Did CMS’s Partnership to Improve Dementia Care Reduce the Inappropriate Prescribing of Antipsychotic Drugs for Nursing Home Residents as Much as CMS Claims? Two Recent Studies Say No

The Centers for Medicare & Medicaid Services (CMS) has ended its campaign to reduce the inappropriate use of antipsychotic drugs for long-stay residents in nursing facilities (formally called the National Partnership to Improve Dementia Care in Nursing Homes) for facilities that reduced their antipsychotic drug usage by 34% by the end of 2016 (from 23.9% of long-stay residents in the first quarter of 2012 down to 15.7% of long-stay residents in the first quarter of 2017).  No more is expected of these facilities; CMS encourages them to “continue their efforts and maintain their success.”[1]  Only facilities not yet achieving the reduction – facilities that CMS now calls “late adopters” – have new Partnership goals; CMS encourages these facilities to reduce antipsychotic drug use by 15% by 2019.[2]  The American Health Care Association (AHCA), the largest nursing home trade association, also touts the reduction of antipsychotic drugs by its member facilities.[3]  Are these claims of victory – of actually reducing the use of these drugs – true?  Two recent studies say no; the claims of success are grossly overstated.

A study in the Journal of the American Medical Association Internal Medicine debunks claims that the National Partnership achieved its goals.[4]  First, the rate of antipsychotic drug use in nursing facilities was declining before the Partnership began (from 29% in 2006 to 22% in 2009-2010) so that the Partnership did not “accelerate[] the decline that was already occurring.”  Second, researchers found a lot of drug substitution.  Instead of antipsychotic drugs, residents were given other drugs, particularly mood stabilizers that are not captured in CMS assessment data.  Reviewers based their analysis on 637,426 residents (a 20% sample of nursing home residents using traditional Medicare between January 1, 2009 and December 31, 2014) and the Part D drugs they received.

Recent reports by CNN also confirm that drug substitution is occurring.  In October 2017, CNN reported that Avanir, the manufacturer of Nuedexta – a drug approved by the Food and Drug Administration to treat pseudobulbar affect, or PBS (uncontrollable laughing or crying), a condition affecting less than 1% of the population – targeted advertising at nursing facilities that had high rates of antipsychotic drug use.[5]  Since 2012, “more than half of all Nuedexta pills have gone to long-term care facilities” and “The number of pills rose to roughly 1.4 million in 2016, a jump of nearly 400% in just four years.”[6]  CNN reported that in March 2018, the Centers for Medicare & Medicaid Services issued a memorandum to Part D plans, reminding them, as described by CNN, that “Nuedexta is only approved to treat PBA” and that “Part D insurers are legally required to ensure the drug is only being covered when prescribed for medically-accepted uses.”[7] 

A second study found that the 20% reduction in the use of antipsychotic drugs was overstated because of the increased reporting of exclusionary diagnoses, chiefly schizophrenia.[8]  CMS excludes residents with schizophrenia (and also Tourettes and Huntingtons) from the publicly-reported quality measure that define inappropriate antipsychotic drug use.  Consequently, by including one of the three excluded diagnoses in residents’ assessments, facilities avoid being cited as administering inappropriate antipsychotic drugs.  The researchers conclude, “since antipsychotics prescribed to patients with schizophrenia, Tourette's syndrome, and Huntington's disease is not reported to CMS's quality measure, these findings suggest that a significant portion, perhaps as much as l/5th, of the apparent reduction in inappropriate antipsychotic use in nursing homes, is due to this trend in diagnosis reporting rather than an actual decrease in medication use.”  Researchers based their findings on resident assessment data for Medicaid beneficiaries in Virginia.  The National Institute of Mental Health reports that schizophrenia “is typically diagnosed in the late teen years to early 30s.”[9]  Consequently, the “new” diagnoses of schizophrenia in nursing home residents is likely to be false and fraudulent.    

Conclusion

Human Rights Watch reported in February that 179,000 nursing home residents are receiving antipsychotic drugs now.[10]  Other reports describe other rates of antipsychotic drug use.  Reports vary dramatically, depending on which residents are reviewed (long-stay, short-stay, all) and whether analysts look at residents during a Medicare Part A stay or residents receiving medications under Medicare Part D.[11]  

Regardless of how the numbers are calculated, however, there is no question that many tens of thousands of residents are getting antipsychotic drugs who should not be.  This problem is not solved, despite CMS’s, and the industry’s, claims of victory.

June 28, 2018 – T. Edelman.