude from this study for all of its flaws), this still doesn’t mean they serve no purpose for human health. There’s no question that supplements improve various aspects of health in the context of vitamin or mineral deficiencies, for instance, and compelling evidence links certain vitamins and minerals to reduced risk of particular health concerns such as osteoporosis even in otherwise healthy individuals, which may in turn impact mortality from specific causes.
Are multivitamins the be-all, end-all key to longevity? Certainly not. But whether or not they move the needle in this regard remains an open question – and this present study did little to add to the conversation.
For a list of all previous weekly emails, click here.
References
- Mishra S, Stierman B, Gahche JJ, Potischman N. Data Brief 399: Dietary Supplement Use among Adults: United States, 2017–2018. National Center for Health Statistics; 2021. doi:10.15620/cdc:101131
- Loftfield E, O’Connell CP, Abnet CC, et al. Multivitamin Use and Mortality Risk in 3 Prospective US Cohorts. JAMA Netw Open. 2024;7(6):e2418729. doi:10.1001/jamanetworkopen.2024.18729
- Zhang W, Wang H, Zhang H, et al. Dietary supplementation with omega-3 polyunsaturated fatty acids robustly promotes neurovascular restorative dynamics and improves neurological functions after stroke. Exp Neurol. 2015;272:170-180. doi:10.1016/j.expneurol.2015.03.005
