Objective: To describe the incidence and health resource consequences of self-reported adverse drug reactions (ADRs) and their relation to self-perceived health, number of medications used, and age in a geographically based population 65 years of age and older. Design: A survey of a defined population. Participants: Three thousand, one hundred seventy noninstitutionalized persons 65 years of age and older residing in two Iowa counties. Measurements: Reports of ADRs (elicited using an open-ended questionnaire); names of drugs involved; descriptions of reactions and related health resource use; self-perceived health status (at the first annual follow-up); and number of medications (from the baseline interview). Results: Ten percent (95% Cl, 8.97 to 11.09) of respondents reported one or more ADRs. Three quarters of respondents who reported ADRs contacted a physician. Of these, half indicated that a laboratory test had been ordered, and 7% reported a hospitalization due to the reaction. Advanced age alone was associated with decreased risk in women; a similar trend in men was not statistically significant. However, persons with poorer health status and those who reported the greatest prior use of medications were most likely to report reactions. Conclusions: In this study of noninstitutionalized elderly persons, advanced age did not appear to be associated with increased risk for self-reported ADRs. We could not determine whether the decrease in ADR reports among the oldest respondents represented true diminished ADR occurrence or altered ADR detection and reporting capabilities. When projected to the elderly community-dwelling U.S. population, 2.2 million annual physician visits, 1.1 million laboratory tests, and 146 000 hospitalizations may result from ADRs.
