Has CMS’s Partnership with the Nursing Home Industry Really Reduced the Inappropriate Use of Antipsychotic Drugs by Nursing Home Residents as Much as CMS Claims? No.

In September 2014, the Centers for Medicare & Medicaid Services (CMS)[1] reported that the National Partnership to Improve Dementia Care had reduced the use of antipsychotic drugs with nursing home residents by 15.1%, “exceeding” the Partnership’s 15% drug reduction goal for long-stay residents.  That claim of success was overstated.  When CMS originally announced the initial Partnership[2] goals on March 29, 2012, it called for a 15% reduction in antipsychotic drug use for long-stay residents by December 2012.[3]  When that preliminary, first-year goal was not met, CMS extended the goal for another calendar year. 

The 15% reduction was not actually achieved until about December 2013, a full year after its original expected date.  The limited effect of the Partnership was pointed out in a report by the Long Term Care Community Coalition (LTCCC), a statewide New York group that advocates on behalf of nursing home residents.  LTCCC found that if facilities had met the initial goal, as anticipated, by December 2012 (and if they had carried that same goal forward in 2013 and 2014), an additional 80,000 residents would not be taking antipsychotic drugs now.[4]

More troubling, the initial rates of use of antipsychotic drugs by nursing home residents may have been seriously understated.  CMS’s Partnership used facilities’ self-reported resident assessment data (called minimum data set or MDS).  CMS recently reported that MDS data are often inaccurate, when compared with residents’ medical records.[5]  In addition, a new report by the Government Accountability Office (GAO) evaluated, among other issues, the use of antipsychotic drugs by long-stay residents in 2012, using both Medicare Part D and MDS data. With both measures, the GAO found considerably higher rates of antipsychotic drug use by long-stay residents than the baseline rate relied on by the Partnership.[6] 

National Partnership

The Partnership used MDS data to determine the antipsychotic drug use rate at the beginning of the initiative in March 2012.  Using an average of the last three quarters in calendar year 2011, it described the national rate of use for long-stay residents as 23.9%,[7] later reduced to 23.8%. 

In September 2014, CMS announced new Partnership goals for 2015 and 2016 – a 5% decrease in antipsychotic drug use for each year, resulting in a total 25% decrease by 2015 and a total 30% decrease by 2016 over levels identified at the start of the Partnership in 2012.[8]  If achieved, the new goals “assuming a baseline of 23.8 percent, would lead to a prescribing rate of 16.7 percent [of all nursing home residents].”[9]

Long Term Care Community Coalition Report

LTCCC’s report debunked CMS’s claim of success.  It showed that the original nine-month goal of a 15% reduction is the equivalent of a 20% reduction for a calendar year.  Extending the 20% reduction to the two successive years, 2013 and 2014, LTCCC suggested, “[I]f the first goal had been accomplished in the time-frame announced, and that initial modest rate of reduction simply sustained thereafter, drugging rates would have dropped by 15% in the last nine months of 2012, an additional 20% in calendar year 2013 and 20% more in 2014.”[10] 

LTCCC reported that in the baseline quarter preceding the campaign (fourth quarter 2011), 340,752 of the 1,431,731 residents nationwide were given antipsychotic drugs.[11]   The total number of residents declined to 1,383,700 in 2013 and to 1,368,754 in 2014.[12]   Accordingly, 15,000 fewer residents would have received antipsychotic drugs, regardless of the Partnership, “solely because the nursing home population as a whole declined over the course of these years.”[13] 

LTCCC reported that the Partnership data indicate that 60,000 fewer residents were inappropriately receiving antipsychotic drugs as a result of the Partnership.[14]   It then suggested, “However, if the campaign had just accomplished and kept with its original, modest goal rate of reduction, that reduction would have been over 140,000” – a difference of 80,000 residents.[15] 

GAO Report

The GAO reported higher antipsychotic drug rates for long-stay nursing home residents in 2012 than the Partnership used as its baseline rate.  The GAO used two sets of data.

First, using Medicare Part D Prescription Drug Event data for antipsychotic drugs for long-stay nursing home residents, the GAO reported that in 2012, approximately 33% of long-stay residents with Part D plans were prescribed antipsychotic drugs.[16]  (It also reported that “about 28 percent of long-stay Medicare Part D enrollees with dementia were given three or more 30-day supply prescriptions for an antipsychotic drug over the course of 2012,” with 68% of the long-stay Part D enrollees receiving seven or more 30-day supplies of the drug.[17]  In addition, since 2013, the Medicare Part D Compare Website has reported chronic use [defined as three or more months] of atypical antipsychotic drugs by nursing home residents.  The GAO reports, “The average among all Medicare Part D plans in 2012 was approximately 22 percent of enrollees residing in nursing homes having at least 3 months or more of a prescription [for atypical antipsychotic drugs only].”[18])  

The GAO report did not identify how many long-stay residents had Part D plans.  However, if we assume that many do,[19] it may be inferred that antipsychotic drug rates for long-stay residents in 2012 were considerably higher than the 23.8% rate used by the National Partnership as its baseline.

Next, analyzing MDS data for 2012, the GAO reported “approximately 30 percent of all older adult nursing home residents – regardless of enrollment in Medicare Part D – with a dementia diagnosis were prescribed an antipsychotic drug at some point during their 2012 nursing home stay.”[20]  This 30% figure is also higher than the 23.8% rate used by the Partnership.

Medicare Payments for Antipsychotic Drugs

The GAO reports that Part D plans paid $171 million in 2012 for antipsychotic drugs for residents who have dementia.[21]  Antipsychotic drug costs for residents in a Medicare Part A stay are not separately calculated because the Part A payment is all-inclusive, covering all care and services provided to residents during the stay.  In addition, Medicare pays for the costs of adverse events, including hospitalizations that may result from the inappropriate use of antipsychotic drugs.

Conclusion

The reduction of antipsychotic drugs in nursing homes is critically important.  For residents with dementia, use of the drugs can be dangerous and even life-threatening.[22]  While fewer residents are inappropriately receiving antipsychotic drugs as a result of the Partnership, more than 200,000 residents who should not be taking these drugs continue to take them.  Much more action is needed.  CMS must expand its efforts beyond the Partnership and enforce federal law requirements that prohibit the use of the drugs.

T. Edelman – March 2015