Kenneth Cohen, MD, FACP, executive director of clinical research at UnitedHealth Group Research and Development and senior national medical director at OptumCare, discusses areas where low-value care is more prevalent and the shift to high-value care.
Areas of low-value care can be identified in any specialty, said Kenneth Cohen, MD, FACP, executive director of clinical research at UnitedHealth Group Research and Development and senior national medical director at OptumCare.
Transcript
Are there certain diseases where low-value care is a more prevalent issue?
There's low-value care across the health care system, so you can pick any specialty and identify areas of low-value care. I think [there are] probably 3 areas that I would focus on—and there are more—but one is spine surgery. For example, between 2002 and 2007, the rate of lumbar fusion in this country went up 15-fold. We now have 4 large randomized control trials that show in the average patient that decompression alone is the equivalent of decompression plus fusion. But fusion surgery costs about 4 times as much, so just a huge, huge area of wasted care and other areas of spine management as well. Now there's no doubt that epidural steroid injections help the majority of people who receive them. There is very little evidence that spinal cord stimulators help people. So, all sorts of expensive invasive things we do for spine don't have a strong evidence base of support.
Cardiology is another large one. Whereas cardiac catheterizations and stents have been absolutely lifesaving in patients that are having heart attacks, for the management of stable coronary disease, they have not been found to be any more effective than medical therapy despite the fact that we've looked at this in 4 large randomized control trials. So a huge number of heart catheterizations, nuclear stress tests, and angioplasties and stents are put in people with stable coronary disease who would do just as well with medical management.
Another large area is in our cancer screenings. This can be colonoscopy, it can be mammography, it can be prostate cancer screening—we don't target those screenings. For example, there's no evidence that [prostate-specific antigen] screening helps men over the age of 70. There is no evidence that the little tiny polyps that a lot of people have increase risk of colon cancer, and yet those patients wind up on more frequent colonoscopies. There is not a lot of evidence that our mammogram program improves breast cancer survival any more than anybody else's around the world, except ours cost 2 to 3 times as much. So there are a lot of things that we do on the cancer screening side that aren't really targeted to the individual and wind up being wasteful.
On the reverse side, reducing low-value care requires promoting high-value care. What is needed to ensure this is done?
Shifting from low-value to high-value care is critical and is a natural extension of addressing low-value care. If you think about some of the things that we've talked about, the use of evidence-based medicine is specifically designed to drive high-value care. The use of clinical algorithms is specifically designed to drive high-value care. When you begin to share data transparently and shine a bright light on what can be considered aberrant practice patterns, that transparent sharing is a potent change agent, and by itself drives the adoption of high-value care.
Laundromats as a New Frontier in Community Health, Medicaid Outreach
May 29th 2025Lindsey Leininger, PhD, and Allister Chang, MPA, highlight the potential of laundromats as accessible, community-based settings to support Medicaid outreach, foster trust, and connect families with essential health and social services.
Listen
New Insights Into Meth-Associated PAH Care Gaps: Anjali Vaidya, MD, on Closing the Divide
June 4th 2025Research from Anjali Vaidya, MD, FACC, FASE, FACP, Temple University Hospital, reveals critical care gaps for patients with methamphetamine-associated pulmonary arterial hypertension (PAH), emphasizing the need for early diagnosis and integrated support.
Read More