Brian Baker

The self-referral, appropriate use, and pay for quality conundrum

August 29, 2013
By Brian Baker

The noise level around health care related to cost seems to be increasing of late, while initiatives to drive interoperability and quality are missing the mark and losing their leaders. Meanwhile, individual hospitals are facing millions in cuts while enduring increasingly invasive payment or RAC audits. The bottom line here was expressed recently by a physician at the AHRA Physician Leadership Day: "Medicine isn't the problem, it's the business of medicine."

We have identified the enemy and it is us

Back before DRGs (Diagnosis-Related Groups) were implemented, around 1982, payments in health care were based on cost reporting and were paid on a "cost plus" methodology. Critical Access Hospitals ("CAH") still operate this way today, getting paid 101 percent of reasonable costs. Other reimbursement options exist for CAHs, but that's another article. The DRG system unified the payments based on groupings of services provided. Come in to the ER with a possible heart attack? Medicare has a code that pays one amount, regardless of how many aspirin are consumed. As a result of this payment grouping, providers haven't paid as close attention to the cost reports any longer. They are still filed, as it's a requirement, but may not have the same level of attention to accuracy paid to them. As an organization that consults for hospitals, we rarely see an accurate expression of procedural costs.

Fast forward to today. Cost reports are used by Medicare under advice by groups such as the RUC (Specialty Society Relative Value Scale Update Committee) to adjust Medicare reimbursement. To be fair, that description is a gross oversimplification of the effort and information utilized. But what is important to know is this: The cost reports, as explained recently by one member of the RUC, while variable between submitter, indicate a low cost to deliver things like medical imaging that is not representative of true costs.

So if Medicare and its advisers are using the cost reports, among other things, to help determine reimbursement, and the reports show a steady or declining cost, it might make sense then that Medicare is reducing reimbursement. Have we missed something along the way?

Your credit line has been exceeded

The sustainable growth rate (SGR) originally created in 1997 to help limit spending from Medicare for physician services has been exceeded every year since 2002. The fix was intended to be what are called "negative updates", which means reduced reimbursement. But for the last ten years Congress has passed a series of short term fixes (kicking the can down the road) to prevent these cuts from happening. So today the estimated cost to fix the disparity between the original SGR formula from 2002 and where Medicare spending is now is estimated by the Congressional Budget Office to be significantly more than $140 billion. Yes that is a "B". While the House Energy and Commerce committee voted 51-0 to repeal or fix the SGR on July 31, 2013, no plan has been developed to pay for this fix...yet. If a fix is not in place by the end of 2013, physicians who treat Medicare beneficiaries will see cuts estimated at 25 percent. If history is any indication, expect the can to be kicked again. What we really need is a solution. Most physicians today don't have a profit margin of 25 percent. Without a fix or a kick of the can, this is unsustainable.

Free beer tomorrow...

...said the sign in the restaurant. Sound familiar when you think about health reform and the changes since it passed in 2010? Transparency cried the masses! We need transparency through access to data to determine what is really going on! Yet, call a Medicare Administrative Contractor ("MAC") and ask for de-identified utilization data. Click... buzz is what you will hear. Call Medicare directly and ask the same. After several hoops you can get a 5 percent file. That is one in 20 claims; hardly representative and impossible to determine who is referring what to whom with statistical certainty. HIPAA rules aren't making it any easier. The problem is, in order to get to real automated clinical decision support, we need access to the linear patient records, including outcomes. Another way to think about "pre-encounter" clinical decision support is predictive healthcare. Using the scads of data already available in the hundreds of millions of patient records to analyze treatment and outcome patterns would clearly reveal what works and what does not.

Solving the riddle

Self-referral (of diagnostic testing services) is a complex issue and is driven by several factors. Some legitimate, like proximity to the testing services; other self-referral excuses perhaps not so legitimate. Self-referral, the growing use of diagnostic testing services in general and the increasing cost to payers, including Medicare, are the primary engines behind the cry for appropriate use. Then, in no small part due to the efforts of some clever statisticians who quantified broad gaps in patient outcomes compared to costs in several markets across the country, the light came on flashing "Pay for Performance" ("P4P") not just for doing the work. Quality in other words. Perhaps you see now how wise the statement was in our first paragraph: "Medicine is not the problem, it the business of Medicine". Any business with a poor quality product (Inappropriate use, poor outcomes, high cost) will not last long. Not everyone is on board with pay for performance as a recent study by the JAMA published by Diagnostic Imaging found 70 percent of 2556 respondents not in support of eliminating fee-for-service. However, that same study revealed 89 percent believe they need to take a more prominent role in limiting the use of unnecessary tests.

To get to appropriate use, which drives quality and keeps cost down, we must first completely understand our current care models and outcomes, to determine what changes must be made to improve. The caveat is: you cannot improve what you cannot measure. The HITECH Act is paying providers (Meaningful Use) to report on key measures while no structure or incentive exists to capitalize on and learn from the data now being collected. Providers are being paid to simply report on collecting basic data. It's a wonderful start but there is no process or plan that we know of, to actually learn from the data. Larger than that is the lack of interoperability between information systems in health care which, by the way, is driving some of the duplication of services, waste, and cost through a lack of access to complete patient histories between providers.

Without a standard for collecting and communicating longitudinal patient records for analysis, learning and feedback, it's going to be tough to learn from the data. Compounding the problem is a lack of a national patient identifier and a well-founded concern for protecting health information.

To drive pre-encounter clinical decision support based on actual patient outcomes, to propel appropriate use; to minimize cost and maximize positive patient outcomes, we need to connect patient data nationally and learn from it to make decisions. It's a simple concept, a feedback loop.



Feedback loops, sometimes thought of as Artificial Intelligence, by design, learn from themselves by constantly evolving and adjusting, using historical and new data, as defined by the analytics design, to deliver actionable measures. This technology is no longer Star Trek fiction - it exists today. These approaches can allow the bi-directional connection of health IT systems and deliver the advanced automated analytics clinicians can use to make decisions based on accepted criteria. The result is appropriate tests (bring down cost) that reveal the answers to the clinical questions (based on past and current results) that direct clinicians to treat for the best (proven through data) outcomes. The new outcomes feed back into the data because health care data are standardized and connected. Sound like a circular reference?

Boldly go where no one has gone before; (to) live long and prosper

Disruptive innovation, we are told, is innovation that creates new markets and value networks. Do the goals of the Affordable Care Act ("ACA") represent a corollary? If the ACA is promoting P4P, the fundamental problem of measuring our performance today must first be solved. To get there the measurement capability must be able to quantify the thousands of elements needed to determine quality and performance through the measurement of thousands of outcomes, not simply attesting that the data is being collected. It's time health care boldly solves its own issues following in the footsteps of other dominant industries to demand standardization and communication of health care-related data. Then we must make ourselves and our vendors accountable.

Based in Franklin, Tennessee, Regents Health Resources is a national consulting firm focused on the complexities of medical imaging. As President of Regents, Brian concentrates on business strategy and development focused on industry challenges.