Who's setting the bar–and where is it?

“What's going to happen, ultimately, is that Medicare is going to set the quality metrics,” predicts Minda Wilson, founder of Affordable Healthcare Review, “and providers will have to follow them. The problem is that Medicare hasn't figured out what the right metrics are. Worse, they might not figure them out until long after new quality standards have been adopted.”

And some commercial payers are developing proprietary quality metrics that ultimately could challenge Medicare's authority in the industry.

“There are three ways insurers are determining what quality measures to focus on,” says Ray Desrochers, executive vice president at HealthEdge. “One is following what CMS is doing. Another is following other groups, like [America's Health Insurance Plans], that have been making a number of recommendations. And third—and this is most important—is that payers are doing research on their own and figuring out the right ways for value-based payment to work for their own organizations. Most payers are not going to freely share this information with others, as they believe that their specific approaches could provide a competitive advantage.”

In fact, exactly how these quality metrics are determined has become an area of much interest in the health care arena because alignment and standardization—the buzzwords that represent solid, evidence-based benchmarking information and the ability to collect and share associated real-life data without corruption—are vital for creating effective quality metrics, which in turn are crucial for value-based payment methods.

“Payers might develop quality metrics based on, for instance, performance levels determined nationally by top-tier providers, so they're setting the benchmark very high,” explains Matt Amodeo, a benefits attorney at Drinker, Biddle & Reath. “A commercial payer could look at the highest-scoring, highest-quality cardiology group in the country and use that to set their benchmark for the cardiology providers participating in that commercial payer's [accountable care organization]—and those metrics might look nothing like Medicare's metric.”

Some of the issues go beyond metric development to touch on larger problems of technological infrastructure and data analysis.

“If you're going to have a deal between a health plan and a provider and it's partially or wholly based on outcomes and quality measures, someone needs to keep track of those measures,” says Joe Kuehn, a partner in the health care practice at KPMG.

“So one of the issues for all is not just the availability of the data, with definitions and measurements that everyone agrees on, but having the technology in place and the processes in place to capture it and translate it into real-time information you can use to manage performance over the long haul. If you don't have that data and analytic capability upfront, entering into at-risk contracts could prove disastrous.”

Additionally, all of these assorted metrics to track for each payer has resulted in understandable frustration from the health care provider community, which argues that the administrative burden generated from tracking and analyzing so many different quality metrics is unreasonable.

“Most hospitals over the past five years have had to create almost entire departments with nurses who would collect this quality metric data,” says Dereesa Reid, chief executive officer of the Hoag Orthopedic Institute. “As we get deeper into the electronic age, we begin to be able to mine this data electronically. But it is an onerous task.”

Standard is as standard does

“Individual physicians, group practice managers and hospitals have a legitimate complaint about the proliferation of measures and the lack of standardization,” says Bill Kramer, executive director for national health policy at the Pacific Business Group on Health, who's also on the board of the National Quality Forum, an group created to generate standardized measures for performance, quality, appropriateness of care and other areas of health care, which solicits input from physicians, purchasers, employer groups, consumers and other health care stakeholders.

“I will say that the NQF has accomplished much,” he says, “but there's still much to be done to bring greater standardization to health care measures. Before NQF, it was totally the Wild West. The federal government also should receive credit for standardizing the measures it uses through CMS. Many commercial insurers are using their own measures that are going to be similar to but not the same as the government's standardized measures. And so far, we haven't been able to find a way to align all of the payers and make a dent in this proliferation of non-standardized measures.”

“One of the things we don't have yet in the United States that we hope someday will be more standardized is procedural registries,” Reid notes. “Some different states…have registries that collect clinical outcomes data about, for example, joint replacement patients. But in other countries, they have nationwide registries, and that data is being collected through standardized quality metrics and outcomes measures. What's advantageous about that is your data isn't in a silo, and you're able to look at a much bigger population.”

The International Consortium of Health Outcomes Measurement is a nonprofit organization that collects health outcomes information that's contributed by patients and physicians across the globe; there are about 40 disease types in the ICHOM registry, and care providers in the United States can participate—in fact, the ICHOM annual conference will be held at Harvard for the second year running—but there are no national quality metrics or health outcomes data registries for U.S. health care providers.

The missing links

The lack of standardization in quality metrics is clearly one problem, but there are others associated with these new data collection and analysis initiatives. And the issues with quality measures aren't always standardization—sometimes, there's a greater problem of omission or inability to quantify return on investment.

“We've made progress, but there are still many things that aren't measured well—or, in some cases, aren't measured at all,” Kramer says. “For example, we don't have a good standardized measure for medication errors. We know it's a problem, but we don't have a good measure to track it.”

Kramer also mentions the lack of patient-reported outcomes: There's no systematic way to gather data that indicates, for example, whether a patient who underwent a surgery to repair an ACL torn on the basketball courts was still able to shoot hoops six months after the surgery.

“Patient-reported outcomes are the next frontier,” he says. “We've made some progress on traditional clinical outcome measures, but we don't have a systematic way of capturing patient-reported outcomes.”

