Cooking Up Some Value-based Care? Here Are Some of the Ingredients You Will Need.

A new report finds that successful programs have a third dimension beyond financial incentives and quality measurement.

By Michael D. Dalzell
Senior Contributing Editor

There are hundreds, if not thousands, of experiments taking place with value-based care. Like CBD preparations, value-based care models are everywhere—with innumerable claims about their effectiveness. But as with CBD, rigorous scientific study has yet to catch up with most of those claims.

It is difficult to draw conclusions from the evidence base. Much of the literature examines programs whose singular focus is either to curtail spending or improve quality, but neither cost-saving incentives nor quality measurement alone epitomizes value-based care. Relatively few studies look at programs that rely on both of these influences. Fewer still provide insight into their synergistic effect with a rigor that instills confidence in the outcome.

So, a new report authored by Harvard researchers and funded by UnitedHealthcare is remarkable for both its methodical approach and its accomplishment in deconstructing value-based programs. In it, Alyna Chien, MD, assistant professor of pediatrics at Harvard Medical School, and Meredith Rosenthal, a professor of health economics and policy at Harvard T.H. Chan School of Public Health, systematically reviewed value-based care models implemented over the last decade. They began with more than 4 million publications, culling the pool to 2,887 that focused on both spending and quality. From there, only 82—covering 24 value-based programs—made the cut as empirical evaluations. 

Understanding the components of value-based care can help payers decide what role they want to play, says Alyna Chien, an assistant professor at Harvard Medical School.

The two didn’t set out to try to draw conclusions about particular types of programs. More fundamentally, “my goal was to get my arms around the evidence base for value-based purchasing, especially as it pertained to programs that blended spending reduction and quality improvement incentives together,” Chien told Managed Care in an e-mail.

In the process, they came to see that value-based programs really have three dimensions—not just cost and quality incentives, but also supportive components (what the authors refer to as “infrastructure supports”) that make cost and quality outcomes possible. Even the most rigorous studies didn’t always characterize these components, however, so Chien and Rosenthal reviewed additional sources to identify the key features underpinning the programs they evaluated.

Understanding those features, Chien says, is important to payers that develop value-based care programs. “Payers can decide to leave the transformation entirely up to the provider or may want to participate, help or
even lead.” 

Finding the recipe

The two dozen programs spanned Medicare, Medicaid, and commercial payers. Five were multipayer and three more were all-payer programs.

The Harvard researchers segmented the programs by the financial incentives used to reduce spending and by incentives to improve quality. They also stratified the strength of those inducements. For example, among spending-reduction incentives, global payments (with varying degrees of risk) were viewed as stronger incentives than bundled payments, which are limited to an episode, or rate-setting techniques like DRGs. Among quality incentives, gates (in which providers meet a minimum quality threshold to receive a shared-savings award) were deemed stronger than pay-for-performance measures, and ladders (in which the provider’s share in the savings rises or falls with quality performance) were considered more motivating than gates. 

Types of infrastructure support providers need from payers

In their “3D Model for Value-based Care” report, Harvard researchers Alyna Chien, MD, and Meredith Rosenthal identified six basic types of infrastructure supports that payers typically offer to providers in a value-based care arrangement. The graph depicts the frequency in which they are used in the 24 value-based care programs the two evaluated.

Source: Chien and Rosenthal, “A 3D Model for Value-based Care,” October 2019

From there, the two identified payer-provided infrastructure supports they concluded “may be critical to the success of value-based care arrangements.” There have been few efforts to characterize these supports in a systematic way, so Chien and Rosenthal established a sort of taxonomy, breaking them down into six types (see “Types of infrastructure supports providers need from payers,” above).

Two of those supports relate to data sharing, which the report emphasizes is critical for promoting a working relationship between payers and providers. Lewis Sandy, MD, executive vice president for clinical advancement at UnitedHealth Group, says United will typically set up a joint operating committee with ACO providers for the purpose of sharing data and learning what each other can do to address issues found in the data.

