by 
CMS is urging providers to participate in ACOs with downside risk, but they might be eliminating one of the keys that providers need to prepare. It couldn’t come at a worse time, when providers already stand to lose under risk-based models if they can’t identify where their cost issues lie. That data is only available from claims data made available by payers.
Up until now, practices have had access to indispensable data on costs that are attributed to their providers, showing specifics of where they are above the norm. These were previously part of Quality and Resource Use Reports (QRURs) that have been distributed for years, first under the Value-Based Payment Modifier program and then under its successor, MIPS.
CMS informed providers this month that QRURs were only for PQRS and the Value-Based Payment Modifier (VM), and will no longer be distributed. The VM was incorporated completely in MIPS, but CMS is clearly moving in a direction that will dismantle MIPS in favor of ACOs or other risk models.
What will MIPS providers get instead? So far, a final score with either an incentive or penalty determination occurring in 2019, without explanation. Although CMS seems to be wavering on whether additional data will come later, this is a big departure from the past. What should providers do when this data is no longer available?
How the QRURs Helped Providers
While the purpose of the QRUR was to provide a VM score, the additional exhibits, containing patient-level detail, were beneficial for practices that were interested in identifying and addressing detrimental utilization trends. The information wasn’t actionable on its own, but those who partnered with some advanced Clinical Data Registries (CDRs), such as Roji Health Intelligence, saw their QRURs transformed into meaningful analytics, including the conversion of opaque Medicare IDs into identifiable patient details. By revealing the patient case and individual provider, the data triggered episode-based analytics and identified outliers.
Without these supplementary materials, practices lose the ability to dig into the “why” of their cost score—only the score and high-level information are available. This is significant: although cost was not figured into your 2017 MIPS score or 2019 payment adjustment, it will be worth 10 percent of your overall score in 2018, and 15 percent in 2019 (according to the most recent proposed rule released July 12, 2018). And that’s the optimistic scenario, because providers may well be required by next year to participate in financial risk models like ACOs or direct provider contracting by future administration proposals.
Without receiving the cost data directly from Medicare, providers need alternative means to evaluate and improve cost performance. One way is to involve a Clinical Data Registry to help develop the data and analytics necessary to measure and improve performance in both cost and outcomes.
MIPS Cost Scoring in 2018
So, that’s the new landscape. Understanding how your costs will be assessed in MIPS is critical to developing your strategy for success and ongoing improvement. The 2018 Cost Score will be calculated exclusively using two cost measures:
- Total Per Capita Cost for All Medicare Beneficiaries (TPCC), designed to evaluate efficiency of care provided to all patients attributed to a specific TIN.
- Medicare Spending Per Beneficiary (MSPB), designed to evaluate efficiency of care provided to patients, as it relates to a specific episode of care, where that episode has been attributed to a specific TIN.
Other episodic cost measures have been “field tested,” but these metrics are still being developed and will not be included in MIPS scoring until 2019 (at the earliest).
CMS scores both the TPCC and MSPB measures exclusively using adjudicated claims data from services occurring during the measurement (performance) year. In other words, unlike other MIPS measures, no information separately reported by the practice or another third party (e.g. Qualified Clinical Data Registry, Qualified Registry, EHR, Survey Vendor) is included in the scoring process. It is important to note that, although the other MIPS categories include all patients (unless reporting Quality via the CMS Web Interface), because Cost is calculated by CMS, only Medicare patients are included in the Cost component of MIPS.
The Total Per Capita Cost is a standardized dollar value indicating the average sum of Medicare Part A and Part B costs for each attributed Medicare beneficiary over the measurement year. Medicare Spending Per Beneficiary is a standardized dollar value indicating the average spending associated with an MSPB episode of care. An MSPB episode includes the sum of all Medicare Part A and Part B costs for services beginning 3 days prior to, during, and 30 days after an Inpatient Prospective Payment (IPPS) hospital admission.
Prior to scoring the TPCC and MSPB measures, CMS performs risk and specialty adjustment on claims, and standardizes payments to account for variation unrelated to care (e.g. location). This means that the numeric value of the benchmark is the same across the country, but CMS determines beforehand whether a dollar spent on care for an attributed patient in one practice is equivalent to $1.10 at another practice.
MIPS Feedback Reports Are Summaries, Not Tools
The feedback delivered on 2017 cost metrics contributes to creating a value-based strategy, but is not enough information, alone. The reports show only the highest-level details for each measure. You’ll see the number of attributed patients or eligible cases, the measure score (in dollars), a measure ratio and performance (points and decile). There is an additional summary of Emergency Department Utilization showing the number of “associated patients,” the number of those patients who had an ED visit and the total number of ED visits.
Without the breakdown of patient attribution, (primary care services delivered by a primary care provider, rather than a specialist), a window onto the in-network/out-of-network referral patterns has been slammed closed. It will be impossible for your group to know whether you have a significant problem with patients accessing specialty physicians without any primary care coordination, because you won’t see this data. Therefore, you also can’t execute a targeted strategy by specialty for ensuring that patients are connected to primary care physicians.
The same issue is magnified on the inpatient side, as the listing of admitting hospitals (both for episodic care and total per capita costs) is also absent.