“Around here we talk a lot about the Institute for Healthcare Improvement's 'Triple Aim' initiative—to reduce cost, to improve quality and to improve the patient experience,” says Molly McCarthy, associate vice president of the provider performance network at Priority Health. “And sometimes those activities will result in an ROI, but sometimes those are long-term investments. One example is screening young women for cervical cancer. That benefit is not being realized for 30 years, but it's an evidence-based treatment standard that has great value. However, it might not be value that Priority Health realizes—that patient might not still be with us in 30 years. So some of the ROI is elusive.”

And although many experts agree that initial steps taken by CMS and many commercial payers, such as penalizing hospitals for post-procedural readmissions, some point out a need for more granular analysis of those circumstances.

“If a health system is readmitting a patient right after a patient has been discharged because the care providers didn't perform medication reconciliation, didn't give good instructions or discharged the patient into a situation where they can't manage their health, that's one kind of problem,” notes Scott Wallace, visiting professor of family and community medicine at Dartmouth's Geisl School of Medicine. “But the question is, why are patients being readmitted? If we can differentiate between readmissions that result from mistakes, from failure to discharge the patient as well as we should, and readmissions when something else unrelated to the quality of the care happens, then we've got a better tool to examine health care costs. A more constructive way to structure these metrics is to provide incentives to keep people healthy rather than imposing penalties when people get sick.

“In health care, we tend to roll out big programs out and then tweak them,” he adds. “And it's a lot harder to tweak at scale.”

Tweaking at scale

But payers and providers have been attempting to do that nonetheless, from primary care providers to specialty practices. Reid's organization, Hoag Orthopedic Institute, has been participating in CMS's value-based purchasing program and says the metrics have stayed fairly consistent during the institute's participation.

“Each year they change the metrics a little bit,” she says, “but for us as an orthopedic surgery specialty hospital, the metrics are relatively straightforward because there are two things in most surgeries that you're really monitoring: infection rate and readmission rate. CMS is very straightforward about how you collect the information.

“Now that we have more sophisticated claims information, we know that there is a real cost to an infection,” she continues. “In orthopedics, an infection is a very serious thing, because if a patient gets an infection after a joint replacement, that could potentially result in the whole surgery having to be redone.”

Dr. Allen Nissenson, chief medical officer at DaVita, says that CMS and the dialysis community developed a portfolio over many years of basic quality metrics that apply to kidney patients.

“Most of the commercial insurers have looked to what CMS has developed and are pretty comfortable with those metrics,” he explains. “But we have some challenges with the more generic quality metrics that insurers are used to applying to general population management, because many of those are not comparable with our quality metrics—or are even contraindicated. And those discussions are harder because the health plans typically don't have content experts, and they don't always understand why provider networks are arguing with them about these metrics.”

The setting of standardized quality metrics is very much a work in progress, and many payers are offering resources for providers, as well as a soundboard for feedback.

“We developed our Patient-Centered Medical Home Program in close collaboration with primary care leaders throughout New Jersey,” explains Carl Rathjen, manager of network strategy and program development at Horizon Blue Cross Blue Shield of New Jersey. “In working with them, we also put together a physician advisory panel. So when we want to propose changes to the program, or when providers want to recommend changes to the program, we have a process in place to get physician feedback and have a dialogue about whether we're using the right quality metrics, among other things.”

“Most commercial payers have developed their own proprietary population management programs, and by programs, I mean hardware, software and personnel,” Amodeo says. “They have all the data for the attributed population, and they can more easily distill this data and provide it in reports that are easily digested and understood by their network providers. Sometimes the provider can request certain data elements organized in a specific way, and sometimes the payer can accommodate that and provide ad-hoc reports that are customized. Payers are providing resources.”

“We do offer additional support for our providers,” McCarthy says. “We provided funding to support the development of patient-centered medical homes. We are now providing support for organizations that are building their own care-management capabilities. We also set metrics that are common across other payers so that we're not asking providers to do something that is unique to Priority Health.”

And other organizations, such as the nonprofit FAIR Health, have jumped into the national discussion, too.

“FAIR Health has data in connection with 151 million privately insured lives,” explains Robin Gelburd, president of FAIR Health. “Our data helps inform value-based payment programs and the shaping of quality measures, and one way we do that is to license de-identified utilization data to help stakeholders see what types of wellness programs might be appropriate for different populations. Employers and payers can then use that data to frame wellness programs that emphasize the needs of their targeted populations, and they can see whether the data starts to change in any meaningful way. And payers and third-party administrators can also use the data to design appropriate provider networks. Likewise, they collect charge information, which helps them frame incentive payments so they can shape models that reward providers for meeting certain quality indices.”

The bottom line

Quality metrics are the cornerstone of any value-based payment model, and as the health care industry moves toward value-based payment, it's increasingly vital for benefits professionals to understand the intricacies of these models.

“It would be helpful for brokers to know if there are quality measures inherent in the plans they sell, and what those measures are,” Gelburd says. “For example, are there special features of these health care programs that really facilitate sound care management? What kinds of plan elements are built into these benefit designs to achieve cost savings and bring quality and value?”

“Driving higher quality does save money,” Nissenson says. “So if you focus on the holistic needs of the patient and can find organizations delivering patient-centered care, you're going to find better clinical outcomes and you're going to find lower costs—and that's the sweet spot.”