An Industry Perspective: Dietary Supplements and ...

Journal of Dietary Supplements, Early Online:1–8, 2013 C 2013 by Informa Healthcare USA, Inc. Available online at www.informahealthcare.com/jds DOI: 10.3109/19390211.2013.783661

COMMENTARY

J Dietary Suppl Downloaded from informahealthcare.com by 38.100.3.98 on 05/16/13 For personal use only.

An Industry Perspective: Dietary Supplements and Mortality Rates in Older Women Taylor C. Wallace, PhD, FACN1 , Douglas MacKay, ND1 , Barry W. Ritz, PhD2 , Michael McBurney, PhD3 , Andrew Shao, PhD4 , John Miller5 , James Brooks, PhD6 , & Lewis Hendricks, MS7 1

Council for Responsible Nutrition, Washington, DC, USA, 2 Atrium Innovations, Chadds Ford, Pennsylvania, USA, 3 DSM Nutritional Products – North America, Parsippany, New Jersey, USA, 4 Herbalife International of America Inc, Torrance, California, USA, 5 GNLD International, Fremont, California, USA, 6 Pharmavite LLC, Northridge, California, USA, 7 Innophos Inc, Cranbury, New Jersey, USA

ABSTRACT. This special article seeks to provide balance and clarity to the confusion brought about by the conclusions resulting from a recent study published in Archives of Internal Medicine by Mursu et al. (2011). An examination of three key limitations of the study provides context to why additional research is needed: (a) Nonusers were poorly defined; (b) supplement users were healthier than nonusers; and (c) the number of supplement users increased throughout the study. Although the literature is limited, other similar observational studies have also shown positive effects on the risk mortality for both multivitamins and single-nutrient supplements. Observational trials are an essential component of evidence-based nutrition but do not offer certainty because other data, such as the one generated from randomized controlled trials, are equally important in regard to the totality of evidence. The Senior Scientific Advisory Committee for the Council for Responsible Nutrition, an industry trade group, feels that the conclusions of the study by Mursu et al. (2011) are overstated and suggests that researchers analyze cohort(s) designed to specifically examine vitamin and mineral supplements free of confounding from factors, such as hormone replacement therapy, to better assess their benefits to the general population. KEYWORDS.

dietary supplements, industry perspective, mortality, women

INTRODUCTION Dietary supplement use in the United States has increased substantially over the past ≥20 years, according to the data from the National Health and Nutrition Address correspondence to: Taylor C. Wallace, PhD, Senior Scientific Advisory Committee, Council for Responsible Nutrition, 1828 L Street NW, Suite 510, Washington, DC 20036–5114, USA (E-mail: [email protected].) (Received 19 June 2012; accepted 4 February 2013)

1


2

Wallace et al.

