In the new and constantly shifting healthcare landscape, one in which policymakers are funneling hospitals, physician groups, and integrated health systems into a world in which they will need to take on more risk for their patient populations, the pressure is mounting on providers to deliver better outcomes.
Indeed, over the past several months, federal health administrators have made clear that the healthcare system, in both its current and projected cost state, is not sustainable, as Medicare actuaries have predicted that U.S. healthcare spending will leap from $3.5 trillion in 2017 to $5.7 trillion in 2026—a 62.8-percent growth over nine years. But at the same time, they have also stressed the need to reduce the burden that quality measures put on providers, since industry observers believe that many of these government-oriented measures have little value to clinicians or patients. As such, moving forward, federal leaders must strike a balance between being able to measure outcomes while not overburdening providers even more.
One way in which the federal Centers for Medicare & Medicaid Services (CMS) will aim to do this is through its new “Meaningful Measures” initiative. Announced last October, the new approach to quality measurement “will involve only assessing those core issues that are most vital to providing high-quality care and improving patient outcomes. The agency aims to focus on outcome-based measures going forward, as opposed to trying to micromanage processes,” CMS stated last fall. “We need to move from fee-for-service to a system that pays for value and quality—but how we define value and quality today is a problem,” CMS Administrator Seema Verma stated at the time. “We all know it: Clinicians and hospitals have to report an array of measures to different payers. There are many steps involved in submitting them, taking time away from patients. Moreover, it’s not clear whether all of these measures are actually improving patient care.”
As Verma alluded to, the lack of quality measure alignment across different value-based purchasing programs can be quite frustrating for providers, as a payer in one reporting program might have a different quality requirement than a payer in another. And even in the same program, such as the Quality Payment Program (QPP), under the MACRA (Medicare Access and CHIP Reauthorization Act of 2015) law, a lack of measure alignment creates complexities.
For instance, Kate Goodrich, M.D., CMS' director of the Center for Clinical Standards and Quality and the agency's chief medical officer, said at a meeting earlier this year in Washington, D.C, as reported by the Healthcare Financial Management Association (HFMA), that the lack of that quality measurement alignment has hindered quality reporting during the first years of the MACRA law. “The measures are basically the same, but what people have to do—the rules of the road, if you will—on the scoring are very different between the two,” Goodrich said, referring to hospitals and their employed physicians. “And that creates problems for health systems that use a single [electronic health record (EHR)] to report on behalf of clinicians and to report on behalf of hospitals,” she said, as reported by HFMA.
The notion of harmonization and alignment across quality measures is hardly a new concept, asserts Jeff Smith, vice president of public policy for the Bethesda, Md.-based AMIA (the American Medical Informatics Association). “This is [something] that has been out there for a decade or even more. So, the real question is, how do you get alignment on those measures?” he asks.
Smith believes that quality measures are a microcosm for a much larger issue inside medicine—defining what value is among different physicians. “Let’s say you are asking two different cardiologists what they value. The 80/20 rule is solid in this space; 80 percent of cardiologists would agree on a certain small set of measures that they see as important for most of their patients. But then going from cardiology to another specialty, or certainly in family medicine, measures matter to different people for different reasons. And we have been on this 10-year journey, especially when you think about the electronic environment that is everywhere now, so it’s a large task to figure out if there are measures that can be meaningful across specialties and settings,” Smith says.
Payers and Providers Working Together
Most industry stakeholders believe that payers and providers need to be on the same page when it comes to agreeing on a common set of measures, but to date, this has been tough to achieve. One reason why is that a lack of shared financial risk between payers and providers could lead to more misalignment. For instance, if a payer is assuming all the financial risk in a given value-based contract, its goals will be different than the provider that is in that agreement.
Charles Saunders, M.D., CEO of Florida-based healthcare technology company Integra Connect, and former clinician and executive at Aetna, says that while payers and providers are improving in their willingness to align incentives, it does vary, depending on the payer. “I was with Aetna for five years, and we took a collaborative approach, always looking for a win-win,” he says. “The wrong way to go about this, from a payer perspective, is if you have a shared savings program [such as CMS’ Medicare Shared Savings Program ACO model] and make it so great that it benefits you while the provider will never generate savings unless they do something heroic. That’s not a win-win,” Saunders attests, noting that 75 percent of ACOs (accountable care organizations) in the MSSP model do not generate shared savings. And then, “Cynicism gets generated and it comes across as adversarial with the payer. But then there are those enlightened payers that are taking the more collaborative and proactive approach,” he says.
One example of a payer being proactive and collaborative is in Hawaii, where the largest insurer in the state, Blue Cross and Blue Shield (BCBS) of Hawaii, works very closely with all of the state hospitals to identify where there are quality and safety opportunities for improvement, and then looking individually at each one of those areas, explains Madeleine Biondolillo, M.D., vice president of quality and safety at the Charlotte, N.C.-based Premier, Inc.
Madeleine Biondolillo, M.D.
Biondolillo says that BCBS of Hawaii supports hospitals’ participation in Premier’s QUEST collaborative—a quality improvement initiative that aims to help health systems provide high-value care—by specifically working with them to develop very precise compensation structures to address exact areas that they want to see improvement in. “They are working together and they get the support that is needed. It has to be a collaboration between payers and providers,” Biondolillo says. [BCBS of Hawaii] pulls all the levers they can to provide that support through the hospital’s participation in the quality improvement work, for example. They look closely at the outcomes measures and track them over time,” she adds.
