Most US Patients Have TAVR Options Nearby, but Proximity Isn’t Everything

More than 90% of older Americans live within a hospital referral region that includes a TAVR center, although people living in certain parts of the country have to drive a bit longer to reach one, researchers found. The findings might help inform ongoing discussions about how to balance top-quality procedures—that typically require higher patient volumes—against equitable access.

Out of 305 hospital referral regions—areas that include at least one hospital that performs major cardiovascular or neurosurgical procedures—across the United States, 76.7% included a center capable of performing TAVR. According to census data, 92.1% of people 65 and older lived within one of these regions, lead author Guillaume Marquis-Gravel, MD (Duke Clinical Research Institute, Durham, NC), and colleagues report.

The median driving time for patients who underwent a successful transfemoral TAVR during the study period (June 2015 to June 2017) was 35 minutes, with shorter times in the Northeast and longer times in the South and Midwest.

“At an overall level, if you just look at the population of those who are 65 and older, there seems to be adequate access [to TAVR] if you consider the overall healthcare system of the US, where certain services are centralized,” Sreekanth Vemulapalli, MD (Duke Clinical Research Institute), senior author of the study, told TCTMD. But, he added, it’s still unknown whether that access is equitable because there’s no way of knowing from existing sources how many patients who are eligible for TAVR don’t ultimately receive a replacement.

“That would be the piece that we really need to spend time on as researchers and as a country to make sure that we’re not having disparities among subgroups of patients,” he said.

Geography: One Piece of Access

The commercial rollout of TAVR has been steered by procedural volume requirements in the national coverage determination from the Centers for Medicare & Medicaid Services (CMS), first released in 2012 and updated last summer. As part of the debate in recent years about whether such requirements are the right way to ensure high-quality procedures, concerns have been raised about the potential to impede patient access if lower-volume centers are not allowed to perform TAVR. “You can use certain metrics to ensure quality, and volume is one of them. There are metrics that might be better. But that always has to be balanced against people having access to TAVR services,” Vemulapalli said.

However, there was not much data informing the access question, an issue that not only brings in geography but also many other factors like insurance coverage, socioeconomic status, race/ethnicity, and patient preference.

In this study, published online June 10, 2020, ahead of print in JAMA Cardiology, the investigators zeroed in on geography and how TAVR centers were distributed across the US. They drew on data from the Society of Thoracic Surgeons/American College of Cardiology TVT Registry, CMS, the US Census, and Google Maps, looking at the distribution of TAVR centers in two different ways—by zip code and by hospital referral region.

Of 40,537 zip codes in the US, only 1.2% had a TAVR center. Only 2.6% of the roughly 47.5 million Americans 65 and older had a TAVR center in their home zip code. Coverage was much greater when looking at hospital referral regions.

“If you consider having at least one hospital with a TAVR service as being adequate access, then we’re over 90%. If you consider that it needs to be in every person’s zip code, then we’re at the other end of the spectrum and not doing so well,” Vemulapalli said. “So I think that’s where the debate needs to start being focused now, at least from a geographic standpoint: what are we really aiming for from a healthcare policy standpoint?”

The researchers also assessed driving times from home to implanting center among 31,098 patients (mean age 82.4 years; 47.3% women) who underwent successful transfemoral TAVR, 23.9% of whom lived in a rural area. Median driving time was 35 minutes, although the range was wide—from 2 minutes to 18 hours and 48 minutes. There was also regional variation, with the longest driving time in the Midwest (40 minutes) and shortest in the Northeast (31 minutes).

How Patient Preference Plays In

Commenting for TCTMD, Catherine Otto, MD (University of Washington, Seattle), shifted the discussion from geography to patient preference: “The issue that this paper doesn’t really get at is: how does a patient balance the issue of potentially better outcomes at a more expert center that is less convenient versus the outcomes and the experience at a center that is more convenient or more familiar to them?”

The assumption of this analysis, she said, is that patients would always want to go somewhere close to home. “I think that that may be true if the outcomes are equivalent, but I think we haven’t really asked patients what they prefer in terms of distance or outcome,” she continued. “And it may be that if patients are really given adequate data about outcome, they might choose to go somewhere more distant if the outcomes are significantly different.”

I think we haven’t really asked patients what they prefer in terms of distance or outcome. Catherine Otto

Like Vemulapalli, Otto also brought up the issue of equitable access to TAVR. “The real question is: are patients who have severe aortic stenosis getting diagnosed and referred for treatment, and are those who are candidates for [valve replacement] getting the option to consider TAVI versus surgery?” she said. “So I think it’s really about healthcare equity. I don’t think it’s about geography. I think it’s about race, sex, age, socioeconomic status. Zip code gets at socioeconomic status and some of these other issues to some extent, but I don’t think this addresses the healthcare equity issue. And I don’t think this study can answer that question.”

One of the major challenges in medical research, Otto noted, is identifying people who have a condition but don’t get treated for it, and the same goes for patients with severe aortic stenosis who may have benefitted from a valve replacement but didn’t receive one.

Taking a broader view, Otto said “this type of study stimulates the question of how we move into the future to provide really optimal care for patients with valvular disease.” She highlighted a multisociety consensus statement released in April 2019 that proposed a system of care for valvular heart disease that would include primary valve centers that could treat the majority of patients, with a smaller number of comprehensive centers for more-complex patients. And recent developments might add another layer to that, she indicated.

“I think the COVID pandemic has shown us that we might be able to take this even further and create more of a virtual network where people from both underserved populations and geographically dispersed populations would be able to access care virtually with their primary care doctor and their cardiologist [and then communicate] with these valve centers via telemedicine,” Otto suggested. “And then patients would move to higher levels of care only when they need it.”