The work presented in this dissertation focuses on two important medication-related problems in older adults, that is, untreated indication and drug-drug interactions, specifically with respect to a high-risk medication such as warfarin. Warfarin is a challenge to use in clinical practice due to its narrow therapeutic index, variability in dose-response and its interactions with numerous foods and drugs. This dissertation presents the research from two projects. In the first project the prevalence and predictors of warfarin use in nursing home (NH) residents with atrial fibrillation (AF), and use of secondary stroke prevention strategies was determined, in order to understand the patterns of anticoagulant use in frail NH residents and to identify patient characteristics associated with warfarin use. In the second project the effect of oral antibiotics on anticoagulation outcomes, when prescribed concomitantly with warfarin, was determined, in order to provide evidence on the clinical significance of warfarin-antibiotic interactions in older adults. In the first project a cross-sectional analysis of the prescription and resident files from the 2004 National Nursing Home Survey was done to determine the prevalence of AF and rates of use of warfarin and other anti-platelet agents, such as aspirin and clopidogrel. A multiple logistic regression model was used to determine factors associated with warfarin use. In this sample of older NH residents, 13% of residents had a diagnosis of AF, with indications for warfarin use and no contraindications to warfarin. From these patients, 30% received anticoagulant therapy with warfarin and 23% of the remaining patients received either aspirin or clopidogrel, suggesting that more than 50% of residents with AF did not receive any form of anticoagulant therapy. Non-white race, history of bleeding, and use of anti-platelet medications were associated with reduced odds of receiving warfarin. The second project was a retrospective medical record review of older patients from an outpatient anticoagulation clinic at a Veterans Affairs medical center. Results of the repeated measures ANOVA suggested a significant increase in post-antibiotic INR values with fluoroquinolones, azithromycin and amoxicillin. In addition, the percentage of patients with warfarin dose adjustments was significantly greater with fluoroquinolones and azithromycin as compared to cephalexin. No bleeding events were reported for any of these patients. In conclusion, the results of the projects suggest that there is underuse of warfarin in NH settings. Furthermore, antibiotics may be safely prescribed with warfarin in older adults as long as the INR is monitored closely.