Examination Survey (NHANES) (Gahche et al., 2011). The NHANES 2003–2006 data show that multivitamins are currently the most popular dietary supplement on the market and are reportedly used by approximately one-third of the US population (Bailey, Gahche, & Lentino, 2011). Beyond their pivotal role in helping consumers achieve recommended nutrient intakes when diet alone is not sufficient (Fulgoni, Keast, Bailey, & Dwyer, 2011), multivitamins and other multiple- and single-nutrient vitamin and/or mineral supplements may be particularly important to certain subpopulations (i.e., folic acid and iron in pregnant women and vitamin B12 in elderly individuals) (US Department of Agriculture, Center for Nutrition Policy and Promotion, 2010). Using the Iowa Women’s Health Study cohort, Mursu, Robien, Harnack, Park, and Jacobs (2011) conducted an observational study that was published in the Archives of Internal Medicine in 2011. Their results showed that vitamin and/or mineral supplements, particularly multivitamins, might increase the risk of mortality in elderly women. The publicity from this study has resulted in some health care practitioners changing their recommendations surrounding vitamin and/or mineral supplement use (MedScape WebMD, 2012). This practice could have long-term negative public health implications, particularly for those individuals in certain subpopulations who are at risk for suboptimal intake and those who are deficient in a particular nutrient or nutrients. As members of the Senior Scientific Advisory Committee for the Council for Responsible Nutrition (CRN), the leading trade group representing dietary supplement manufacturers and ingredient suppliers, we feel that it is necessary to provide balance and clarity to the confusion brought about by the conclusions resulting from the study by Mursu et al. (2011). A SECOND LOOK AT THE STUDY BY MURSU ET AL. (2011): THREE KEY LIMITATIONS Nonusers Were Poorly Defined Several factors in the recent article by Mursu et al. (2011) warrant caution in interpretation of the results and further analysis. Most important is how users and nonusers of dietary supplements were defined and segregated. Table 1 of the original manuscript, which is also included in this commentary, compares the baseline characteristics of the 24,329 supplement users versus the 14,443 nonusers in 1986; however, mortality among these two groups is not discussed. The authors fail to report mortality between total users and the true nonusers (participants who did not use any supplement throughout the analysis). Instead, the authors compared mortality in users of a specific multiple- or single-nutrient supplement at baseline with all other women in the cohort of 38,772 individuals. As a result, a nonuser TABLE 1. Changes in Users and Nonusers of Supplements from 1986 to 2004 Year

Users

Total

Change (%)

1986 2004

24,329 16,278

38,772 19,124

66.9 85.2

Adapted from Mursu et al. (2011).


Industry Perspective: Supplements and Mortality

3

of a multivitamin could have been a user of ≥1 other supplement, such as individual vitamins. In other words, when Mursu et al. (Mursu et al., 2011) compared multivitamin users with nonusers, all study participants who were using any number of single-nutrient supplements were classified as nonusers in the study analysis. Clearly, the authors did not account for women who were using >1 of the specific supplements being investigated (e.g., users of folic acid were compared with users of multivitamins that almost always contain folic acid). Instead, the authors define nonusers of a particular nutrient (e.g., vitamin C) as those individuals who were not using that specific single-nutrient supplement, regardless of their use of other supplements (e.g., multivitamins). Thus, the size of the nonuser groups was artificially increased, likely affecting the mortality analysis of the so-called nonuser cohort. The authors failed to address the mortality rate of a specific multiple- or single-nutrient supplement compared with the true 14,443 nonusers (or the portion that continued to be true nonusers throughout the study) of dietary supplements. Subsequent to the original publication and at the request of the scientific community as well as the Senior Scientific Advisory Committee for the CRN, Mursu and Jacobs (2012) have since provided an analysis of multivitamin users, other supplement users, and true nonusers in the form of a letter to the editor of the Archives of Internal Medicine. The authors contend that “no inverse association of multivitamins with total mortality emerged from the analysis.” The analysis also revealed benefit for users of other supplements aside from multivitamins (Mursu & Jacobs, 2012) debatably driven by calcium, which is also present in most multivitamins. These conflicting data suggest that this cohort and/or the methodology used were too confounded to appropriately examine vitamin and mineral supplements. Supplement Users Were Healthier than Nonusers Numerous observational analyses, including the study by Mursu et al. (Mursu et al., 2011), have confirmed that supplement users tend to be “healthier” than nonusers and practice a variety of healthy lifestyle practices (e.g., more likely to exercise, less likely to smoke, etc) (Bailey et al., 2011; Council for Responsible Nutrition, 2008; Ervin, Wright, & Kennedy-Stevenson, 1999). The authors’ own findings demonstrate that supplement users had a lower prevalence of diabetes mellitus, high blood pressure, and smoking habits at baseline. Supplement users also had a lower body mass index and waist-to-hip ratio, were more physically active, were likely to have achieved a higher educational level, and were more likely to use estrogen replacement therapy compared with nonusers. In addition, supplement users had a lower intake of energy, total fat, monounsaturated fatty acids, and saturated fatty acids as well as a higher intake of protein, carbohydrates, polyunsaturated fatty acids, alcohol, whole-grain products, fruits, and vegetables compared with nonusers (Mursu et al., 2011). The reported increase in mortality rates among elderly women was only statistically significant after the multivariate adjustment (except copper) for these healthy lifestyle factors in which supplement users have been traditionally shown to exhibit better habits versus nonusers (Bailey et al., 2011; Council for Responsible Nutrition, 2008; Ervin et al., 1999; Mursu et al., 2011). The majority of the hazard ratios for the vitamin and/or mineral supplements included in the study had a 95% Confidence Interval (95% CI) that spanned 1.0 in the age- and energyadjusted model as well as both of the multivariate-adjusted version 1 and 2 models

