- Mary Washington Healthcare’s clinically integrated network (CIN) uses a physician scorecard to promote effective population health management.
- The scorecard rates physicians on activity-based metrics that encourage them to provide enhanced access and care to patients of the CIN.
- The CIN plans to eventually use the scorecard to guide referrals to high-value specialists in the network.
Mary Washington Healthcare has prioritized population health throughout the seven-year tenure of Travis Turner, chief population health officer of the health system and COO of the organization’s health plan.
In recent years, a physician scorecard has been one of the Fredericksburg, Va.-based organization’s key tools in its population health ventures. The scorecard shows physicians in the health system’s clinically integrated network, Mary Washington Health Alliance (MWHA), how they are performing in measures of cost and quality.
The plan is to eventually use the scorecard as a means for primary care physicians (PCPs) to assess the value of care provided by specialists in the network, allowing them to make referrals accordingly.
“It’s great to have the population health initiatives that you hope to qualify for, but you need to also have a strategy and plan in place to execute on,” Turner said. “You can’t strategize around hope.”
“You’ve got to create value, you’ve got to change behavior, you’ve got to have accountability,” he added, to succeed in initiatives such as those sponsored by CMS (see the box at the bottom of this article).
In separate interviews, Turner and Rick Lewis, MD, medical director of MWHA, provided details on the scorecard and how it helps drive behaviors that promote high-value care.
Q: Describe the scorecards for the employed and independent physicians who are part of Mary Washington Health Alliance.
Travis Turner: We feel transparency is extremely important in order to change behavior. [The scorecard gives] a comparison of provider to provider within the same specialty. And then it’s a provider to their individual practice. And then it’s that provider to the network.
We also partner with a vendor that provides national data because it’s great to say we want to compare ourselves to ourselves and improve, but we don’t know if we’re starting from a good point or a bad point unless we look at the national benchmark.
Q: What types of metrics do the scorecards incorporate?
Turner: The Alliance provides an activity-based incentive methodology that incentivizes our PCPs to provide additional access, as well as enhanced visits, to our attributable lives.
It also promotes accurate and appropriate documentation of acuity levels that assists in accurately reflecting the risk of our population. If there were conditions in the prior year, it’s important for our primary care physicians to evaluate and confirm whether or not those conditions are still prevalent, whether they’re acute or chronic — that the appropriate acuity levels are provided based on RAF [Medicare Risk Adjustment Factor] and HCC [Hierarchical Condition Category] adjustments.
We have Annual Wellness Visit incentives included in these activity-based incentives. These payments to providers are in addition to what CMS pays on a per claim basis to see those beneficiaries.
As part of our activity-based incentives, some of that is quality-dependent, some of it focuses on a more sophisticated level of care called Advance Care Planning, and another is cost PMPM [per member per month].
When we’re just managing cost, we may come in below the cost target, but we still need to know where our variability is. And so, it’s managing those high-opportunity, low-value services that create that variability.
Q: What types of incentives are used in conjunction with the scorecard?
Turner: We provide incentive payments to the network based on annual performance. It’s typically weighted 70% to primary care, 30% to specialists, and then that split gets down in the weeds to individual performance. The specialists’ 30% is just split across the board, equally. It’s tough to assign dollars individually at the specialist level because you don’t have the [patient] attribution logic.
Rick Lewis: We used to have just one annual distribution. In other words, work done in 2019 would be rewarded in April 2020. But now, we’re able to do real-time chart audits. Instead of waiting for claims to drop, which puts you three to six months behind, we’re able to monitor their activities and performance almost in real-time. We reward them on a quarterly basis, so it’s much more timely performance-based reimbursement.
We’ve seen a number of positive changes in physician behavior, including improved utilization of Annual Wellness Visits and Advance Care Planning and improvement in almost all of the Medicare Shared Savings Program/Next Gen ACO quality metrics — especially HbA1c control, tobacco screening and cessation counseling, and depression screening. It’s been gratifying to see targeted payments result not only in better physician engagement but also in improved patient care.