Take care fragmentation, for instance. The ACO will know what’s happening with its attributed population internally, says Sandy, but not necessarily what’s happening outside its network. “We’ll know more about that because our programs follow patients wherever they go. And so we might say, ‘Did you know that a certain number of your high-risk patients aren’t actually getting all of their care from your system?’ Many systems find that hard to believe. We show them the data and they say, ‘What are they doing over there? Are we missing something? Is there some opportunity that we can work together on?’”

Two supports relate to the provision of technical assistance, such as hands-on training, or financial or in-kind support (such as the MEOS—Monthly Enhanced Oncology Services—payment in the Oncology Care Model) for building the capacity to execute population health management. Two more supports, risk management tools (such as stop-loss provisions) and care-management devices, round out the list.

A joint operating committee between UnitedHealth Group and ACO providers can help with data sharing and care fragmentation, says United’s Lewis Sandy, MD.

Among the 24 programs, the researchers found that 13 used the strongest spending- and quality-incentives combination, which was global payment with ladders. Nine of those 13 had at least four of the infrastructure supports, which indicates to Sandy that “you really have to invest in this infrastructure dimension [of value-based care] if you want to be successful.” Participants in 12 of the 13 programs received data analyses from payers, while in 11 of them payers offered technical assistance and in nine they provided raw data or payments for capacity building.

Care management tools were the least-utilized support, with fewer than half of the 13 (and only one of the other nine programs) having access to them. And though care management is a critical component for executing population health management, it can be a point of contention between providers and payers. 

Clinical leaders whose opinions were published in the UnitedHealthcare report expressed the view that care management is a not a payer responsibility but a provider function and an expansion of their work. For his part, Sandy doesn’t view it as an either/or proposition. 

“Care management is a complex function,” he says. “It requires sophisticated data and analytics, the ability to do predictive modeling, analyze what you need, what the population needs, and where the opportunities are, and then you have to have the capacity to actually do it.” The offer of care management support, he thinks, should depend on a system’s capabilities. “I think a collaborative approach—like the joint operating committee—can be very effective in working this through.”

On the basis of their findings, the Harvard duo developed some recommendations for future creation of value-based care programs. (See “Five recommendations for upping your value-based care game,” page 18.) There’s much that a payer can do to support the change process necessary to pull off value-based care, but success is more likely when clinicians and their leaders, too, go all-in. Sandy says that’s because value-based care requires a change in mindset—“to move from a focus on what I’m doing in my office, one patient at a time, to really thinking about population health and an entire population and how to deliver better care at lower cost.”

Enduring relationship 

Clinical leaders interviewed within the report understand the strategic imperative. “You’re going to have a harder and harder time getting the full reward in the old way; that cannot be your business model,” said Chris Crow, MD, president of the Catalyst Health Network, a Plano, Texas-based clinically integrated network of primary care physicians. “So, at some time you’re going to have to shift. Do you want to be forced to do it?” Forcing anything is not the stuff of partnership. Therein lies the value of the infrastructure supports and how Sandy says they shape UnitedHealthcare’s approach to value-based care. “Sometimes, these things are looked at as a financial incentive or a contractual innovation, or as an experiment or the next generation of a transactional relationship—which it is—but our view is that we want a different kind of relationship. A multidimensional and enduring relationship.”

Five recommendations for upping your value-based game

Harvard researchers Alyna Chien, MD, and Meredith Rosenthal concluded their review with five suggestions for getting more value out of health care delivery. 