4

Wallace et al.


(see Table 2 in the study by Mursu et al. (Mursu et al., 2011)). Associations with relative risks 1.20. Furthermore, the authors did not include covariates for individual supplements in their multivariate regression models. The Number of Supplement Users Increased Throughout the Study Separately, there seems to be a marked difference in the survival of nonusers of supplements during the 18-year observational period. At baseline, supplement users (24,329 of 38,772) comprised 62.7% of the study population. By 2004, surviving supplement users in the study (16,278 of 19,124) comprised 85.1% of the study population, during which time the supplement users should have experienced greater mortality according to the study’s conclusions. In 2004, approximately 66.9% of supplement users and only 19.7% of nonusers were available for follow-up (Table 1 in the study by Mursu et al. (Mursu et al., 2011)). The NHANES population survey data have shown an increase in dietary supplement use among women from 44% in 1988–1992 to 53% in 2003–2006 (Bailey et al., 2011). However, this increase is still too small to explain the extremely large difference in the surviving number of users versus nonusers available for follow-up in 2004. Moreover, the study fails to reconcile its conclusions with well-documented human behavior with regard to supplement usage. Individuals often start taking vitamins and minerals after a disease diagnosis; they do not become ill after starting a multiple- or single-nutrient supplement (The Nutrition Source, 2012). Mursu et al. (2011) did not exclude women with illness at baseline; consequently, the authors cannot discern whether the women were already taking vitamins and/or minerals when they became ill or they became ill and then started taking dietary supplements (Li, Kaaks, Linseisen, & Rohrmann, 2011). Dosages of Iron Reported Are Extremely Unlikely in OTC Supplement Products The reported outcomes for use of iron supplements established the highest hazard ratios, reaching 2.01 for those participants who reported using >400 mg/day in the final 4-year of the study. However, the reported dosages of iron at >400 mg/day would be extremely unlikely in over-the-counter iron-containing supplement products (multivitamins contain ∼18 mg of iron). If in fact there was a correlation between high-dosage iron use and mortality risk, this would most certainly have comprised cases in which the participants were under a doctor’s care for anemia of chronic disease or other blood iron deficiencies, and were receiving treatment using prescription-only iron-containing medications. The analysis by Mursu et al. (2011) ignores the likelihood that the presence of an underlying disease may have contributed to the mortality rate of this group of supplement users. Interestingly throughout the subsequent commentary, the authors encourage


5

high-dose supplementation to treat symptomatic deficiency disorders and discourage over-the-counter selections of supplements at far lower doses.


