NCHS Data Brief
■ 7 ■
interviewers recorded the supplement name, manufacturer, and address, and collected information
on duration of use, frequency of use in the past month or past 30 days, and the amount taken
when used. Product label information was obtained by NCHS nutritionists and is also publicly
released with the dietary supplement-use data.
NHANES data for adults aged 20 and over were used to calculate the prevalence of dietary
supplement intake for three time periods (1988–1994, 1999–2002, and 2003–2006) by
demographic characteristics for ve supplement categories: 1) any supplement, 2) multivitamin/
multimineral (contains at least three vitamins and may or may not contain minerals), 3)
supplement with folic acid, 4) supplement with vitamin D, and 5) supplement with calcium,
including antacids containing calcium. Demographic variables used in the analyses were age,
gender, and race and ethnicity. Race and ethnicity were self-reported and categorized as Mexican
American, non-Hispanic black, non-Hispanic white, and other. People in the “other” racial and
ethnic category were included in analyses of the total sample, but results for this group are not
shown separately. Pregnant women were excluded from the analytic sample. The analytic sample
size was 37,596: 18,504 in NHANES 1988–1994; 9,660 in NHANES 1999–2002; and 9,432 in
NHANES 2003–2006. Response rates were 81% in NHANES 1988–1994, 77% in NHANES
1999–2002, and 73% in NHANES 2003–2006.
All analyses used the interview sample weights, which account for unequal probabilities of
selection, person-level nonresponse, and a poststratication adjustment to the estimated U.S.
population. Taylor series linearization was used to calculate standard errors. All estimates that
were not age-specic were age-adjusted to the 2000 U.S. standard population using three age
groups: 20–39, 40–59, and 60 and over (5). Trends were tested to evaluate changes in estimates
across survey periods. Differences among groups were assessed using a univariate t statistic.
To account for multiple comparisons, statistical signicance was set as a P value less than
0.0175, using the Bonferroni method to adjust for the three comparisons. Statistical analyses
were conducted using SAS version 9.2 (SAS Institute, Cary, N.C.) and SUDAAN version 10.90
(Research Triangle Institute, Research Triangle Park, N.C.).
About the authors
Jaime Gahche, Vicki Burt, and Jeffery Hughes are with the CDC’s NCHS, Division of Health and
Nutrition Examination Surveys. Regan Bailey, Elizabeth Yetley, Johanna Dwyer, Mary Frances
Picciano (deceased, August 2010), and Christopher Sempos are with the National Institutes of
Health’s (NIH) Ofce of Dietary Supplements. Margaret McDowell is with NIH’s National
Institute of Diabetes and Digestive and Kidney Diseases.
1. Heimbach JT. Using the national nutrition monitoring system to prole dietary supplement
use. J Nutr 131(4 Suppl):1335S–8S. 2001.
2. Radimer K, Bindewald B, Hughes J, Ervin B, Swanson C, Picciano MF. Dietary supplement
use by US adults: Data from the National Health and Nutrition Examination Survey, 1999–2000.
Am J Epidemiol 160(4):339–49. 2004.
3. Briefel RR, Johnson CL. Secular trends in dietary intake in the United States. Annu Rev Nutr