Surgical Specialization May Trump Volume as Measure of Quality

An analysis of Medicare data for 25,000 surgeons concludes that specialization may be a stronger of measure of quality than volume.

Strong evidence and common sense support the idea that volume is a measure of surgical competency. 

A group of academic medical centers have even taken the volume pledge, agreeing not to perform high-risk procedures if they don't have the volume. They encourage others to do the same.

But there has been pushback.

Smaller hospitals wonder whether they'll meet volume thresholds. Others say the possible, unintended consequence of volume rules could lead to consolidation of services and limited access.

Now comes a study suggesting that the link between surgical volume and quality may be more complicated than a function of practice-makes-perfect.

An analysis of Medicare data for 25,000 surgeons concludes that "specialization" may be a stronger of measure of quality than volume.  

Related: Limits Urged on Surgeries by Low-Volume Providers

Volume is measured as the number of times a surgeon performs a procedure. Specialization is defined by researchers as the number of times a surgeon has performed a certain procedure, divided by the number of times a physician has performed any surgical procedure. 

While the two measures dovetail, specialization predicts 30-day mortality independent of volume.

John Birkmeyer, MD, of the Dartmouth Institute for Health Policy and Clinical Practice in Lebanon, NH is one of the study authors. He performed much of the research supporting volume as a quality measure, and is a proponent of the volume pledge.

The new findings do not mean specialization should replace volume as a quality measure, says co-author and Harvard University economist Nikhil Sahni. "We are by no means saying that volume should be replaced by this metric. We are saying that this is a nice metric to have alongside it."

One big benefit: Like volume, specialization is simple and inexpensive to measure.

While the researchers used data from Medicare claims, hospitals have access to discharge data from all payers. So they can use their own health information technology systems to measure different surgeon's specialization and share it with referring physicians, Sahni says.

Relative Risk Reduction

The researchers looked at the impact of specialization on 30-day mortality rates for four cardiac procedures and four cancer operations. Measured by relative risk, the ratio of overall risk to the risk of mortality with a specialist, they found that the relative risk reduction from greater specialization ranged from 15% for coronary artery bypass grafting to 46% for valve replacement.

For the four cancer surgeries, the relative risk reduction in mortality ranged from 28% for lung resection to 48% for esophagectomy, removal of all or part of the esophagus.

For the six procedures with "statistically significant relative risk reduction," the absolute risk reduction—the difference between overall risk and risk of mortality with a specialist—ranged from 0.3% for carotid endarterectomy to 2.8% for abdominal aortic aneurysm repair.

The research may offer good news for smaller hospitals that have a hard time reaching volume thresholds.

"You have a critical access hospital in the mountains, and you're only going to see so many heart attacks," Sahni says. "With specialization, because it is independent of volume, it brings up this point: Maybe I just have my surgeons specialize instead of having a cardiovascular department with ten different surgeons. Each of them does whatever (cases) come in."

Since specialization is easy to measure and independent of volume, it could offer a way for smaller hospitals to improve mortality rates. It could also guide hiring decisions, Sahni said.

The BMJ paper is important because volume is an incomplete measure of quality, said David Chang, PhD, MPH, MBA, of the Codman Center for Clinical Effectiveness in Surgery, at Massachusetts General Hospital in Boston. He is one of the authors of an editorial entitled: The Hidden Consequences of the Volume Pledge: "No Patient Left Behind"? and posted pre-publication in Annals of Surgery.

Chang agrees that the paper offer options for smaller hospitals that may suffer under volume restrictions.

"The implication would be, if you are a low volume hospital/surgeon, you may be able to compensate by specializing," he says. "Again, the research is incomplete there."

The research may also be welcomed by providers suffering under the growing number of quality measures, who worry this might just be another one.

Chang says the reason there are so many quality measures is that the measures in use are imprecise. Before adding more incomplete measures, "we should take a step back and really be thinking about how we measure quality."