Between 2007 and 2015, the volume and proportion of primary prevention ICD implantations that did not meet the Centers for Medicare and Medicaid Services (CMS) National Coverage Decision (NCD) decreased at all hospitals across the U.S., according to a study published July 3 in the Journal of the American Medical Association. There were larger declines at hospitals that later settled with the Department of Justice (DOJ) after a DOJ investigation into potential overuse of ICDs.

For the study, Nihar R. Desai, MD, MPH, FACC, et al., examined changes in the number of ICD implantations that did not meet NCD criteria after the DOJ investigation was announced in January 2011, both at hospitals that did and did not settle. Using data from ACC's ICD Registry, the researchers identified 309,850 primary prevention ICD implantations for ischemic and nonischemic cardiomyopathy among Medicare beneficiaries between 2007 and 2015 at 1,809 hospitals. The study's primary outcome was the proportion of primary prevention ICD procedures that did not follow NCD criteria at each six-month interval throughout the study period. The researchers calculated the proportion of ICD procedures that did not meet NCD criteria for each six-month interval and over the entire study period, stratified by whether the hospital later settled. To examine associations with the DOJ investigation, the researchers calculated the proportion of ICDs not meeting NCD criteria in three time periods – January 2007 to December 2009, before the DOJ announcement of the investigation; January 2010 to June 2011, during the DOJ investigation announcement; and July 2011 to December 2015, after the DOJ investigation announcement.

According to the results, there were "significant declines" in the proportion of primary prevention ICD procedures that did not meet NCD criteria at all hospitals. Following the announcement of DOJ investigation, the declines were larger and more rapid at hospitals that later reached settlements. In January 2007, ICDs did not meet NCD criteria at 25.8 percent of hospitals that did settle and at 22.8 percent of hospitals that did not settle. During the study period, there was a 62.7 percent relative decline and 16.1 percent absolute decline in proportion of ICDs that did not meet NCD criteria at the hospitals that later settled. In comparison, hospitals that did not settle experienced a 53.2 percent relative decline and 12.1 percent absolute decline.

The researchers attributed the overall declines to three distinct trends. Between January 2007 and December 2009, prior to the DOJ investigation, the proportion of primary prevention ICDs that did not meet NCD criteria was relatively stable and the rate of decline was modest and similar among hospitals that did and did not settle. Second, between January 2010 and July 2011, there were significant declines in ICDs that did not meet NCD criteria in both hospital groups. Finally, between July 2011 and December 2015, the proportion of ICDs that did not meet NCD criteria was similar and stable in both groups.

According to the authors, further research is needed into the agreement between the NCD, clinical practice guidelines and appropriateness. The authors conclude that additional research into whether the DOJ investigation led to "potential unintended consequences" is needed, adding that studying hospital responses to the investigation "could offer additional insights."

As the availability of big data and reliance on appropriate use criteria increase, whistleblower complaints similar with the one that led to the DOJ investigation "may be anticipated" in the future, Paul A. Heidenreich, MD, MS, FACC, writes in an accompanying editorial. "Physicians and hospitals can lead … by developing appropriate use criteria and participating in registries tracking appropriate use," Heidenreich concludes.

The DOJ investigation did not involve ICD Registry data. In February 2018, CMS released a final policy decision that included minimal revisions to the NCD, which remains in effect for ICD implantation for Medicare fee-for-service patients.

Keywords: Centers for Medicare and Medicaid Services, U.S., Medicare, Defibrillators, Implantable, Registries, National Cardiovascular Data Registries, ICD Registry, Cardiomyopathies

< Back to Listings