Abstract and Introduction
Background The effect of antioxidant vitamin and mineral supplementation on health is one of the most controversial issues in human nutrition. Our objective was to investigate the effect of nutritional doses of a combination of antioxidant vitamins and minerals on health-related quality of life (HRQoL) in a sample of healthy French adults.
Methods SU.VI.MAX is a randomized, double-blind, placebo-controlled, primary prevention trial in which a total of 8112 participants received a single capsule daily containing either placebo or vitamin C 120 mg, vitamin E 30 mg, beta-carotene 6 mg, selenium 100 μg and zinc 20 mg. Participants completed HRQoL questionnaires (SF36 and GHQ12) at baseline and after a mean of 76.0 ± 4.2 months.
Results Scores for physical dimensions tended to decrease over time, whereas those for mental dimensions tended to improve. No differences in changes over time were observed between the supplement and placebo groups. Participants who believed that they received placebo had lower HRQoL scores than did those who thought they had received supplements [SF36 Bodily pain (−3.3), General health (−2.2), Vitality (−1.6) dimensions and physical component summary score (−1.1) in men, and in SF36 Social functioning (−2.3), General health (−1.4) dimensions and physical component summary score (−0.7) in women].
Conclusions Long-term supplementation with antioxidant vitamins and minerals had no beneficial effect on HRQoL in this trial. This is contrary to conventional beliefs and claims that such an effect exists.
Trial Registration "Primary Prevention Trial of the Health Effects of Antioxidant Vitamins and Minerals." NTC n 00272428 https://www.clinicaltrials.gov
The effect of antioxidant vitamin and mineral supplementation on health is one of the most controversial issues in human nutrition. Randomized, placebo-controlled trials of long-term, high-dose administration of antioxidant micronutrients have failed to detect a benefit;[1–5] and two studies involving high-risk participants[2,3] suggested that the treatment may even be harmful. A recent meta-analysis concluded that high-dose (≥400 IU/d) vitamin E supplements have a harmful effect on all-cause mortality. Combining 47 low bias risk randomized trials, Bjelakovic et al. showed a 5% mortality increase [95% confidence interval (CI) 1.02–1.08] in the group receiving antioxidant supplements.
Nevertheless, the use of antioxidant vitamins and minerals is widespread in the USA,[8–11] and is increasing in European countries.[12–16] Most people who take supplements do so because of the conventional beliefs that vitamins and minerals may have positive effects on health, well-being and quality of life, and because of claims made by the manufacturers of the countless products currently on the market. The lack of controlled data to support the perception of benefit among consumers, and the numerous and diverse claims of manufacturers (particularly those related to perceived health), may be due to the lack, until recently, of reliable tools with which to assess health as anything more than the absence of disease. However, advances in the assessment of health-related quality of life (HRQoL) now make it possible to measure health as perceived by participants themselves (an increasingly common secondary endpoint in clinical trials).
The present paper reports the application of generic measures in the 'SUpplementation en VItamines et Minéraux AntioXydants' (SU.VI.MAX) trial to assess the effect of long-term antioxidant supplementation on HRQoL used as a secondary endpoint. The primary objective of SU.VI.MAX was to test in a randomized, placebo-controlled trial whether an adequate and well-balanced intake of antioxidant nutrients reduces the incidence of cancers and ischaemic CVD in a middle-aged general population. Concerning the main results regarding primary criteria, no effects were observed on ischaemic cardiovascular events, but a decrease in cancer incidence and in global mortality was observed in antioxidant recipient men during the supplementation period compared with the non-recipients. The nutritional doses used indicate that this result could equally well be achieved by dietary changes. The beneficial effect was not observed in women.
Int J Epidemiol. 2011;40(6):1605-1616. © 2011 Oxford University Press
Copyright 2007 International Epidemiological Association. Published by Oxford University Press. All rights reserved.