We read with much interest the article by Bates and coworkers1 in the November 1999 issue of the ARCHIVES. The authors searched for patient risk factors for adverse drug events in a series of hospitalized patients with a mean age of less than 60 years. They concluded that patient characteristics, chiefly advanced age and polypharmacy, should not be used for risk stratification. In addition, they stated that impaired renal function is a relatively infrequent problem. Adverse drug reactions (ADRs) currently represent a major threat to older patients, since these patients are the major drug consumers in Western countries, and the consequences of ADRs might be most severe in the frailest subjects. Thus, caution is mandatory when denying the necessity of focusing on older patients for preventing ADRs.