THE VITAMIN AND MINERAL CONTEXT Although the literature is limited, other similar observational studies have also shown positive effects on the risk mortality for both multivitamins (Park, Murphy, & Wilkens, 2011; Rautiainen et al., 2010) and single-nutrient supplements (DIPART (Vitamin D Individual Patient Analysis of Randomized Trials) Group, 2010; Lee, Willett, & Fuchs 2011; Patterson, White, Kristal, Neuhouser, & Potter, 1997). The Multiethnic Cohort Study most notably showed a decrease in cancer incidence, and thus mortality, with long-term consumption of multivitamins (Park et al., 2011). Multivitamins were inversely associated with myocardial infarction in a cohort of Swedish women, especially when used for long term by women without cardiovascular disease (Rautiainen et al., 2010), directly contradicting the findings of Mursu et al. (Mursu et al., 2011). Reduction in myocardial infarction was greatest among women who reported use of multivitamins for >5 years (relative risk = 0.59; 95% CI = 0.44–0.80) (Rautiainen et al., 2010). Observational trials are an essential component of evidence-based nutrition but do not offer certainty because other data, such as the one generated from randomized controlled trials, are equally important with regard to the totality of evidence. To date, few randomized controlled trials have addressed the efficacy of multivitamin use in chronic disease prevention and health promotion in the general US population. In 2006, the US National Institutes of Health examined the available randomized controlled trial data, excluding many observational trials from its review, and concluded that there was inadequate evidence to recommend either for or against the use of multivitamins for the reduction of chronic disease; however, there was some evidence to suggest that these dietary supplements may be protective against cancer in individuals with suboptimal nutritional status (National Institutes of Health, 2006). These findings were based on the conclusions of three manuscripts published from two large multiple nutrient trials. In China, overall cancer incidence and mortality rates were significantly reduced, as were incidence and mortality rates for the two leading types of cancer, esophageal and gastric, in an arm of the study that included vitamin E, ß-carotene, and selenium. Another arm of the study, on zinc and vitamin A, was associated with reduction in non-cardia gastric cancer, although other gastric and esophageal cancers were not reduced (Blot et al., 1993). In France, an intervention consisting of vitamin E, selenium, vitamin C, ß-carotene, and zinc was associated with a reduction in overall cancer incidence in men only, but no individual cancer was clearly reduced. Overall mortality rates in men were also lower in the intervention group (Hercberg et al., 2004). Among men in the French study with normal prostate-specific antigen levels, the intervention was associated with a lower incidence of prostate cancer, but prostate cancer incidence was higher among men with high prostate-specific antigen levels at baseline (Meyer et al., 2005). The US Agency for Healthcare Research and Quality found no consistent pattern of increased adverse effects of multivitamins except for skin yellowing by β-carotene (Huang, Caballero, & Chang, 2006). It should be noted that the current literature on multivitamins as well as many other vitamin and/or mineral

6

Wallace et al.

supplements and their potential to influence chronic disease among the general population is conflicting and certainly not robust. Although there is no doubt that vitamin and mineral supplementation is beneficial in certain subpopulations, additional research has been funded by both industry and the National Institutes of Health to confirm the potential long-term benefits in consumers of products such as multivitamins (see www.clinicaltrials.gov).


CONCLUSION We and others have reported on the limitations of the analysis and conclusions offered by Mursu et al. (2011). Other expert opinions are available on the “Rapid Responses to the Article” section of the Archives of Internal Medicine website (Archives of Internal Medicine, 2011) as well as on the Harvard University School of Public Health website (The Nutrition Source, 2012) and in the peer-reviewed literature (MacKay, 2011). Because of these inherent limitations, Mursu et al. (2011) provide little credible evidence to support the conclusion that vitamin and/or mineral supplements in modest doses can increase the risk of mortality. The authors of this commentary judge that the position of Mursu et al. (2011) should be viewed as a hypothesis that warrants further research to validate whether some risks are indeed present. It is also important to note that the findings of Mursu et al. (2011) contradict other clinical findings for nutrients, such as vitamin D, for which at least two highly regarded studies, a meta-analysis and systematic review, show a protective effect against mortality (Autier & Gandini, 2007; Bjelakovic, Gluud, & Nikolova, 2011). The Senior Scientific Advisory Committee of the CRN encourages the authors, the National Institutes of Health, and the principal investigator of the Iowa Women’s Health Study to develop a data-sharing plan of this publicly funded dataset (concealing the identity of the participants) so that other academic researchers and epidemiologists can independently examine the data. In addition, there is value in continuing to study other large cohorts and subpopulations to inquire if similarities in results exist. It is recommended that future studies intended to evaluate risks and benefits of dietary supplements be designed in a manner that allows reliable evaluation of clinical outcomes, including proper ascertainment of events and comprehensive documentation of known risk determinants. Designing a cohort to specifically examine vitamin and mineral supplements free of confounding from factors such as hormone replacement therapy would give researchers better tools to assess the risks and benefits to the general population, as large randomized controlled trials are already underway. Declaration of interest: All authors are currently employed by the dietary supplement industry. The Council for Responsible Nutrition funded the organization and preparation of the manuscript. CRN is the leading trade group representing responsible dietary supplement manufacturers and ingredient suppliers. ABOUT THE AUTHORS Dr Taylor C Wallace, PhD, FACN, and Dr Douglas Mackay, ND, are affiliated with Council for Responsible Nutrition, Washington, DC. Dr Barry W Ritz, PhD,


