ARS | Publication request: Variability of four nutrients evaluated in adult multivitamin/mineral (MVM) products

Submitted to: Experimental Biology
Publication Type: Abstract
Publication Acceptance Date: November 21, 2006
Publication Date: April 25, 2007
Citation: Schweitzer, A., Andrews, K., Roseland, J.M., Zhao, L., Holden, J.M., Perry, C., Douglass, L., Dwyer, J., Picciano, M., Fisher, K., Saldanha, L., Yetley, E. 2007. Variability of four nutrients evaluated in adult multivitamin/mineral (MVM) products. EExperimental Biology, April 25, 2007, Washington, D.C.

Technical Abstract: Nutrient Data Laboratory, USDA, is collaborating with Office of Dietary Supplements to create an analytically verified Dietary Supplement Ingredient Database (DSID). As part of the DSID, variability within and between lots was assessed in MVM products. Using NHANES 2001-02 supplement use data, the three most common %Daily Value (DV) levels were chosen for iron (Fe), vitamin C (VC), folic acid (FA), a-tocopherol (VE). Three MVMs at each %DV level (9 products/nutrient) were chosen randomly. Samples of 2 lots per product were sent to an independent laboratory for analysis. Within-lot sample size (Fe n=2, VC and FA n=3, VE n=4) was estimated using preliminary USDA data on within-day variability in order to detect a 10% difference between lots with 95% confidence. Product coefficients of variability (CV) within-lots were estimated from data expressed as mg/serving (Fe, VC, VE) or µg/serving (FA). CVs by product were 1.7-10.2% for Fe, 1.2-16.4% for VC, 6.3-24.1% for FA, and 2.7-25.2% for VE, indicating lower within-lot product variability for some products and nutrients than other products. ANOVA indicated sample lot-to-lot variance within products and nutrients accounted for <1% of total variability. Although lot-to-lot variance was minimal, additional lots and/or nutrients may need to be tested to assess the impact of between and within lot effects. Funded by USDA & ODS/NIH Y1CN5010