Ups and Downs of High Volume

Healthcare leaders agree that experience and expertise are essential, but some caution that the development of volume-based thresholds could have unintended effects.

This article first appeared in the December 2015 issue of HealthLeaders magazine.

Peter Pronovost, MD, PhD, FCCM, can pinpoint the moment he knew hospitals needed to focus more attention on the role volume plays in outcomes.

Pronovost, who now serves as director of the Johns Hopkins Armstrong Institute for Patient Safety and Quality and senior vice president for patient safety and quality at Johns Hopkins Medicine in Baltimore, a $7.7 billion integrated global health enterprise, was caring for an elderly woman at another hospital who had had an esophagectomy that went "terribly wrong."

During the patient's final days on a breathing machine, Pronovost asked if she had been told she was at a high risk in having the procedure done there because that hospital had only done one or two of them. To his dismay, the answer was no.

Her death is just one example of why Pronovost says he began to promote volume standards and to collaborate with peers at Dartmouth-Hitchcock Medical Center and the University of Michigan Health System to create the "Take the Volume Pledge" campaign, which was announced earlier this year. All three systems and their hospitals have developed minimums on certain complex surgeries so that only surgeons who do them frequently enough to be proficient would be permitted to perform them. They aim to have the rules in place by year-end.

The pledge lists 10 procedures, including bariatric staple surgery, carotid artery stenting, and hip and knee replacement. "These are the top procedures with the strongest evidence that low volume impacts surgical outcomes," Pronovost says.

With intense pressure on providers and payers to strive for optimal outcomes, volume seems like an obvious metric. The belief is that more equals better, or practice makes perfect. Three decades of data show that volume is a clear indicator of outcomes, Pronovost says.

At Dartmouth-Hitchcock, John Birkmeyer, MD, a surgeon, outcomes researcher, executive vice president for enterprise support services, and chief academic officer, determined that for five common procedures and conditions, "as many as 11,000 deaths nationally might have been prevented from 2010 to 2012 over the three years analyzed if patients who went to the lowest-volume fifth of the hospitals had gone to the highest-volume fifth," according to U.S. News & World Report.

But volume has its downsides, including potentially overburdening surgeons and hospitals that meet the threshold and punishing newer surgeons or surgeons at rural hospitals who do not have the opportunity to perform high volumes of many types of surgeries.

Aware that the threshold campaign has focused on procedures that work best in an academic teaching hospital or center of excellence hospital, Pronovost acknowledges that rural hospitals might not have a deep enough bench of surgeons to meet larger hospital standards, leaving patients with no other options.

"This is a very nuanced area, and any top-down regulation would get it wrong," he says. Therefore, he wants health systems, hospitals, and physicians to develop the limits themselves rather than having a federal regulatory body get involved.

Pronovost's goal was to "start the discussion," and he knows there will have to be accommodations for various factors. In a May HealthLeaders Media online article, Birkmeyer said, "Our intent with these thresholds is not to set such a high bar that every patient must go to one of five places for surgical care. What we're trying to do is minimize the number of patients who wind up getting their care by so-called hobbyists, surgeons and hospitals that seldom do these procedures—certainly not enough to attain a high level of honed proficiency."

Avoiding unintended outcomes
Martin Makary, MD, MPH, surgical director for Johns Hopkins' Multidisciplinary Pancreas Clinic and professor of surgery at the Johns Hopkins School of Medicine, says he agrees that hospitals "need to rein in the small fraction of surgeons that operate outside of their expertise when the option to refer to somebody more skilled is available."

The healthcare industry itself is somewhat to blame, Makary says. "There's an intrinsic financial disincentive in most practices to refer a patient to another surgeon," he says, calling it "a serious problem" that has to be addressed. "Many hospitals don't have a system that rewards teamwork or sharing."

He says the goal of volume standards should be to create a system that directs patients to the right doctor, but he stresses that volume standards have to be done carefully to ensure that "we don't have an unintended outcome from the implementation."

