To Boost Access and Outcomes With New PCI Centers, Location Matters
Regions already at high capacity for PCI services saw no patient benefits, with a decrease in admissions to high-volume centers.
Expanding the availability of PCI-capable centers may improve access to care for some acute MI patients, as well as potentially decrease mortality, but the benefits are only realized in areas with unmet demand, a new study suggests.
“Previous research has found evidence of systematic duplication of new PCI facilities, and our findings extend to show that these new facilities in high-capacity markets provide little to no benefit in terms of patient outcomes,” said senior study author Renee Y. Hsia, MD, MSc (University of California, San Francisco).
The study looked at PCI-capable centers that opened, as well as those that closed, in communities that already had high PCI capacity at baseline compared with those with average PCI capacity at baseline.
Over the course of the study, there was a 5.5% relative increase in same-day revascularization and a 7.6% relative increase in the likelihood of in-hospital revascularization in average-capacity markets where new PCI centers opened. There was also a 2.5% decrease in mortality across all time points, and a 10.4% relative increase in admission to high-volume PCI hospitals, in these average-capacity areas. In high-capacity markets, however, no such changes were seen.
Previous research has shown that while the addition of new PCI centers is intended to improve access to PCI services for those in the areas of most need, they have a tendency to expand also in areas that already have sufficient PCI services, higher rates of private insurance, wealthier hospitals, and less state regulation of new cardiac catheterization labs, Hsia and colleagues note.
Steven R. Bailey, MD (LSU Health, Shreveport, LA), writing in an accompanying editorial, notes that the opening and closing of hospitals and their impact on acute MI patients is a “novel and narrow metric” that doesn’t shed light on the overall quality of the care patients receive, and in particular whether the trend will have a meaningful impact—good or bad—on patient outcomes.
“Although this is an interesting method, the fact that capacity was based upon all admissions and not either CV admissions or PCI procedures provides uncertainty regarding the level of care for ST-segment elevation myocardial infarction (STEMI) patients,” Bailey writes. “Unfortunately, acute MI care is not based upon simply having a hospital in a close location to patients, and the assumptions used may not be able to assess other acute MI processes related to this question.”
No Mortality Benefits in High-Capacity Areas
For the study, published online last week in JACC: Cardiovascular Interventions, Hsia and colleagues led by Yu-Chu Shen, PhD (Naval Postgraduate School, Monterey, CA), used standardized data from the US census and American Community Surveys to determine geographic location and demographic information for over 2.7 million acute MI patients with Medicare. They also created a database using patient and hospital zip codes to enable analysis based on patients living within a 15-minute drive to a PCI-capable center that performed at least five procedures per year.
High-capacity markets were regions that had a higher percentage of patients admitted to PCI-capable hospitals, rather than hospitals without PCI services, while regions with lower percentages were considered average-capacity markets. By hospital referral region, high-capacity markets were located in regions where the percentage of all patients admitted to a PCI-capable hospital was in the top 25% of all regions. In contrast, average capacity markets were communities with hospital referral regions in the bottom 75% in terms of the percentage of patients admitted to a PCI hospital. High-volume centers were defined as those performing approximately 150 PCIs per year.
Between 2006 and 2018, the availability of PCI-capable hospitals within a 15-minute driving distance expanded by 30% in average-capacity markets and by 9.6% in high-capacity markets. Closures occurred in 11% of high-capacity areas and in 7% of average-capacity areas.
After a new PCI center opened in an average-capacity market or a high-capacity market, patients in average-capacity markets had a 0.4 percentage point reduction in 30-day mortality, a 0.5 percentage point reduction in 90-day mortality, and a 0.8 percentage point reduction in 1-year mortality compared with patients in average-capacity markets who did not have a new PCI center opening, adding up to a 2.5% total decrease. No statistically significant mortality differences were seen in patients in high-capacity markets when a new PCI center opened in their area.
The study also found that while new PCI centers contributed a 2.6% decrease in being admitted to a high-volume PCI facility for acute MI patients living in average-capacity areas, those living in high-capacity areas saw an 11.6 percentage point decrease, which translated to a 20% drop in the likelihood of being admitted to high-volume centers.
Not only are they not driving prices down, but there is this perverse incentive to add capacity when capacity is already over supply. Thomas W. Concannon
Additionally, the proportion of patients admitted to high-volume PCI hospitals in average-capacity markets was 5 percentage points above the national average before a new PCI hospital opened, but decreased to the national average after the opening.
The researchers say the findings “have implications for the quality of acute MI care, providing support for the idea that harmful consequences may result from the preferential adoption of PCI in markets where such services are already saturated.”
“We know a significant portion of hospital revenue is derived from cardiac services, which likely motivates some of the PCI hospital openings and closures observed in our study,” Hsia told TCTMD. She added that future research should address financial and other incentives that may contribute to the imbalanced distribution of cardiac services across communities and suggest potential policy interventions to reduce these disparities.
Ten years ago, a study led by Thomas W. Concannon, PhD (Rand Corporation, Boston, MA), showed that areas of the United States that already had sufficient numbers of PCI hospitals were steadily adding more, amounting to systemic duplication of services in areas without need.
Speaking with TCTMD, Concannon agreed with Hsia that financial incentives and competition have stymied progress in prioritizing equitable distribution of PCI services.
“The assumption has been that competition and markets will drive prices down so far that there will be no incentive to add new capacity. That is not the case,” he said. “Not only are they not driving prices down, but there is this perverse incentive to add capacity when capacity is already over supply.”
Concannon said among the potential solutions are regionalized STEMI networks like North Carolina’s successful RACE network, which allows emergency medical services systems to bypass certain hospitals on the agreement that the follow-up care returns to the community hospital, negating the need to expand PCI services in the area.
Regulatory interventions, specifically price pressure and certificate of need (CON), are other possibilities. CONs invoke mechanisms at the state level by refusing to certify new initiation of PCI services until demand can be established.
“The state of Michigan has a really successful Certificate of Need program in place [for PCI] and Massachusetts had a good policy until a new administration came in and wiped it out at the urging of the Massachusetts Hospital Association,” Concannon noted. “Michigan is a really good example of what can be achieved by restraining diffusion of medical technology.”
Shen Y-C, Krumholz HM, Hsia RY. Do PCI facility openings and closures affect AMI outcomes differently in high vs average-capacity markets? J Am Coll Cardiol Intv. 2023;Epub ahead of print.
Bailey SR. PCI facility openings and closures: if you build it, will they benefit? J Am Coll Cardiol Intv. 2023;Epub ahead of print.
- Shen, Hsia, Bailey, and Concannon report no relevant conflicts of interest.