  1. Focus on all three aspects—spending, quality, and infrastructure supports—when designing value-based programs. The infrastructure supports are critical because these are what enable provider organizations to succeed in value-based care delivery.
  2. Increase the proportion of total reimbursement tied to value-based care. “There is a consensus that for these [programs] to work, you have to move from that one-foot-in-the-boat, one-foot-on-the-dock point of view and say, ‘We are going to do this and do it in a big way,’” says Lewis Sandy, MD, executive vice president for clinical advancement at UnitedHealth Group. UnitedHealthcare last year funneled $69 billion—almost half of its annual provider reimbursement—through value-based care models, but Sandy acknowledges that not everyone is ready make this migration. “It’s like the famous quote—‘The future’s already here, it’s just not evenly distributed.’ There are provider organizations that say, ‘Yeah, we think this is the future and we want to move there as quickly as possible.’ Others are in a more exploratory phase or hedging their bets.”
  3. Use infrastructure supports to strengthen payer–provider relationships. Among the programs Chien and Rosenthal examined, those with the strongest incentives for changing provider behavior come with multiple supports—which, to Chien, suggests that “payers have not necessarily left providers completely to their own devices.” Data sharing, in particular, can help to foster collaboration, though senior health system executives whose views were captured in the UnitedHealthcare report expressed frustration with the way this works in practice. “Every payer delivers its data in a different way on a different timetable,” said Shari Rajoo, MD, medical director of population health at Ballad Health in Johnson City, Tenn. “The time lag is something we really struggle with.
    I might be looking at data about something that happened at the beginning of the year and now the year is almost over, so it’s really hard for me to effect a change in anything for the remainder of that year.”
  4. Align quality measures for consistency. Providers spend substantial time and effort tracking and reporting quality performance. Standardizing measure specifications across payers can help to reduce this burden. “Every payer wants things slightly differently and has a slightly different definition of their metrics,” said Rajoo. “Diabetes control is a good example; one is looking for the proportion of your patients over 9% hemoglobin A1c, another for 8%. A lot of work goes into satisfying that. But the last thing I want is for a physician on the front line to be wondering whether he or she has to do a different thing for one payer than for another.”
  5. Align individual physician incentives with value-based contract incentives. Most physicians are shielded from the risk their organizations take on and are rarely incentivized in parallel within their own caseloads. That creates an inherent conflict, says Sandy. “You can’t really get the culture you want to promote population health and higher-value care if you say, ‘Yeah, I get that, but now I’m being scorecarded by my own organization on productivity.’” Chien thinks provider organizations should develop value-based compensation models for physicians, but the form that should take hinges on details and organizational culture. “Physician compensation should match the organization’s goals in producing health, but it’s an open question whether that works better when physician compensation mirrors the VBP contract. I can see that playing out well or poorly.”

Wellness visits may be a waste

Results of a study of Medicare annual wellness visits reported in the November 2019 issue of Health Affairs cast further doubt on the value of routine checkups. Using 2008–2015 Medicare claims data, Ateev Mehrotra, MD, an associate professor at Harvard Medical School and a member of the Managed Care Editorial Advisory Board, compared Medicare beneficiaries seen by practices that provided wellness visits to those seen by practices that didn’t. The outcomes they looked at included screening, both appropriate (for example, mammograms for women ages 50–74) and low-value (Pap smears in women ages 66 and older with no cervical cancer history), emergency department visits, and hospitalizations for ambulatory care–sensitive conditions. They found some differences between the two groups in appropriate screening and emergency department visits, but those differences vanished when they factored in how the beneficiaries used services prior to the wellness visits. 

“In sum, we found no substantive association between annual wellness visits and improvement in care,” concluded Mehrotra and his coauthors. 

One of their suggestions: A more targeted approach that would encourage people who are less engaged in their health care to get the wellness visits. 

Some commentators say the problem is with the content of the wellness visits and what the annual checkup has become. In this critique, they don’t produce good results because doctors rush through them and are only concerned about “checking the boxes.” 

A blueprint for high-volume, high-quality lung cancer screening that is detecting cancer earlier—and helping to save lives

Clinical Brief

Multiple Sclerosis: New Perspectives on the Patient Journey–2019 Update
Summary of an Actuarial Analysis and Report