Information on care coordination is similarly missing. Groups could formerly see the breakdown by type of care that patients received. Practices that saw lower-than-expected ambulatory and post-acute care and higher-than-expected ED utilization could infer that care was not sufficiently coordinated. Additional emphasis on care transitions after a procedure may mean decreased emergency department utilization or hospital admissions.
Without actionable data, results will be difficult to leverage. There is no way to determine the areas in which you performed well or that need work. In order to lead the field, organizations will need to use their own data to understand and improve spending and utilization patterns. This is a perfect opportunity to work with a Clinical Data Registry. Why? Because the CDR is specifically designed to assist in performance improvement, integrating both cost and quality performance. The CDR, which blends analytics, performance measurement and improvement activities, is built on data that is aggregated and customized to performance improvement projects.
Five Strategies for Taking Control of Your Costs
No matter how you interpret the recent CMS decision not to release detailed cost data, one thing is clear: Medicare Fee-for-Service, and possibly MIPS, will erode, replaced by a risk-based payment system. Therefore, providers who have focused on maximizing revenues in the past must turn their attention to managing efficiency of encounters, together with maintaining or improving outcomes. That will require major changes in the way providers work now.
What do you need to do to succeed? Start with developing available data and then optimize value. Consider these five tactics:
1. Choose technology to focus on cost and quality performance, or use the services of a Clinical Data Registry as a technology vendor.
Current provider systems are not generally optimized to look at performance and, especially, costs. A CDR that can aggregate comprehensive transactional and clinical data from multiple data sources and facilities, along with the analytics to visualize what’s important, can develop the tools you’ll need to measure and reduce spending.
2. Understand your services, costs and inbound/outbound referrals.
A CDR with the ability to ingest years of historical billing data can illustrate the types of procedures provided (and trends) and can identify common conditions and progression. Many groups—even those with EMRs—have been stymied by a lack of basic data on group and provider volume by procedure, as well as costs.
3. Build medical and procedural episodes as the unit for evaluation of discrete costs and clinical quality.
In order to undertake risk, groups must evaluate production statistics from an episodic rather than a by-service standpoint. Episodes allow comparison of variations in services and costs by provider, and offer a mechanism to evaluate what is going right or wrong in the delivery. Additionally, episodes create the mechanism to examine variation, spending by types of procedures or diagnoses. Episodes are more instructive than focusing on total costs in both the TPCC and MSPB measures, because they lead to actions in specific areas. If conservative treatment, or a more minimally invasive (but equally successful) procedure can be utilized, spending can be reduced while improving patients’ qualify of life. For example, studies have shown that laminectomy with spinal fusion does not improve outcomes more than the laminectomy on its own.
Lacking payer or Medicare data, your technology can still identify patients who have had a specific procedure and then attach other services delivered within a specified window. Even if the patient has a different MRN at multiple practices, or if different practices use different EHRs, a CDR should be able to track a patient from one location to the next. If all components of the episode (surgery, radiology, anesthesia and medicine) are present and accounted for, it’s possible to build a preliminary bundle.
With a CDR, it’s also possible to incorporate feedback to examine the reasons why procedures were performed, in a peer review context. With the technology and episodic measurement, you can identify variation by provider and procedure, and begin the conversation on change.
4. Identify out-bound referrals and out-bound patients.
Your technology or CDR should be able to calculate where system outflow is occurring. This is essential not only from a cost control standpoint, but as an indicator of clinical quality or customer service. Since out-of-network services can be attributed to your organization, understanding this will be critical to your cost levels.
5. Pilot and roll out performance improvement projects—best in collaboration with health plans—that incorporate cost and quality and permit practicing cost control.
While participation in the QPP (either through MIPS or an APM) is required, there’s no reason to limit your focus. By looking at procedures by payer (e.g. how many total knee replacements were covered by one health plan) and payers by procedure (the mix of health plans that have covered a specific procedure), you’re in a position to negotiate your own arrangement with that health plan. If you are able to obtain patient-level and aggregate-level data related to a certain procedure or population, your CDR can incorporate those claims into its technology to provide you with previously unavailable insights, and then incorporate and track interventions designed to improve performance.
While the MIPS Feedback and Score Reports were very clear and concise, they never gave providers the full information needed to improve cost performance. Translation and integration with other data was always required to identify cost issues. These five strategies will work regardless of data source or data set. The availability of cost data has never been the primary stumbling block to addressing costs, although it has often been used as an excuse.
There is no short cut to creating better cost performance in health care, and the lack of CMS detailed data, while disappointing, should not be a severe deterrent. While claims data will be needed in the future, some health plans are more willing to engage in programs that help providers develop the technology, data and process acumen, and the cost results they need. Providers will have to address their costs across the board, regardless of payer, and it doesn’t much matter where they start—it only matters that they do.
To view the CY 2019 Quality Payment Program proposed rule, please click here.
Founded as ICLOPS in 2002, Roji Health Intelligence guides health care systems, providers and patients on the path to better health through Solutions that help providers improve their value and succeed in Risk. Roji Health Intelligence is a CMS Qualified Clinical Data Registry.
Image Credit: Ryan McGuire