Aim To evaluate the potential drug–drug interactions (PDDI) between drugs used by older adults, any associated factors and recommended clinical management. Methods A cross‐sectional, population‐based study was carried out through a home survey of 934 older adults (from December 2009 to April 2010). A questionnaire was applied, and the participants were asked to show all the drugs used and their respective prescriptions, thus providing data to identify polypharmacy, self‐medication and PDDI. PDDI, their consequences, severity and clinical management were identified using Micromedex. Results Overall, 2846 drugs and 665 PDDI were identified, 71.0% of which were moderate and 22.4% serious. The prevalence of PDDI was 36.9%. Drugs with a narrow therapeutic index were involved in 17.0% of the PDDI. The variables female sex (PR = 1.11, 95% CI 1.02–1.20), age ≥80 years (PR = 1.15, 95% CI 1.03–1.28), no polypharmacy (PR = 0.72, 95% CI 0.67–0.78) and no hospitalization in the past year (PR = 0.90, 95% CI 0.82–0.97) remained associated with the presence of three or more PDDI in the final multivariate analysis model. Conclusions Most PDDI were related to routinely used drugs (enalapril, hydrochlorothiazide, calcium, captopril, levothyroxine and simvastatin), and more than one‐third of the older adults were exposed to PDDI with the possible risk of serious health consequences. Drugs with a narrow therapeutic index were involved in several PDDI, with increased risk of toxicity. The clinical management procedures most recommended are dose adjustment and dosing changes, control of the drugs’ serum levels, and monitoring of the clinical conditions. Geriatr Gerontol Int 2017; ••: ••–••.