Harris Interactive's HarrisPollOnline (HPOL) was used to conduct a cross-sectional survey of adults in the United States that evaluated individuals' knowledge about drug safety, the importance they accorded drug safety knowledge, and their preference for active participation in prescribing decisions. Respondents provided free-text response to the question “What does the term drug safety mean to you?” Two authors (K.B.V. and S.L.W.) independently determined whether each category addressed a drug side effect or adverse drug events (ADEs), drug effects appropriate for the condition being treated (physiologic effects), both, or neither. To assess the respondent's level of knowledge about drug safety, we used a composite variable derived from 3 questions, each of which had a 5-level Likert response format to rate their knowledge about drug side effects, serious safety events, and drug-drug interaction. We also asked respondents to rank how important knowledge of drug side effects and drug-drug interactions was to them. By weighting respondent's knowledge of drug safety by its importance, we developed a 3-level categorical variable: low, medium, and high level of knowledge and importance (henceforth noted as weighted knowledge). Our primary outcome was whether the respondents preferred to choose the medications to be prescribed (as appropriate for their condition) or to have the physician select their medication. A secondary outcome was what occurred during the last physician-patient interaction when a medication was prescribed. We evaluated the following covariates for both confounding and effect modification: age, sex, race/ethnicity, education, previous ADEs, respondent-defined drug safety, and whether the respondent sought information on drug safety or was interested in participating in a drug safety monitoring program. We used logistic regression to estimate odds ratios for the 2 outcomes of interest.