• cross-sectional study;
  • dietary supplements;
  • herbal medicine;
  • elderly

OBJECTIVES: To analyze baseline data from the Ginkgo Evaluation of Memory (GEM) study, in which information was collected on the use of all dietary supplements.

DESIGN: Cross-sectional regression analysis.

SETTING: GEM study sites in California, Maryland, North Carolina, and Pennsylvania.

PARTICIPANTS: The GEM study enrolled 3,072 ambulatory individuals aged 75 and older between September 2000 and June 2002.

MEASUREMENTS: Self-reported use of dietary supplements and use identified through bottles brought to the clinic.

RESULTS: Respectively, 59.4%, 66.6%, and 27.4% of the GEM study cohort used a multivitamin, at least one individual vitamin or mineral supplement, and some type of nonvitamin/nonmineral dietary supplement (NVNMDS). In logistic regression models, multivitamin use was associated with female sex, a higher income, a higher modified Mini-Mental State Examination score, difficulty with mobility, and asthma history; use of any other vitamin or mineral was associated with female sex, white race, nonsmoking, more years of schooling, difficulty walking, a history of osteoporosis, and reading health and senior magazines; and NVNMDS use was associated with residing in California, having difficulties with muscle strength, and reading health and senior magazines.

CONCLUSION: There were substantial differences between individuals who used vitamins and minerals and those who used NVNMDS. These data require that trial investigators pay close attention to participant use of off-protocol dietary supplements. In addition, these findings may help identify elderly individuals likely to combine NVNMDS and prescription drugs.