Eric Schneider, M.D., M.Sc., executive vice president of the National Committee for Quality Assurance (NCQA) Quality Measurement and Research Group, recently spoke with Healthcare Innovation about the significance of the Centers for Medicare & Medicaid Services’ “Universal Foundation” of quality measures.
The Universal Foundation is intended to focus providers’ attention on measures that are meaningful for the health of broad segments of the population, reduce provider burden by streamlining and aligning measures, and advance health equity. CMS executives noted that the streamlined measure set would be used across CMS programs and populations, to the extent that they are applicable and in keeping with legislative statutes.
The nonprofit NCQA works to improve healthcare quality through the administration of evidence-based standards, measures, programs, and accreditation. Among other things, Schneider leads NCQA’s effort to advance the move to digital quality measurement.
Schneider said he has been working on quality measures since the 1990s and that he has found that there tends to be a cyclical nature to measure profusion and streamlining.
Initially there weren’t enough measures for employers to compare health plans or for pay-for-performance programs to measure clinical quality, he said. The AMA got into the game in the mid-2000s, and created a bunch of measures so CMS could have quality measures for all the pay-for-performance programs it was rolling out in those years. Then there were too many measures and a loss of standardization, he said. “Long story short, it's like a cycle of expansion — some of it is innovation, some of it is new capability or new measures that get developed. Then measures need to be retired, and NCQA has a history of looking at its own set and retiring measures.”
Creating alternative payment models for different specialties also led to a proliferation of new measures. “In addition, the federal government has its own version of this, because each agency is developing measures for its own purposes, whether it's HRSA or CDC, and it leads to that sort of proliferation,” Schneider said.
Schneider called the development of the Universal Foundation quite remarkable. “I wouldn't have imagined that the centers across CMS could really get on the same page around a Universal Foundation measure set. So that was quite an achievement. That's very hard work to do to get that level of alignment.”
CMS is caught between two competing objectives, Schneider explained. One is the health of the population that they serve; the other is the set of providers, hospitals and other organizations, such as Medicare Advantage plans, that they have to administer in some way, and the payment models that they're using. “As a payer, they need to do the value-based contracting around that, but as a population health manager, they have to use a different lens, and I think the Universal Foundation tries to marry those two pretty effectively.”
The fact that CMS came out with this statement is a strong signal because they also have an implementation arm, Schneider added. “They're actually now going to change their programs in terms of what's required reporting, and it helps to put some alignment in place. The other thing that was gratifying for us at NCQA is that 70 percent of the measures in the Universal Foundation set are HEDIS measures. That's really a tribute to the consensus process and the careful evaluation of the importance of the measures and also the feasibility to implement those measures.”
Five of the measures in the Universal Foundation will be part of NCQA’s Digital Content Services Early Adopter Program. I asked Schneider to talk about the importance of digital quality measure reporting and details of the Early Adopter Program.
This is something he has been working on for a long time, he said. In 1999, he published a paper in JAMA called “The Digital Future of Performance Measurement.” “We were pretty far ahead of our time. In fact, so far that none of the technology existed” he said with a laugh. But the digital ecosystem has really evolved since then, with more exchange of standardized digital data across health systems. “We're on the verge of having the thing we proposed back in the 1990s, which is no more manual record reviews or sending a document that a human being has to interpret,” he said. We are moving to support computing the measure directly without a lot of human interaction, he added. “The more you can make it computable, the more likely it is that you'll get accurate responses and reduce the cost of this whole thing dramatically. It is orders of magnitude less expensive.”
Part of the promise of the Digital Content Services Early Adopter Program is that NCQA has taken several of its existing measures through a process of making them computable, providing the code and a Clinical Quality Language using FHIR health data standards, and putting that into software that can actually be run by participating organizations, Schneider said. “We just came through a relatively small pilot proof of concept. Now we're recruiting organizations to participate as early adopters in a learning phase to determine what needs to be tweaked in order for this to be successful,” he added. “We are really excited because there is a principle in innovation called the hockey stick effect, where there is a long flat period followed by a sharp upturn. I feel like we're probably getting toward that hockey stick upswing of things coming into place to make this operational.”
I mentioned that at conferences such as NAACOS, people have presented on being pioneers with electronic clinical quality measure reporting, and they have found it quite difficult.
Schneider said he was just involved in an interoperability roundtable with ACO participants at the ViVe meeting in Nashville. He said one ACO leader said it took them six months to do the first eCQM and six weeks to do the second one, so there is a learning curve. Also, he said, ACOs are relatively newly formed and often involve systems that have different EHRs. They might be managing seven to 10 different systems. The FHIR data exchange model is meant to solve that problem, but there's still a set of operational steps that have to happen before they are smoothly exchanging data internally within the ACO, Schneider said. “I think that's been a struggle for NCQA, too. We've spent years developing patient-reported outcome measures for depression. if you go too early, the infrastructure is really not there to support it. And then people have to create workarounds and manual inputs and it is just way more effort than it seems to be worth when it's at that stage, whereas as the data and technology infrastructure and architecture solidifies, it becomes a lot easier.”
CMS has put a big emphasis on health equity and said that part of the Universal Foundation's goal is to advance equity with the use of these measures. I asked Schneider if NCQA’s work is aligned with them on that front, too.
“In terms of social needs screening and intervention, we have a measure that's in the field,” he replied. “We actually think it's important for intervention to be linked to the screening because simply screening and reporting how many people have a need doesn't really solve the need. It is a step that creates some transparency, but we also have concerns about how that data will be recorded and whether people will be honest. There are a lot of difficulties in collecting some of the information — about interpersonal violence, for example. It's not something you just ask on a survey and you're done. It needs a high level of trust. But I think the concept is right and we're moving in the right direction. We looked across the field and there are probably a dozen or more different screening instruments that organizations are using. We got codes for those so that they can be represented in electronic data systems, and will start to create some standardization around the social needs screening and intervention measures. We expect we'll be able to align on digital definitions as well. There's an effort called Sync for Social Needs where the three quality organizations —National Quality Forum, Joint Commission and NCQA — have come together to create better alignment on a digital standard.”