OBJECTIVES: To evaluate appropriateness of prescribing medicines using Beers criteria 2003 and 2012 and determining predictors of potentially inappropriate medications (PIMs) prescribing in elderly in-patients. METHODS: cross-sectional study was conducted at public hospital and baseline data were collected. Elderly in-patients from medicine wards (≥60 years) were included. Multivariate logistic regression analysis was used to determine the predictors of PIMs prescribing. RESULTS: 500 patients were recruited during I year of study period;60% were males and 66% were between 60-69 years of age with mean (SD) of 68 (7) years. Mean (SD) number of diagnoses and medications were 3 (1) and 9 (4), respectively. 81 (16%) patients were prescribed with at least ≥1 PIMs according to modified AGS updated Beers criteria 2012, compared to 11% according to Beers criteria 2003. On multivariate regression, important predictors for the PIMs prescribing were found to be age ≥80 years (Odds Ratio (OR) 2.46, 95% CI 1.27-3.12; p = 0.03), male gender (OR 1.35, 95% CI 1.06-1.84; p = 0.03), more than 3 diagnoses (OR 2.47, 95% CI 1.59-3.39; p = 0.04), ≥6 medications prescribed (OR 1.16, 95% CI 1.02-1.35; p = 0.03) and ≥10 days of hospital stay (OR 1.59, 95% CI 1.09-2.31; p = 0.02). CONCLUSIONS: Results indicate that PIMs prescribing is common among hospitalized Indian elderly patients. It is feasible to reduce this practice through provision of appropriate unbiased information to healthcare professionals. Beers criteria is a well established method for evaluating prescribing appropriateness. Results also show the capture of more number of PIMs through the use of Beers criteria 2012 due to the addition of new medications in the list like spironolactone in heart failure and removal of capping of maximum dose of alprazolam, clonazepam and lorazepam from Beers criteria 2003.