By Dónal P. O’Mathúna, PhD
Senior Lecturer in Ethics, Decision-Making & Evidence, School of Nursing and Human Sciences, Dublin City University, Ireland
Dr. O’Mathúna reports no financial relationship to this field of study.
Synopsis: An association was found between decreased death from cardiovascular disease and U.S. women’s use of multivitamin-mineral supplements when taken for at least 3 years. A similar association was not found for men, nor for either gender after use of multivitamins without minerals.
Source: Bailey RL, et al. Multivitamin-mineral use is associated with reduced risk of cardiovascular disease mortality among women in the United States. J Nutr 7 Jan 2015; ePub ahead of print. doi: 10.3945/jn.114.204743.
- The mortality from cardiovascular disease (CVD) was significantly associated with taking multivitamin-mineral (MVM) supplements for 3 years or longer for healthy U.S. women.
- A similar statistically significant association was not found for men.
- No significant associations were found for use of multivitamins without minerals and CVD mortality.
- Several statistical tests were performed to check the results’ reliability, and these confirmed the study’s conclusions. However, the findings need to be interpreted with caution because of the nature of the study design involved.
The overall general finding from this analysis of a representative sample of U.S. adults was that no significant associations were found between the use of multivitamin-mineral supplements (MVMs) or multivitamins (MVs) and cardiovascular disease (CVD) mortality when all supplement users were compared with nonusers. However, when users were grouped according to how long they had been using supplements, some significant associations were found. Those who used MVMs for more than 3 years had a significant reduction in hazard ratio (HR) compared to non-users (HR, 0.65; 95% confidence interval [CI], 0.49-0.85). When further subdivided by gender, the association between women using MVMs for more than 3 years and non-users was strengthened (HR, 0.56; 95% CI, 0.37-0.85). No significant association was found for men (HR, 0.79; 95% CI, 0.44-1.42). Significant associations were not found for MVs and CVD mortality.
In this study, data from the National Health and Nutrition Examination Survey III (NHANES III) were matched with mortality data from the National Health Index (NHI) through the end of 2011 to examine potential associations between MVM or MV use and CVD mortality. NHANES III allows the selection of a nationally representative sample of the U.S. population. Participants were enrolled between 1988 and 1994 and were interviewed at baseline about their supplement use. During the initial interview, participants showed researchers the containers and labels of all supplements they were taking. This allowed the collection of more detailed information on supplement contents and doses than is often available in studies. Based on length of use reported, participants were assigned to one of three groups: those using supplements for < 1 year, for 1-3 years, or ≥ 3 years. For this study, an MVM was defined as a product containing three or more vitamins plus one or more minerals. An MV was defined as any vitamin combination that did not include minerals.
The current analysis was based on adults older than 40 years of age in 2011 who had no history of CVD, myocardial infarction, stroke, coronary heart disease, or chronic kidney disease at baseline. Pregnant and lactating women and anyone with missing dietary supplement data were excluded. The final study sample was 8678 participants. The median length of time between NHANES III baseline and NHI report was 18.7 years. Among the population analyzed, 45.3% had used any form of dietary supplement in the previous 30 days. The most frequently used type of supplement was an MVM (21.2%), followed by an MV (14.2%). Up to the end of 2011, 4122 NHANES III participants had died, with CVD being the most common cause of death (1636 participants or 40% of deaths).
MVs are the most frequently used dietary supplement in the United States, with Americans spending an estimated $11.8 billion on them each year.1 The Bailey et al study provides important evidence about the effectiveness of MVM supplements for the prevention of death from CVD. At the same time, its findings raise additional questions about MVMs and CVD and how best to study their effectiveness. The U.S. Preventive Services Task Force concluded in 2013 that there was no evidence of an effect of nutritional doses of vitamins or minerals on CVD, cancer, or mortality in healthy individuals.1 However, this systematic review identified only two randomized, controlled trials (RCTs) upon which to base its recommendations. One of these RCTs enrolled only male U.S. physicians and found no association between MVM use and CVD incidence or mortality after 11 years. The other RCT enrolled men and women, and found no statistically significant effect on CVD incidence after 7.5 years, but used a supplement containing antioxidants and was not an MVM. On the other hand, a prospective cohort study in Sweden found that MV use by women for more than 5 years was associated with a reduced risk of myocardial infarction.2 An RCT of MVM use by women and CVD mortality is not available.
Untangling the results of research on MVs from MVM products is difficult. Part of the problem behind the inconsistent findings has to do with the study designs. The RCT is regarded as the gold standard for determining whether an intervention is effective. But as the study becomes more controlled, the setting becomes less like everyday life and the subjects become less heterogeneous. RCTs are more narrowly focused, which can make it more difficult to apply their findings to the general population. On the other hand, observational studies like this one include a more representative spread of people from all walks of life. They allow people to continue living their usual lifestyle, but then it becomes more difficult to separate the impact of the intervention from the many confounding factors. To take account of this, Bailey et al ran several statistical tests and found similar findings each time.
The researchers did highlight some additional important limitations with their study. The participants’ use of supplements was self-reported and only collected at baseline in NHANES III. Over the course of the next 18 or so years, participants could have changed their use of supplements in any number of ways. This is a major limitation with the findings, especially since the intervening years are the most important for any assessment of the impact of the supplements on CVD. In addition, NHANES III was conducted before the passage of the Dietary Supplement and Health Education Act of 1994. Since then, dietary supplement use has increased in general in the United States.3 In addition, those who had been taking supplements for 3 years prior to NHANES III may have been “early adopters” who were proactive about their health in other ways.
Another reason for variation in the findings of different studies arises from the different supplements used for different lengths of time. Although NHANES III identified the products each participant used, they varied widely. In addition, other studies have administered individual vitamins and minerals in varying doses, or MVs. Although MVMs are the most widely used supplements in the United States, few RCTs have used MVM supplements. Results will inevitably remain variable given that the interventions are variable. Why MVMs have benefits not found with MVs is unclear. Minerals (like magnesium or copper) may have their own beneficial effects on CVD, although many of these individual effects are themselves uncertain. The reasons for gender differences are also unclear. The limitations of observational studies must be kept in mind, and also the challenges of sub-group analyses.4 As more sub-groups are analyzed within study results, the more likely it is to find spurious correlations and significant findings. Sub-group analyses are best used to generate hypotheses that are tested in RCTs.
Overall, Bailey et al found some evidence that MVMs may have a protective effect against CVD mortality in healthy U.S. women when used for at least 3 years. At the same time, the study provides little evidence of benefit for men against CVD mortality, and no evidence of benefit from MV products. While this study was well-designed, it has important limitations given the data available for analysis. Healthy women can be encouraged that MVMs may protect against CVD mortality, although a clearer recommendation will have to await the results of further studies, especially the results of RCTs of specific MVM supplements.
- Fortmann SP, et al. Vitamin and mineral supplements in the primary prevention of cardiovascular disease and cancer: An updated systematic evidence review for the U.S. Preventive Services Task Force. Ann Intern Med 2013;159:824-834.
- Rautiainen S, et al. Multivitamin use and the risk of myocardial infarction: A population-based cohort of Swedish women. Am J Clin Nutr 2010;92:1251-1256.
- Gahche J, et al. Dietary supplement use among U.S. adults has increased since NHANES III (1988–1994). NCHS Data Brief 2011;61:1-8.
- Horton R. From star signs to trial guidelines. Lancet 2000;355:1033-1034.