Current issues of ACP Journal Club are published in Annals of Internal Medicine


Vitamin and mineral supplement use was not associated with mortality

ACP J Club. 1993 Sept-Oct;119:54. doi:10.7326/ACPJC-1993-119-2-054

Related Content in the Archives
Vitamin and trace-element supplementation in healthy elderly persons reduced the frequency of infection and nutritional deficiencies and improved immune response

Source Citation

Kim I, Williamson DF, Byers T, Koplan JP. Vitamin and mineral supplement use and mortality in a US cohort. Am J Public Health. 1993 Apr;83:546-50.



To examine the association between reported vitamin and mineral supplement use and mortality.


Cohort study of adults in the First National Health and Nutrition Examination Survey (NHANES I) with mean follow-up of 13 years.


United States.


11 348 of 14 407 adults who were noninstitutionalized and aged 25 to 74 years when recruited from 1971 to 1975 provided initial dietary data. Participants with missing data on education or drug abuse (n = 108) or on vital status in 1987 (n = 474) or date of death (n = 8) were excluded. 10 758 persons participated (mean age 50 y, 6483 women).

Assessment of risk factors

Vitamin and mineral supplement use was reported on the baseline and first follow-up questionnaires. Supplement use was defined as regular (daily use), irregular (< daily to once/wk), and none (< weekly). Potential confounders were age, race, marital status, education, special diet (ordered by a physician because of a specific disease), body mass index, weight status (underweight, normal, overweight), alcohol use, smoking, and medical conditions at baseline. A 24-hour dietary recall measured total calories, protein, carbohydrates, fat, vitamins A and C, thiamin, riboflavin, niacin, calcium, phosphorus, and iron. Estimated nutrient needs had 3 categories: none, 1 or 2, and ≥ 3 nutrients < 80% of the 1989 recommended allowances. Participants were interviewed once from 1971 to 1975 and again during 1981 to 1984.

Main outcome measure

Mortality data from the National Center for Health Statistics.

Main results

22.5% of the participants reported regular use of supplements; 10%, irregular use; and 67.5%, no use. 833 participants died (181 from cancer). Using Cox proportional hazards analyses adjusted for all confounders, regular supplement use was not associated with an altered risk for all-cause mortality (relative risk [RR] for men who used supplements regularly 0.94, 95% CI 0.82 to 1.06; RR for women 1.02, CI 0.90 to 1.17). The risk for mortality from cancer did not differ for regular users of supplements compared with nonusers (RR for men 0.90, CI 0.68 to 1.19; RR for women 0.85, CI 0.64 to 1.14). For women who were underweight, the risk for mortality was increased with regular supplement use (RR 1.97, CI 1.23 to 3.16). Supplement use was not associated with mortality in any of the subgroups defined by dietary intake at baseline.


Vitamin and mineral supplement use by adults in the United States was not associated with a change in mortality.

Source of funding: Not stated.

For article reprint: Dr. I. Kim, DHMH/LFHA, 201 West Preston Street, Baltimore, MD 21201, USA. FAX 410-333-7106.


There is little question that dietary quality and diversity play a major role in reducing the risk for chronic disease (1) and mortality (2). Many Americans assume that supplementing the diet with “essential” nutrients further improves health and longevity. Most follow-up studies assessing the effects of supplement use on mortality have been limited to elderly persons or small groups. Larger cross-sectional studies are unable to capture information on mortality. It is, therefore, heartening to see a well-conducted study on a large, representative sample of adults in the United States.

Kim and colleagues confirmed long-term supplement use (over more than 10 years) with follow-up interviews and carefully controlled for a number of possible confounding factors. They concluded that vitamin supplements do not improve longevity, that healthy persons do not require nutritional supplements, and that supplement use had no effect even on those with estimated dietary intakes below the recommended dietary allowances in ≥ 3 nutrients.

Of course, nonrandomized studies such as this one cannot prove or disprove cause. Because nutrient levels or actual intakes were not measured, we do not know what the supplement users actually took. It is possible that large doses of specific nutrients such as vitamin C, vitamin E, or β-carotene can reduce the risk for certain chronic diseases, and several randomized controlled trials are under way to examine these questions. This study reinforces the conclusions of earlier studies on supplements and mortality and allows physicians to continue to not recommend popping such pills for a long and happy life.

Wayne B. Jonas, MD
National Institutes of HealthBethesda, Maryland, USA

Wayne B. Jonas, MD
National Institutes of Health
Bethesda, Maryland, USA


1. National Research Council. Diet and Health: Implications for Reducing Chronic Disease Risk. Washington, DC: National Academy Press; 1989.

2. Kant AK, Schatzkin A, Harris TB, Ziegler RG, Block G. Dietary diversity and subsequent mortality in the First National Health and Nutrition Examination Survey Epidemiologic Follow-up Study. Am J Clin Nutr. 1993;57:434-40.