CMS Mandate for Shared Decision-Making Didn’t Affect ICD Use

Easing some concerns about access, a research letter published online this week in JAMA Internal Medicine shows that a US Centers for Medicare & Medicaid Services (CMS) requirement for shared decision-making prior to placing an implantable cardioverter-defibrillator (ICD) for certain indications did not influence use of the devices over ensuing years.

Primary-prevention ICD use among Medicare beneficiaries had already been declining for at least 2 years before the mandate was introduced in a February 2018 update to the agency’s national coverage determination (NCD) for ICDs, and that trend did not change significantly over the next 2 years.

“Our results do not support the hypothesis that a shared decision-making mandate might reduce access to ICDs,” Joshua Rager, MD (University of Michigan, Ann Arbor), and colleagues write.

The CMS mandate for a shared decision-making encounter between a patient and either a physician or a qualified nonphysician professional (physician assistant, nurse practitioner, or clinical nurse specialist) prior to ICD implantation also includes a requirement to use of an evidence-based decision tool. It applies to ICDs used for primary—but not secondary—prevention of sudden cardiac death.

“We believe that a shared decision-making encounter prior to initial ICD implantation is a critical step in empowering patient choice in their treatment plan. While ICDs have remained a common treatment option for many years, the strength of evidence for an ICD benefit is different for different patient populations,” CMS explained in the NCD, noting that a patient’s health goals, preferences, and values aren’t typically covered during the usual process of informed consent.

Whether this new requirement would create an undue drop in ICD implantations had remained unclear, Rager told TCTMD.

For the study, his team examined data from a random 20% sample of the Carrier and Medicare Provider and Analysis Review files. The analysis included 40,085 Medicare beneficiaries (mean age 72 years; 72% men) who received a de novo ICD between January 2016 and February 2020.

In the current cohort, 59% of ICD recipients had a primary-prevention indication. Those with a secondary-prevention indication were included as comparators.

The rate of primary-prevention ICD use was already decreasing at a rate of 2.37 procedures per 100,000 potentially eligible beneficiaries per month in the 2 years before the CMS mandate was implemented, and the rate of decline after that point (0.77 procedures per 100,000 per month) was not significantly different (P = 0.26).

Although there was a significant reduction in primary-prevention ICD implantations after the mandate when compared with the rate of secondary-prevention implantations, the investigators note that the difference can be attributed to a slower-than-expected decline observed in the latter group.

“We do not believe this reflects a true decrease in primary-prevention ICDs associated with the mandate,” Rager et al write. “This plateau may, however, suggest that the mandate unintentionally incentivized billing different, non-primary-prevention diagnoses to avoid shared decision-making requirements, although more research is needed.”

Why No Impact?

As for why the change in CMS policy wouldn’t have an impact on ICD use, Rager said that “it may not have actually resulted in the shared decision-making that I think of when I think of shared decision-making.” In addition, he said, the aim of requiring this process isn’t necessarily a change in device use.

“Really the goal of shared decision-making is to align patient preferences with treatment received,” he said. Regarding a change in device use, he added, “maybe that’s not the goal we’re really looking for here.”

Moreover, it’s possible that electrophysiologists were already having these types of discussions with patients in the years leading up to the CMS decision. Of note, in 2016, the DANISH trial showed that ICDs did not reduce the risk of death in patients with nonischemic cardiomyopathy, Rager pointed out, suggesting that those results and an increasing emphasis on shared decision-making in major guidelines were already influencing how physicians were talking about ICDs with their patients.

Discussing the overarching motivation for requiring shared decision-making, Rager said that “patients have varying preferences that can be reasonable for devices, and devices can be consequential, meaning that that person then lives with this thing and that decision matters. And it’s good to think about the longer-term outcomes of making that decision.”

One consideration, he indicated, is that clinicians need to be empowered to have these types of discussions with their patients, which add time to visits. “We need to have mechanisms that allow for that sort of compensation, and we want to incentivize that sort of process. So I think efforts to do that would be good, and I think future mandates need to be mindful of how we’re going to make sure this is actually happening, aligning patient preferences ultimately with treatment received.”

There’s work that still needs to be completed, especially around how to track whether such mandates are accomplishing what they’re designed to do, Rager added, “but I think it’s sort of an overall big push to [be] really leveling with patients, getting their perspectives, understanding their values, and aligning that with care.”