7


is associated with Atrium Innovations, Chadds Ford, Pennsylvania. Dr Michael McBurney, PhD, is associated with DSM Nutritional Products – North America, Parsippany, New Jersey. Dr Andrew Shao, PhD, is with Herbalife International of America Inc, Torrance, California. John Miller, BS, is with GNLD International, Fremont, California. Dr James Brooks, PhD, is associated with Pharmavite, LLC, Northridge, California. Lewis Hendricks, MS, is associated with Innophos Inc., Cranbury, New Jersey. Dr. Wallace wrote the manuscript. MacKay, Ritz, McBurney, Shao, Miller, Books, and Hendricks contributed equally to the planning, preparation, and writing of the manuscript. All authors read and approved the final version.

REFERENCES Archives of Internal Medicine. Rapid responses to the article. Available at: http://archinte. ama-assn.org/cgi/content/short/171/18/1625. 2011 Accessed March 18, 2012. Autier P, Gandini S. Vitamin D supplementation and total mortality: a meta-analysis of randomized controlled trials. Arch Intern Med. 2007;167(16):1730–1737. Bailey RL, Gahche JJ, Lenino CV, Dwyer JT, Engel JS, Thomas PR, Betz JM, Sempos CT, Picciano MF. Dietary supplement use in the United States, 2003–2006. J Nutr. 2011;141(2):261–266. Bjelakovic G, Gluud LL, Nikolova D, Whitfield K, Wetterslev J, Simonetti RG, Bjelakovic M, Gluud C. Vitamin D supplementation for prevention of mortality in adults. Cochrane Database Syst Rev. 2011;(7):CD007470. Blot WJ, Li JY, Taylor PR, Guo W, Dawsey S, Wang GQ. Nutrition intervention trials in Linxian, China: supplementation with specific vitamin/mineral combinations, cancer incidence, and disease-specific mortality in the general population. J Natl Cancer Inst. 1993;85:1483–1492. Council for Responsible Nutrition. Healthy lifestyle habits of supplement users & nonusers. Washington, DC: Council for Responsible Nutrition; 2008. Available at: http://www. crnusa.org. Accessed March 18, 2012. DIPART (Vitamin D Individual Patient Analysis of Randomized Trials) Group. Patient level pooled analysis of 68,500 patients from seven major vitamin D fracture trials in US and Europe. BMJ. 2010;340:b5463. Ervin RB, Wright JB, Kennedy-Stevenson J. Use of dietary supplements in the United States 1988–1992. Vital Health Stat. 1999;244(11):1–14. Fulgoni VL, Keast DR, Bailey RL, Dwyer JT. Foods, fortificants and supplements: where do American’s get their nutrients? J Nutr. 2011;141(10):1847–1854. Gahche J, Bailey R, Burt V, Hughes J, Yetley E, Dwyer J, Picciano MF, McDowell M, Sempos C. Dietary supplement use among U.S. adults has increased since NHANES III (1988–1994). Hyattsville, MD: National Center for Health Statistics, 2011. Hercberg S, Galan P, Preziosi P, Bertrais S, Mennen L, Malvy D. The SU.VI.MAX Study: a randomized, placebo-controlled trial of the health effects of antioxidant vitamins and minerals. Arch Intern Med. 2004;164:2335–2342. Huang HY, Caballero B, Chang S, Alberg A, Semba R, Schneyer C, Wilson RF, Cheng T-Y, Prokopowicz G, Barnes GJ II, Vassy J, Bass EB. Multivitamin/multimineral supplements and prevention of chronic disease. Evidence report/technology assessment No. 139. (Prepared by The Johns Hopkins University Evidence-based Practice Center under Contract No. 29002-0018). Baltimore, MD: Agency for Healthcare Research and Quality, 2006. Available at: http://www.ahrq.gov/downloads/pub/evidence/pdf/multivit/multivit.pdf. Accessed March 18, 2012. Lee JE, Willett WC, Fuchs CS, Smith-Warner SA, Wu K, Giovannucci E. Folate intake and risk of colorectal cancer and adenoma: modification by time. Am J Clin Nutr. 2011;93(4):817–825.


