Trends in Overall and Nutrient-Containing Dietary Supplement Use Among the U.S. Population (≥1 y), NHANES 2007–2018

OBJECTIVES: DS use is widespread in the U.S. and contributes large amounts of micronutrients to users. The prevalence of DS use over time has been characterized; however, most studies have relied on data from one assessment method. Combining multiple methods of assessment enhances the ability to capture nutrient exposures from DS; additional studies updating prevalence of DS use estimates using this approach are needed, as well as documentation of a classification system for DS products. Thus, the objective was to characterize the prevalence of DS use over time among a nationally-representative sample of the U.S. population (≥ 1y), using a combined approach of DS assessment and a DS classification system. METHODS: NHANES obtains DS use via a 30-day frequency method and an in-home inventory (Dietary Supplement and Prescription Medicine Questionnaire (DSMQ)), and two 24-hr recalls (24 HR). All NHANES cycles with 24 HR and DSMQ data were used to characterize DS use and examine trends in any and nutrient-containing (VM-only) DS use, using the DSMQ or ≥1 24 HR. Trends in the prevalence of use and selected types of products used were estimated for the population and by sex, age, race/Hispanic origin, family income (PIR ≤ 130%, 131 to 350%, and >350%), and household food security (food secure vs. insecure). Linear trends were tested using orthogonal polynomials; significance was set at p < 0.05. RESULTS: DS use increased from 50% in 2007 to 56% in 2018 (p = 0.001). While use of VM-only products increased (46% to 49%; p = 0.029), multivitamin-mineral use decreased (p < 0.001). Single-nutrient DS (e.g., magnesium, vitamins B12 and D), also increased in use (all p < 0.001). Among adults (≥19 y), any (54% to 61%) and VM-only (49% to 54%) DS use increased, specifically among men, non-Hispanic Blacks and Hispanics, and low-income (PIR ≤ 130%) adults. Among children (1–18 y), any and VM-only DS use remained stable at ∼38%, except for children living with food insecurity, where use increased from 24% to 31% over the decade (p = 0.035). CONCLUSIONS: The prevalence of any and VM-only DS use in the U.S. increased over time and may be attributable to increases in the use of single-nutrient DS. Differences in DS use exist by population subgroup. This study contributes to a small body of literature that utilizes multiple methods of DS assessment designed to increase the accuracy of DS exposure assessment. FUNDING SOURCES: N/A.