OBJECTIVE: The Minimum Data Set (MDS) is an international facility-reported database used to standardize the assessment process and improve the quality of care in nursing homes. Specific MDS assessments are used as quality indicators to determine the prevalence of antipsychotic prescribing as well as potentially inappropriate antipsychotic usage. MDS classifies low-risk residents as receiving an antipsychotic medication but having no documented psychotic or related condition and no cognitive impairment with behavioral symptoms. Without an audit of resident records, it is impossible to determine if the MDS accurately reports the prevalence of antipsychotic medications and whether all low-risk residents represent inappropriate prescribing. The purpose of this study was to determine MDS validity for identifying low-risk residents in Texas. METHODS: Facility-based MDS assessments were compared to independent on-site assessments. A team of pharmacists reviewed the chart records of low risk residents and determined whether a resident received an antipsychotic medication and the presence or absence of a valid indication. These findings were compared to the statewide MDS quality indicator prevalence rate. RESULTS: Based on MDS assessments, 10,163 residents from 1021 nursing homes were identified as low risk. The chart records revealed that 12.4% of these residents were not receiving antipsychotic medication. Of the 8904 residents who were receiving an antipsychotic, 48.2% had a valid diagnosis or indication and were mistakenly identified as low-risk by the MDS database. Of the remaining 4610 residents, 84.4% were missing a diagnosis and 15.6% had an inappropriate indication. CONCLUSION: For this population of residents, a lack of accurate essential diagnostic information in MDS assessments and a high prevalence of inappropriate antipsychotic prescribing was found to be a widespread problem. These findings indicate that nursing facilities were inaccurate in their responses to MDS assessment items and therefore, the MDS had poor validity as a quality indicator for antipsychotic use.