8

Wallace et al.

Li K, Kaaks R, Linseisen J, Rohrmann S. Vitamin/mineral supplementation and cancer, cardiovascular, and all-cause mortality in a German prospective cohort (EPIC-Heidelberg). Eur J Nutr. 2012;51(4):407–13. MacKay D: Dietary supplements and mortality rate in older women. Nat Med. 2011;J3(11):1. MedScape WebMD. The most significant medical advances and events of 2011. Available at: http://www.medscape.com/features/slideshow/year-in-medicine/2011. 2011. Accessed March 18, 2012. Meyer F, Galan P, Douville P, Bairati I, Kegle P, Bertrais S. Antioxidant vitamin and mineral supplementation and prostate cancer prevention in the SU.VI.MAX trial. Int J Cancer. 2005;116:182–186. Mursu J, Jacobs DR, Jr. The importance of translating research skillfully to benefit the publicreply. Arch Intern Med. 2012;172(5):449–450. Mursu J, Robien K, Harnack LJ, Park K, Jacobs DR Jr. Dietary supplements and mortality rate in older women: the Iowa Women’s Study. Arch Intern Med. 2011;171(18):1625–1633. National Institutes of Health. NIH State-of-the-Science Conference Statement on Multivitamin/Mineral Supplements in Chronic Disease Prevention. NIH Consens State Sci Statements. 2006;23(2):1–30. Park S-Y, Murphy SP, Wilkens LR, Henderson BE, Kolonel LN. Multivitamin use and the risk of mortality and cancer incidence: the multiethnic cohort study. Am J Epidemiol. 2011;173(9):906–914. Patterson RE, White E, Kristal AR, Neuhouser ML, Potter JD. Vitamin supplements and cancer risk: the epidemiologic evidence. Cancer Causes Control. 1997;8(5):786–802. Rautiainen S, Akesson A, Levitan EB, Morgenstern R, Mittleman MA, Wolk A. Multivitamin use and the risk of myocardial infarction: a population-based cohort of Swedish women. Am J Clin Nutr. 2010;92(5):1251–1256. Szklo M. The evaluation of epidemiologic evidence for policy-making. Am J Epidemiol. 2001;154(12 Suppl):S13–S17. The Nutrition Source. Nutrition insurance policy: a daily multivitamin – what should I eat? Boston, MA: Harvard University. Available at: http://www.hsph.harvard.edu/nutritionsource/ what-should-you-eat/multivitamin/index.html. Accessed March 18, 2012. US Department of Agriculture, Center for Nutrition Policy and Promotion. Report of the dietary guidelines advisory committee on the dietary guidelines for Americans. Washington, DC: USDA, 2010. Available at: http://www.cnpp.usda.gov/DGAs2010-DGACreport.htm. Accessed March 18, 2012.