The federal government wants to know how outcomes data that hospitals and doctors now submit to specialty society registries might be folded into Medicare performance measures that assess quality of care.
That's what the Centers for Medicare & Medicaid Services asked in its February "request for information" about such registries, and whether the agency should approve such clinical data registries as being "qualified," ultimately satisfying federal requirements under its Physician Quality Reporting System, or the Electronic Health Record (EHR) Incentive Program.
To be such a federally recognized registry, what requirements would a professional organization have to meet?
These voluntary reporting registries, a growing trend in specialty care especially for surgery and other invasive procedures, amass and risk-adjust, sometimes audit, and usually report back results to participating doctors and hospitals so they can benchmark their own performance.
"What would be the benefits and shortcomings involved with allowing third-party entities to report quality data to CMS on behalf of physicians and other eligible professionals?" CMS asked in one part of its request.
Specialties that operate or are starting such registries range from cardiology and cardiothoracic surgery, to orthopedics and neurosurgery.
In about 100 responses submitted by the April 8 deadline, many of those organizations expressed concerns about how this would all play out.
"We think that if CMS allows registries to serve as physician quality reporting efforts, the agency should make sure the data is accurate and rigorously collected," says Clifford Ko, MD, director of the American College of Surgeons National Surgical Quality Improvement Program, or NSQIP, which now has 700 hospitals participating in a variety of outcomes registries for surgery.
Already, CMS is working with NSQIP, and posts rates of adverse events in surgery, such as infections, blood clots, stroke, and pneumonia, for 79 hospitals that volunteered this information, on Hospital Compare.
Just collecting and feeding the data back to participating physicians and hospitals, however, "is not enough. These registries must be integrated into a quality improvement system or program, a lesson we learned early," says Ko.
"When we just sent back the data, it didn't really help anyone get better. We had to then develop the second arm of NSQIP, to help teams improve communication and culture, and integrate clinical guidelines for people who may believe in them or may not believe in them."
Ko says that without an improvement component to these registries, hospitals are left to figure it out for themselves. "That's difficult for them to do, and they don't do it efficiently."
Eventually, the system could evolve to a point where it is registry outcomes data, risk-adjusted and publicly reported, that determines major chunks of hospital and physician payment, similar to the way the process measures and patient experience scores now affect value-based purchasing algorithms for hospital reimbursement. "That's what the eventual goal is," Ko says. He concedes it may take years.
Some organizations, like the Society of Thoracic Surgeons, which began such a database 20 years ago and now boasts the largest in the country, are especially enthusiastic about using registries for federal reporting.
"We've been trying to get everyone's attention for years," says STS Secretary, Keith Naunheim, MD. "We thought we have a winning proposal here, and we think it ought to be generalized to specialties as a whole."
In heart surgery, the STS database now holds five million patient records. The registry captures rates of post surgical morbidity, such as stroke, renal failure, infections, prolonged time on a ventilator, and reoperation for any reason; in-hospital mortality; use of mammary artery in bypass procedures rather than a vein; and whether correct medications were given to patients at discharge.
The problem, Naunheim says, is that the STS database only captures what happens inside the hospital. "What's also exceedingly important is what happens when (the patient) steps outside the hospital," he says.
Did the patient get readmitted? Is the patient off angina medications or did they have to go back on? Did they go to rehab? How long did they live — three or five years after surgery? And when you compare a stent patient with a bypass patient, how much did it cost?
"All that information we could only get if we could link our databases with the CMS administrative database," says Naunheim, who also is Chief of Cardiothoracic Surgery Saint Louis University Medical Center.
Ko says several big questions must be answered before CMS comes to a decision. Should CMS receive this data from these registries, and interpret it for the public, or should it leave that up to professional societies, like the ACS or STS, to run the numbers and feed it back to the providers.
"Or do we give them all the data, and they do the analysis?" Ko says.
And at what point will reporting on Hospital Compare begin, so the public and payers can know what now is only known to these societies and their participating providers?
In its letter to CMS, ACS Executive Director David Hoyt, MD, wrote that its four other programs, the Surgeon Specific Registry, its Metabolic Bariatric Surgery Accreditation program, and its National Cancer Data Base, and its Trauma Quality Improvement Program, all keep statistics on millions of cases, including patient outcomes.
The American Hospital Association's Ashley Thompson, Vice President and Deputy Director, wrote that CMS "should develop criteria ensuring registries have the technical capabilities needed to report quality data. Moreover, those registries must be transparent in their methodologies and able to provide timely feedback to providers on performance."
Other organizations responding to CMS' request include major professional groups such as the American College of Cardiology, which also has an expansive quality reporting database for stent procedures; American Academy of Orthopedic Surgeons; the American College of Physicians, the American Association of Neurological Surgeons, which began a smaller database last year; the American Board of Medical Specialties, the American Academy of Ophthalmology, and the American Medical Association.
The American College of Cardiology's National Cardiovascular Data Registry, which began in 1997 to monitor quality of cardiac catheterization procedures, now "captures data on the majority of percutaneous coronary intervention procedures performed in this country," said the letter by ACC President John Gordon Harold, MD. It now keeps seven registries, all for different procedures including carotid endarterectomies and implantable cardioverter defibrillators.
Large hospital organizations such as Kaiser Permante, the Cleveland Clinic, and Gundersen Lutheran Health System also sent in responses about how they view any change in the current quality reporting system.
"At the hospital level Cleveland Clinic participates in a significant number of clinical registries," wrote Robert S. White, MD, the clinic's Associate Chief Medical Information Officer.
"This included STS, ACC, NSQIP, GWTG (the American Heart Association's Get With The Guidelines) and Cerner Critical Outcomes. In addition to these registries, Cleveland Clinic participates in specialty registries coordinated by the specialty boards. These registries currently may not be physician specific currently, but can be altered to meet the reporting needs of physicians."
Within orthopedics, AAOS President, John Tongue, MD, wrote "a growing number of AAOS members participate in the American Joint Replacement Registry, which collects data that would be useful in quality reporting."
His organization has concerns, however, saying, "the relationship between the current generation of quality measures and actual patient outcomes has not been established. Thus, AAOS does not believe existing quality measures will make a meaningful contribution to the public domain of provider performance data, and that misleading information could be used in medical liability actions."
Tongue added, however, that "any plan to link performance measures to Medicare payments be created and endorsed by the medical specialty associations and their members, to fully meet the requirement of being meaningful in a clinical context."
Naunheim, of STS, emphasizes that for most other specialties, and even for some parts of cardiothoracic surgery, the field of quality measurement is still young. "There are very few specialties that do this on a regular basis, but that's the coming trend. That's what the public is going to demand; it's what CMS wants and what third party payers want. We're just one of the first groups to start putting our results out there."
The STS now posts one, two or three-star scores for medical groups and hospitals that do open heart cases and agreed to public voluntary reporting.
Might some professional societies turn away, hoping the drive for more database reporting will quietly disappear? Naunheim says there are some who "aren't happy about it. But to be brutally honest about it, we weren't happy about it either. We started ours when New York State published raw mortality data for heart surgery that wasn't risk adjusted."
It was, he said, "a kick in the pants."
So they developed their own risk adjustment model and started collecting the data themselves. "It woke up cardiothoracic surgery as a specialty, because we said to ourselves, if we don't do it, they're going to do it, and besides, it's within our professional responsibility to do it."