番茄社区

Outcomes after high-risk elective surgery did not have a clear relationship with a hospital's local competitive landscape, a study using nationally representative Medicare data suggested.

Hospitals in the highest-competition U.S. markets had mixed postsurgical outcomes compared with those in the lowest-competition markets: risk-adjusted 30-day mortality was lower for pancreatectomy (OR 0.93, 95% CI 0.91-0.95), rectal resection (OR 0.92, 95% CI 0.86-0.98), and lung resection (OR 0.88, 95% CI 0.86-0.90); higher for mitral valve repair (OR 1.11, 95% CI 1.07-1.14) and carotid endarterectomy (OR 1.06, 95% CI 1.03-1.09); and not different for open aortic aneurysm repair, bariatric surgery, esophagectomy, knee replacement, and hip replacement.

Trends in 30-day readmissions similarly suggested a mixed bag in terms of the performance of high-competition hospitals -- depending on the exact operation -- relative to the lowest-competition hospitals, reported Andrew Ibrahim, MD, MSc, of the University of Michigan in Ann Arbor, and colleagues in .

It was clear, however, that hospitals in high-competition metropolitan areas were more likely to treat higher-risk patients. Compared with low-competition hospitals, the high-competition counterparts were more likely to have older patients (mean age 74.4 years vs 74.0 years, P<0.001) and people with more comorbidities (67.1% vs 63.9% with two Elixhauser comorbidities, P<0.001) undergoing high-risk elective surgery.

"Taken together, our findings challenge the common assumption that hospital competition may be good for care as it relates to complex surgical procedures," Ibrahim and colleagues concluded. "Because there was no clear association between hospital competition and improved outcomes, policy makers may not rely on hospital competition as a rationale for trying to improve care, especially as it pertains to surgical care."

Care quality aside, patient access and hospital costs have been cited as problems leading to hospital closures and hospital mergers, both of which have surged in recent years and are ongoing. The last hospital group standing in an area may be described as being in an ultra-low-competition market, quantified by Ibrahim's group as having a Herfindahl-Hirschman Index of 8,000 to 10,000.

The lowest-competition markets tended to cluster in the South and were less likely to be teaching hospitals compared with the highest-competition markets (Herfindahl-Hirschman Index of 2,000 or lower). They also had fewer racial minorities undergoing high-risk elective surgery (5.6% vs 17.3%, P<0.001).

However, most hospitals in the U.S. can be defined as low-competition without reaching this extreme level. Revising the threshold for low-competition hospitals down to a Herfindahl-Hirschman Index of 2,500 -- the for antitrust proceedings -- puts 90% of U.S. hospitals in low-competition markets, commented Christopher Childers, MD, PhD, of the University of Texas MD Anderson Cancer Center in Houston, and colleagues.

"In this new normal, efforts moving forward should focus on the steps available to increase quality and reduce cost in the absence of hospital competition," they wrote in an .

"It would be interesting to see whether the analysis would have yielded the same results using DOJ thresholds," Childers and colleagues wrote. "Nevertheless, the authors add important literature on a topic that should be of great interest to policy makers."

"Should we accept the wave of hospital mergers and assume that healthy competition is unnecessary? Not quite. We need more data on outcomes such as length of stay, oncologic control, and quality of life. And there is a growing body of literature supporting the fact that hospital consolidation does not improve quality and, perhaps more importantly, may increase costs," they added.

Ibrahim and team performed a retrospective analysis of outcomes for over 2.2 million Medicare beneficiaries (mean age 74.1 years, 59.3% women) who underwent one of 10 high-risk surgical procedures from 2015 to 2018.

The three highest-volume procedures represented were knee replacement, hip replacement, and open aortic aneurysm repair.

"We selected elective procedures that could potentially be influenced by hospital market competition: specifically, procedures that are elective and that have known variation in outcomes based on the hospital where the procedure is performed," the authors noted.

By their definitions, of the 3,166 hospitals included in the study, 14.8% were deemed to be in high-competition markets and 34.5% in low-competition markets.

Chief among the limitations of the study was its questionable generalizability to other age groups undergoing different surgical procedures.

Comment