OBJECTIVES Adverse drug event (ADEs), hospitalizations, and emergency department (ED) visits are important sequelae of inappropriate prescribing.  Explicit measures are useful methods to detect inappropriate prescribing in the elderly.  This study compares the predictive validity of the 2002 Beers, 2012 Beers, and the STOPP criteria for these outcomes. METHODS A retrospective cohort design was implemented using commercial claims data between 2006 and 2009.  Subjects included those 65 years and older continuously eligible for medical and pharmacy benefits for at least 6 months to assess baseline comorbidities and 3 months of follow-up.  Time varying cox proportional hazard models were estimated using a monthly time interval where indicators of inappropriate prescribing exposure in month(i-1) where related to outcomes in month(i) to minimize temporal ambiguity between exposure and outcome.  Measures of model discrimination (c-index) and hazard ratios (HR) were calculated to compare un-adjusted and adjusted models with the 2002 Beers, 2012 Beers, and STOPP exposures models.   RESULTS The final cohort included 174,275 contributing 361,621 person years of follow up.   The prevalence of inappropriate prescribing was 34.1%, 32.2%, and 27.6% for the 2012 Beers, 2002 Beers, and the STOPP criteria and the rates of ADEs, ED, and hospital visits were 21.0, 140.3, 67.2 per 1000 person years.  All inappropriate prescribing criteria modestly discriminated ADEs in unadjusted analyses:  STOPP  (HR=2.89 [2.68-3.12]; C-index=60.7%), 2012 Beers (HR=2.51 [2.33-2.70]; C-index=60.3%),  2002 Beers (HR=2.65 [2.46-2.85]; C-index=60.5%).  Similar results for model discrimination measures were observed for ED visits (STOPP [C-index=59.0%], 2012 Beers [C-index=58.5%], 2002 Beers [C-index=58.5%]) and hospitalizations (STOPP [C-index=59.8%], 2012 Beers [C-index=59.0%], 2002 Beers [C-index=58.8%]).   CONCLUSIONS All three of the criteria were modestly prognostic for ADEs, EDs, and hospitalizations with the STOPP criteria slightly outperforming the Beers criteria.  Inappropriate prescribing is common in the elderly and significantly increases the risk of subsequent hospitalizations, ED visits, and ADEs.