In theory, vertical integration in health care makes sense: when physicians work directly for hospitals, rather than in independent practices, there should be greater efficiencies through economies of scale, and better quality of care for patients through coordination and information sharing. And indeed, the healthcare system in the United States has been headed in that direction for years now, consolidating at a rapid rate. The number of doctors who have gone from working in independent practices to working in hospitals has doubled in the past decade and that trend continues.
But a new study by a group of researchers including Harvard Kennedy School Associate Professor of Public Policy Soroush Saghafian finds that in one representative area of medicine, vertical integration is leading doctors to change the way they approach patient care, with consequent adverse effects on patient health, and is also inflating costs. The problem, they conclude, lies in a system of financial reimbursements that incentivizes the wrong behavior—and addressing that issue may offer a solution.
The study was conducted by Saghafian with researchers from Harvard University, Harvard Medical School, Massachusetts General Hospital, and University College London.
Zooming in on a particular specialty and a particular procedure—colonoscopies performed by gastroenterologists in the fee-for-service Medicare program—they looked at millions of randomly selected doctor-patient interactions across a broad geographic spread and across a number of years (2008-1015). They also looked at physicians working in vertically integrated hospitals as well as independent practices.
"Analyzing more than 2.6 million patient visits, we found that physicians significantly alter their care process after they vertically integrate,” Saghafian said. “What is more, this results in substantial increase in patients’ post-procedure complications. Using our findings, we offer some levers for policymakers to mitigate these consequences of vertical integration—a fast-growing trend in the healthcare sector."
The study found that when independent physicians integrated with a hospital, they changed their care practices (for example, by reducing the number of patients they put under deep sedation) and increased their throughput (measured by the number of patients they treated). Specifically, the integrated physicians reduced their use of deep sedation by about 3.7 patients for every 100 treated. However, patients of integrated physicians experienced “a significant increase in both major post-colonoscopy complications such as bleeding (3.8 per 1,000 colonoscopies) and other complications such as cardiac or nonserious GI symptoms (5.0 and 3.3 per 1,000 colonoscopies, respectively).”
The researchers found that the reduced use of deep sedation “at least partially explains the increase in adverse outcomes” and that it was “driven mainly by hospitals no longer allocating expensive anesthesiologists to relatively unprofitable colonoscopy procedures.”
Moreover, integration increased the number of patients a physician was able to treat and elevated reimbursement per procedure—integrated doctors were reimbursed about $127 more per colonoscopy procedure than independent doctors, or about 48% more.
The fee differentials between provider settings are well known. For example, in the case of colonoscopies, Medicare reimburses $917 on average for procedures in hospital outpatient departments, but only $413 for those in physician offices. And these differentials have been criticized for helping to push hospitals toward integration. But policymakers have so far lacked evidence on how vertical integration can affect care delivery.
In addition to providing important evidence on the adverse effects of vertical integration on care, the study also recommends that policymakers correct the incentive system currently in place by paying a bit more for deep sedation. Counterintuitively, this additional expense would result in improvements to both patient well-being and the volume of patients cared for, and therefore would provide notable savings in the long term.
The results, the researchers conclude, “provide a cautionary message that when physicians financially integrate in response to these policies that use financial incentives, it does not guarantee that integrated practices will achieve superior patient outcomes. To achieve superior patient outcomes, there should be additional measures to (a) monitor the post-integration physician behavior and quality, and (b) align post-integration financial incentives.” Given the rapid speed at which the healthcare industry is being vertically integrated, the study urges policymakers to take quick but suitable actions.
Photograph by Friso Gentsch/picture-alliance/dpa/AP Images