Makary also is concerned about the potential fallout if nonsurgeons create the thresholds. The standards, he says, should be determined by two things: The data that shows when volumes are increased by a certain number that outcomes improve, and the wisdom of practicing surgeons who do that surgery and can speak to how many procedures you would need to perform to be proficient.

For instance, he performs pancreas islet transplants and notes, "If I don't do them frequently, I could get rusty." Practicing surgeons, he says, "have an incredible bird's-eye perspective on proficiency credentials needed to practice that specialty skill set."

In addition to who sets the limits, Makary worries about who will be unnecessarily impacted by those limits, including new surgeons who need the opportunity to learn. He recommends having two surgeons operate together, calling this approach "one of the most underutilized ways to drive improvements in healthcare." Working together is not only a way to achieve volume proficiency but also keeps both surgeons' skills fresh and active, he says.

Surgeons in rural or underserved areas also could suffer in a volume-based surgery system. Some patients won't or shouldn't travel great distances from their support system of friends and family and, therefore, are treated by their local surgeons. "Those surgeons shouldn't be punished because the complication rate is above the mean," Makary says.

Another potential problem: labeling surgeons as low-volume when they do a high volume of surgeries in a comparable skill set. "If someone can repair a ruptured aneurism, they are skilled enough to treat a varicose vein," he says.

Considering the surgical team
Jay Redan, MD, FACS, immediate past president of the Society of Laparoendoscopic Surgeons and medical director of minimally invasive general surgery at Florida Hospital Celebration Health—a 203-licensed-bed facility that is part of the Orlando-based Florida Hospital network, which has 22 campuses and 4,899 licensed beds—says that for volume standards to work, surgeons have to self-police better. Hospitals should do internal audits, which he says would help weed out the low-volume, poor-outcome surgeons.

"We have to stress quality volume, not just doing as many surgeries as you can. There should be a stricter look that we're doing the proper thing for the proper patient," Redan says.

He is adamant, too, that volume should be a reflection of the surgeon and the hospital together. "Certain operations are demanding and require the comprehensive total of the care system, including a good nursing staff, anesthesiologist, and more. The patient won't survive if the whole health system isn't working together."

Redan gives the example of a doctor who performed minimally invasive hysterectomies at a small community hospital and became so popular he was asked to do them elsewhere. But when he went to other hospitals, the surgeon didn't have the same outcomes. Surgeons, he says, are a package with the hospital's surgical teams.

Currently, Florida Hospital Celebration Health is applying volume standards to a handful of procedures as part of an upcoming deal with a well-known local company, which wants to partner directly with the health system for employee care. This move is forcing specialty teams to identify their top-performing surgeons.

The downside of volume metrics
While volume standards can root out poor-performing surgeons, they also have the potential to burn out the good ones and cause care delays, according to Robert Wachter, MD, professor and interim chairman of the Department of Medicine at University of California San Francisco School of Medicine, which has an operating budget of $2 billion.

"The volume outcome curve is not relentlessly positive," he says. "There is a point above which you have fatigue, overload, or access problems."

Volume can flood a surgeon, practice, or hospital pipeline, leading to increased surgical loads or patients waiting far too long for their procedures. Also, if insurance companies begin to use volume as a standard metric, patients could run into problems. "Certain communities or certain specialties might not have enough doctors," he says. If the more experienced person has a three-month delay, a patient might have to choose to go with the less experienced one.

Wachter says volume is a temporary metric that will go away once big data analytics matures. "Volume is easy to measure; clinical measures are harder to quantify," he says. "The science of clinical measures is getting better quickly as clinical data on every patient is being entered into electronic medical records."

He predicts that in the not-too-distant future, healthcare systems will be able to analyze which surgeon is best matched to a patient. "We'll see that surgeon X has a significantly lower mortality rate and readmission rate, even when readjusted for patient-related factors," he says. "You can't do this with paper records and bills."

For Johns Hopkins' Makary, the volume discussion is a good first step in addressing what he sees as the much larger problem: credentialing. "The credentialing system is deeply flawed. You take a written exam and you're OK to practice a skill set with no coaching, training, or monitoring for the rest of your career," he says. "If done correctly, volume standards can bring the credential issue to the forefront."

Reprint HLR1215-6