CMS Could Improve the Data It Uses To Monitor Antipsychotic Drugs in Nursing Homes

WHY WE DID THIS STUDY

CMS has oversight of Medicare- and Medicaid-certified nursing homes that are responsible for the health and safety of vulnerable residents. CMS is required to monitor nursing home activities, including how nursing homes use antipsychotic drugs to treat residents' various conditions. These drugs can be effective in treating a range of conditions, but they carry risk and must be prescribed appropriately. CMS uses the Minimum Data Set (MDS)—i.e., data that nursing homes self-report—as its sole data source to count the number of nursing home residents receiving antipsychotic drugs. CMS has acknowledged the risk for inappropriate use of antipsychotic drugs. CMS has taken important steps to reduce the use of antipsychotic drugs in nursing homes and could further that progress by collecting more complete data on residents' use of these drugs.

Nursing home residents and their families rely on nursing homes to provide quality care in a safe environment; however, there are reasons for concern specific to the use of antipsychotic drugs. Previous OIG work in 2011 raised quality and safety concerns regarding the high use of antipsychotic medications among nursing home residents. Since then, CMS has taken important steps to monitor the use of these drugs in nursing homes. However, CMS has acknowledged the potential for inconsistencies in the data—self-reported by nursing homes—that it uses to monitor quality and the safe use of antipsychotic drugs.

HOW WE DID THIS STUDY

We compared Medicare claims to MDS records for nursing home residents age 65 and older in 2018. The claims data include prescription drug event records for Part D drugs; Medicare Part A and B claims data from the National Claims History File; and Part C encounter data. We determined the number of residents who had a Part D claim for an antipsychotic drug as compared to the number of residents whom the MDS reported as receiving these drugs. These analyses are not comparable to the CMS quality measure. Further, we determined the extent to which there were inconsistencies between residents' MDS-reported diagnoses of schizophrenia, Huntington's disease, or Tourette's syndrome-three diagnoses that exclude residents from CMS's quality measure regarding antipsychotics-and these residents' Medicare claims in 2017 and 2018 for each respective diagnosis.

WHAT WE FOUND

We found that CMS's use of the MDS as the sole data source to count the number of nursing home residents using antipsychotic drugs may not always provide complete information. This means some residents' use of antipsychotics may not have been detected by CMS's quality measure intended to monitor these drugs. By analyzing a separate data source—Medicare claims—we found that using the MDS did not always result in a complete assessment of the number of residents who are prescribed antipsychotic drugs. Specifically, in 2018, 12,091 Part D beneficiaries who were long-stay residents age 65 and older—5 percent of all such beneficiaries—had a Part D claim for an antipsychotic drug but were not reported in the MDS as receiving an antipsychotic drug. Further, nearly one-third of residents who were reported in the MDS as having schizophrenia—a diagnosis that excludes them from CMS's measure of antipsychotic drug use—did not have any Medicare service claims for that diagnosis. Finally, even for those residents included in the MDS counts, the MDS does not provide important details about the drug use (e.g., which antipsychotic drugs were prescribed; at what quantities and strengths; and for what durations).

These findings suggest that CMS could enhance the information it uses to monitor antipsychotics in nursing homes by using additional data sources in its measurement of this complex issue that is critical for resident health and safety.

WHAT WE RECOMMEND

So that CMS can enhance the information it uses to monitor antipsychotic drugs in nursing homes, we recommend that CMS (1) take additional steps to validate the information reported in MDS assessments and (2) supplement the data it uses to monitor the use of antipsychotic drugs in nursing homes. CMS concurred with both recommendations.