Nonetheless, Biondolillo readily admits that this type of partnership and collaboration is not happening as quickly as many might hope, a sentiment that Mary Barton, M.D., vice president for performance measurement at NCQA (the National Committee for Quality Assurance) agrees with. Barton, a general internist who has been working at NCQA for six years, notes that there are fundamental issues to overcome in regard to aligning incentives for outcomes measures.
For instance, she offers, an orthopedic surgeon who is being measured on a 30-day post-hip surgery recovery provides an easy way to get an outcome measure into a physician-level program. But with diabetes, on the other hand, Barton says that measuring care for diabetes could be akin to “watching grass grow.” She notes, “Things don’t happen very quickly. So, when someone has a bad long-term effect and goes blind based on decades of poor sugar control, who will you put the outcome on? Will you blame the doctor taking care of that patient at the time he or she turns blind, or will you go back and figure out how to hold all the other doctors in the patients’ past responsible for that?”
Why Analytics and Data Flow are Mission-Critical
Recent case studies have found that health systems that have adopted a systematic approach to using data and analytics to guide their improvement initiatives have documented major improvements in quality, cost and efficiency. Research from Salt Lake City, Utah-based data and analytics company Health Catalyst found that healthcare organizations—ranging in size from small community hospitals to large integrated delivery systems—using the company’s technology and improvement processes experienced significant improvement in key clinical outcomes such as reduced mortality, complication rates, readmissions, infections, and length of stay.
“The data analytics are critical. The more sophistication around that that can be brought to bear, the better,” Biondolillo asserts. Saunders agrees, noting that about 5 percent of patients drive 45 percent of costs. “So, you need to identify that 5 percent up front,” he says. For instance, in the case of CMS’ Oncology Care Model (OCM), in which participating physicians receive a bundled payment for providing chemotherapy services to cancer patients, “You don’t have unlimited resources to expend on every patient, so you would like to focus on engaging those patients who are likely to incur those costs, risks and bad outcomes. You can use predictive analytics to take their claims history and look at simple things like age, gender, cancer diagnosis, chemotherapy choice that they are on, and whether they have been admitted in the past 30 days. There are a variety of things to look at to predict risk, stratify those people, and engage them with case management,” Saunders says.
Another metric that Saunders points out is the number of cancer patients who died within less than three days of entering hospice care. Predictive analytics can be leveraged to determine who is at very high risk of dying in the next 90 days, and then from there, significant health impacts can be realized, such as making sure that those patients have palliative care lined up as well as providing psychological and social behavior support, he says.
Charles Saunders, M.D.
“The richer the analytics, the better,” Saunders attests. “If it’s just claims-based and based on the provider’s 835 file, that doesn’t contain the diagnosis. But if it’s based on the 837 payer claim, it does contain the diagnosis. CMS claims are good claims, but they come three to six months late; you can get commercial payer claims every 30 days,” he says. What’s more, access to EHR data is also rich since that can be obtained in real-time, “allowing you to identify diagnoses, comorbidities, and prescribed therapies patients are on,” which then “enables you to predict things such as risk of treatment failure, complications with chemotherapy, ER visits, hospitalizations, and depression. And then you can jump on that early to make an impact,” Saunders says.
Saunders also believes that for effective transitions of care, if a hospital has an HL7 ADT (Admit, Discharge and Transfer) feed, it can plan for discharge two to three days before it happens and then make sure that when patients arrive at their destination, they get proper medication reconciliation and home health services, if needed. Saunders contends that when hospitals leverage predictive analytics and share data, they are seeing their avoidable readmission rates—one of the top outcome measures tracked by CMS—drop by as much as 50 percent.
As NCQA’s Barton points out, “Seeking better flow and more timely flow of information is crucial to having the health system do better for patients,” and that could mean that the payer has to let the clinician know when a patient has been hospitalized. “If you ask clinicians, they say that they don’t know when their patient has been hospitalized. But the payer knows; the hospital tells them quickly. So why not set up information sharing that makes it easier for clinicians to look after that patient once they are discharged? If I have a primary care physician taking care of me, I’d much rather they be in charge of my reintegration into the community than have some hospital haltingly and stumblingly reinsert me into the community,” Barton says.
Mary Barton, M.D.
This is where incorporating social determinants of health comes into play as well, says Biondolillo, who recommends drilling down into the data and looking at factors such as one’s functional ability, one’s caregiver support, and one’s social determinants of health. It’s what she calls a “whole person approach.” She adds, “You and I might have both heart failure and you might be at vastly different risk for readmission and/or mortality simply because you are much more connected than me in society.”
In the end, the key question going forward that providers, payers and others will have to grapple with, is how to make measurement work that goes along with the value-based payment arrangements that will encompass healthcare’s future. “I am confident that as quality measurement moves into the next stage, it is going to be markedly less burdensome to clinicians,” says Barton. “We have to somehow jumpstart those [payer-provider] relationships being built so that the data will be shared. That’s the answer to this burden question. It’s actually bigger than alleviating burden—our mission is that measurement be used in real time to help clinicians improve quality,” she says.
Nonetheless, providers would still like to be more at the table when it comes to making these decisions, Biondolillo says. “It is super challenging to have so many measures, and more so than measuring what matters, which is important, [we need] alignment between measurement specification.” But, she adds, “One does not perform differently for different patients based on how the outcomes measurement system is constructed. Physicians aim to do the right thing every time. You can’t function any other way and still sleep at night.”