Q: From a culture standpoint, how did you get buy-in for this type of initiative?
Turner: Our board is very supportive — we have a 20-member board, 15 of whom are independent-provider-based physicians. They understand this [initiative], and they’re really the physician champions and endorse our strategy. The other five board members are system-based Mary Washington Healthcare representatives, and of those five, two are also physicians and fully support the physicians in this. It really arms us with additional support to provide this to the network.
Lewis: Whenever you’re putting data in front of physicians, the most important thing is that they trust that the data is accurate. Otherwise they’ll say, “I’m a high spender because you’re wrong.” It’s nice to show that we care about the data and that we stand behind it, and that we’ll get them an answer right away [if they question the data].
That’s one of the important things about having a medical director. When the analytics team comes up with any discrepancies, they don’t put it out there. We go over it first. We look at the clinical background and the potential explanations.
Q: What steps do you take to encourage physicians to act on the information they get from their scorecards?
Turner: It’s voluntary for now — we’re providing transparent data to the network, and then we’re kind of behind the scenes policing them and enforcing conversations. I’m available and my analytics team is available as part of the discussion, should the data be questioned.
But what’s more important is the peer-to-peer, doc-to-doc discussion that is happening: “Hey, why is your generic fill only at 32% when all of your peers are at 87%?” The fact that we’ve established a consistency in creating these reports allows for those discussions to take place.
Q: Where can you take the scorecard from here in the context of patient referrals in the network?
Lewis: We’re getting [physicians] used to seeing that cost data and seeing which patients have higher spend associated with them. The next step is actually doing research and doing analytics on the specialists in our network and seeing whether there are outliers in terms of utilization of tests, of laboratory, of advanced imaging. Then if you have higher- and lower-value specialists, and we make that data known, you can start talking about encouraging referrals from PCPs to higher-value specialists within the network.
That’ll be an interesting new day. If you’re going to use data to influence physician referral patterns, that data will need to be transparent, accurate and substantiated. We’ll probably have a minimal penalty at first, just to open people’s eyes to the fact that that type of referral system can be done. Hopefully, you’d see improvement in lower-value specialist performance, avoiding the need to move on to more onerous penalties.
Q: Is there a process for physicians to review and potentially appeal their data?
Lewis: The specialists see their claims, their risk score, the risk score of their peers, and they see their costs. That has engendered some conversations. For instance, a pulmonologist was in a group of seven. The other six were pretty uniform, and he really stood out — his costs were high.
It turned out that when rounding in the ICU, he was using intensive care unit codes rather than follow-up codes. Most of his peers, once they did the initial evaluation and the patient was stable, would drop down to that lower code.
We educated him on that, his next [scorecard] came out and he was back down with everybody else.
Turner: We obviously answer all questions, reach out, we’ll meet face to face, we’ll bring the data. It becomes educational for both of us, and you tend to find data that is standing out. It builds trust as well because we’re not trying to sit there and point the finger, and they’re not doing it intentionally eight or nine times out of 10. These reports help them because otherwise they’re just performing their job and trying to provide great care, and they may not know where they stand among others.
We’ve got some tremendous data points at the practice level, within their specialty — so now they know how other groups in their specialty are performing — and then we provide that national perspective as well. So it’s been a really great engagement tool.
About Mary Washington Health Alliance
The clinically integrated network of Fredericksburg, Va.-based Mary Washington Healthcare includes:
- 80,000 covered lives (commercial and Medicare)
- 450 physicians (including 150 employed)
- A 3-to-1 ratio of specialists to primary care providers
The network’s population health initiatives include:
- Medicare Shared Savings Program and Next Generation ACO, with plans to apply to participate in the MSSP Enhanced Track
- Bundled Payments for Care Improvement (“Classic” and Advanced models)
- Mary Washington Medicare Advantage (launched in 2020 with more than 1,500 covered lives)