Background: Treatment with AI therapy for at least five years is recommended to reduce risk of cancer recurrence in postmenopausal women with hormone-positive breast cancer, but up to 50% prematurely discontinue their AI. Prior studies have demonstrated higher rates of nonadherence in older patients. Higher rates of comorbidities are also present in older patients, many of which require treatment with prescription medications. However, in prior studies, the impact of polypharmacy on AI persistence in older women has been variable. We evaluated the relationship between use of prescription and over-the-counter (OTC) medications and AI discontinuation in older women in a randomized controlled trial of AI therapy. Methods: In the Exemestane and Letrozole Pharmacogenetics (ELPh) trial, postmenopausal women with stage 0-III hormone-positive breast cancer were randomized to receive exemestane or letrozole and followed for two years. Prescription medication and OTC supplement usage was prospectively collected. Only enrolled participants at least 65 years of age were included in this analysis. Univariate and multivariable Cox proportional hazards models evaluated the association between baseline prescription medications, OTC medications, initial assigned AI medication, and time to discontinuation of initial AI. Results: 131 women at least 65 years old with baseline medication and OTC supplement information were identified. The average age was 70.4 (SD 5). 58 participants (44%) were randomized to exemestane and 73 (56%) to letrozole. 21% had previously received tamoxifen and 18% received taxane-based chemotherapy. The average number of prescription medications at AI initiation was 4.95, and 127 patients reported taking at least one. The average number of OTC medications was 3.23, and 107 patients reported taking at least one. There was no difference in number of prescription medications or OTC medications by study arm. As previously reported, exemestane was associated with premature discontinuation (HR 1.92 [95%CI 1.06-3.50], p = 0.032). Controlling for AI, the risk of discontinuation decreased by 12% for every additional prescription medication (HR 0.88 [95%CI 0.78-0.99], p = 0.038). OTC medication use was not significantly associated with AI duration. Conclusions: Increased prescription medication usage at AI initiation was associated with decreased risk of premature discontinuation of AI. Further research is needed to evaluate this interaction and identify characteristics that impact AI persistence in older women with breast cancer. Clinical trial information: NCT00228956.