CMS is now poised to roll back quality reporting requirements in 2021, vastly altering the direction of quality measurement. Simultaneously, CMS will reduce the weight in Value formulas dedicated to quality, transferring the balance to Cost over the next five years. As providers face risk-based reimbursement, what protections are needed to ensure that patients get the right care? Does streamlining the program give providers a pass on quality? And, how do patients choose providers when there is no standardized measurement?
In this second in our series on whether Value-Based Health Care is on track to meet its mission, we take a closer look at Quality and its role in defining Value.
What’s the Point of Quality in Value-Based Health Care?
Most industry people connect Value-Based Health Care with money, purchasing, affordability, services—the finances and transactions of health care. But Quality is really the heart of Value. Quality and its more measurable results, Outcomes, are what Value-Based Health Care was intended to incorporate—for real people. It is a benefit-to-cost ratio that involves consumers as well as third party purchasers.
Value-Based Health Care is not just about controlling costs; it’s about what we get for our investment. If we spend more than any other country on Earth, yet our life expectancy is lower, we have a deficit of public health benefits in relation to what we are spending. More evidence: the highest maternal death rates among all industrialized countries, the highest opioid addictions, and inequities in outcomes for African Americans. Throughout our health care system, outcomes are falling short in relation to what we pay.
That is where we now stand with Value-Based Health Care. Quality was supposed to be a key component, measured by whether individual patients’ care and results reflect Quality. However, the imperfect design of Quality measurement, the use and abuse of its “scores” by provider organizations and CMS, and the flaws in the Reporting system for Quality have each worked to sabotage meaningful efforts to measure and implement good care.
How Did Quality Measurement Go So Wrong?
The early and voluntary Physician Quality Reporting Initiative (PQRI) was the first formal Quality reporting to Medicare, based on newly adopted measures accepted by the American Medical Association and provided to CMS. Managed care plans had used HEDIS measures approved by the National Committee for Quality Assurance (NCQA) for years, with data sometimes compiled by providers and reported to health plans. PQRI measures resulted from work groups, with ultimate oversight by the AMA, which brought together specialty physicians and experts to agree on standards. Existing measures from the National Quality Foundation (NQF) and Association for Health Research and Quality (AHRQ) as well as HEDIS measures formed the basis of PQRI measures, but other organizations also came into play.
One key precept was that quality measurement in medicine should be universally applied. That meant that measures were constructed for all specialties and all physicians, so that everyone could participate in quality measurement.
Let’s think about the effect that has had on today, with 250 MIPS measures in 2019 (actually reduced significantly from previous years).
Quality measurement was designed, despite individual patient measurement, to be physician-and specialty-focused. The “fairness” rule was applied by the founders of the system: it was unfair for one group of physicians to be measured and another not. To accomplish that across about 38 medical specialties and subspecialties—including some for which quality mattered to patients more than others—required, at the outset, a system with many measures.
How Provider Organizations’ Applied-Quality “Scores” Hurt Physicians
Administrators eager to push their organizations to achieve higher Value-based incentives used raw measure results as cudgels, embarrassing or punishing physicians with poorer measure results, without understanding the data’s limitations.
There are many reasons why data is incorrect or not captured correctly. Data is not pure, complete, or even correct because it comes from a database. It’s often just bad data or not well-collected, and good and bad results alike need to be reviewed and validated by some provider-driven process.
Having received data from thousands of systems, our company once provided pretty awful “Data Integrity” scores back to provider clients. Our intention may have been good, to help moderate the use of the measure data and to improve data capture and storage. But it was still a bad idea that we stopped quickly. Our clients were also victims of their vendor’s systems and their implementation by long-gone staff, not to mention that change is slow.
Making Reporting Easy Also Makes Quality Impossible to Compare
One of the obvious goals of PQRI, then PQRS and MIPS, was to use Quality results not only for Value-Based incentives, but to improve. Also, in the early days of Value-Based Health Care, optimists believed that comparing quality measure results by practices would help patients choose doctors. The CMS Physician Compare site was established to realize that vision.
But CMS allowed “give” in quality reporting from the beginning. A quick and easy sampling of patient quality results was given the green light in lieu of all-patient reporting. A transitional year was introduced under MIPS to permit providers to report just one measure. Group reporting versus individual provider reporting, reporting through various organizations or directly through EMRs, or, for ACOs, submitting fewer quality results on a sample of patients were all okayed. Finally, since providers could select measures to be reported, they could optimize their results under reporting.
In each case, lenient reporting has reflected the reality of a health care system that is too complex and too varied to be easily pigeonholed. Quality in Value-Based Health Care was idealistically designed, but has proved to be unwieldy.
Calculating the Appropriate Burden of Quality Measurement
There is no doubt that organizations have hired legions of administrators to oversee quality measurement and data capture, affecting support staff and, sometimes, clinicians in practices. Clinicians have also had to ensure that they have captured the detail supporting measures in EMRs during or after patient visits. Practices without EMRs have had to find alternatives at great expense and manually input data into other systems. And sometimes those systems don’t store the data in a way that is easily retrievable, so post-audits of clinical records are necessary.
The work of quality measurement can’t be minimized, but we should separate the underlying systems issues, such as:
- How can certified systems be unable to produce stored data in standard formats? Shouldn’t minimum data reporting functionality be required as part of certifications?
- If we have determined that electronic record systems serve the public good (which I think is the case), shouldn’t providers ensure that practitioner use of the system results in retrievable quality and clinical data?
The fact is that if EMRs are well implemented and maintained, clinical and quality data is retrievable and of high value, with very little burden on physicians or staff. That is demonstrated by excellence in most of our clients’ data, which can be processed quickly and seamlessly flows into quality measure results.
Limitation of data capture to claims will dumb-down the Value associated with Quality in a big way. Instead, our efforts should focus on making real Quality reachable and easier to attain.
Should We Abandon or Roll Back Quality Measurement?
The famous Peter Drucker quote, “We can’t improve what we can’t measure,” is truth. Without measuring specific results, we don’t know how close we come to the standard of care, nor can we address variations. Before the current Quality effort, there was virtually no standardized method that required physician performance measurement.
There should be concern, under financial Risk, that patients who are sick will be steered away from organizations that are paid per patient. It happened under previous Risk models. Our economic system has a perfected acumen for maximizing profits, and as health care systems have consolidated into mega health enterprises, the bottom line is watched closely.
These questions are essential:
- What unintended effects will occur if providers are no longer required to submit quality scores?
- If we require providers to internally capture and measure performance but do not require reporting, would effectiveness diminish or accuracy suffer?
- What other measures should we establish to protect patients from poor quality under Risk?
- Is universal physician quality measurement necessary—or even desirable?
- Can administrative data (claims), under CMS proposals for Quality in 2021, provide enough to support quality measurement without clinical outcomes data?
- How should quality measurement deal with new scientific studies on effectiveness and/or unintended effects?
- Where do patients weigh in on quality of care? How can we make patient-reported outcomes practical and add to the value of measuring Quality?
We need these questions answered in the process of revamping quality measurement and reporting. Quality measurement should be addressed not only by providers affected by the process, but also by consumers and advocates. As we think forward to how to best configure Quality as a part of Value, we should do less roll-back and more reshaping. We’ll tackle that challenge in future articles.
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: Ella Olsson