By David Fletcher, MPH
Vice President of Innovations, Streamline Health, Inc.

In healthcare circles, there’s a lot of talk about merging clinical and financial data to face the new realities of value-based reimbursement. At a basic level, we think of clinical data as lab results, medication orders, findings and problems. Financial data typically includes charges, payments, denials and cost of operations. In between clinical and financial, we have admission/discharge/transfer (ADT) or “administrative” data that includes patient info, encounter, diagnosis and procedures. I LOVE data but merging all these data domains by itself will serve no one (see Peter Rivera’s take on the Data Warehouse Field of Dreams). A more interesting question is “What problems are we aiming to solve by merging these data, deriving insights and recommending action?”

The problems start with the shifting reimbursement models coming from CMS. Ours is an industry that is heavily influenced by policy changes in Washington. The latest broad legislative mandate came in June 2015 when the Medicare Access and CHIPS Reimbursement Act (MACRA) was signed into law. Meaningful Use program incentives and penalties will be combined with other programs into the Merit-based Incentive Payment System (MIPS). The net impact of these developments is that it will no longer be enough to meaningfully use an EMR. New MIPS incentives will include clinical practice improvement, quality and resource use factors, along with meaningful use of the EMR. Additionally, Alternative Payment Models (ARMs), such as Accountable Care Organizations (ACOs), Patient Centered Medical Homes (PCMHs) and bundled payment models are covering more and more of the Medicare patient population. While many will continue to debate how much of the Medicare population will be covered by MIPS and ARMs by 2018, few will argue against the premise that clinical quality is quickly becoming central to reimbursement optimization. In my option, these are good problems to have. As Richard Farson points out, many problems are borne from the achievements of the past, so embrace them.

These days, the Revenue Cycle Director, the Clinical Documentation Improvement (CDI) Director, the Health Information Management (HIM) Director and the Quality Director find themselves collaborating in new and unprecedented ways. To chart these new waters, let’s think about some of the ways that integrated data, strong analytics and key workflows can help. The CDI program seems like a logical place to start. Typically, a CDI team is active while a patient is in the hospital, looking for ways to improve documentation that will legitimately optimize a bill. In our new reality, the goal itself does not change but the focus on how to achieve it does. In the merit-based reimbursement world, there’s a shift in the things one must document and take action on. For example, under the new reimbursement models, we need to track and address comorbidities and chronic conditions even when a patient is coming in for an acute new problem. And when we do, we can increase reimbursement because these patients are more challenging to care for. Therefore, running what in the past might have been quality improvement analytics against a clinical data repository can now help to scour patient history for indications of comorbid chronic conditions that analytics against the billing data for those same encounters might not pick up.

For a good practical example of this concept, consider patients with diabetes. Medication prescriptions for Metformin, A1C values in abnormal ranges or a documented problem in the problem list are all indications of diabetes. By alerting your CDI team to in-house patients who have one such clinical indicator of diabetes but no historical billing charges for the same represents an opportunity for the organization. Those patients represent both a gap in care and potential reimbursement dollars left on the table.

I think we can go even further to inject “what if” scenarios into our estimation of reimbursement for key patient sub-populations. Most provider organizations amass a wealth of data about their patients—clinical and financial, including encounters, treatments, diagnoses, charges, payments, costs, lab results, and more. By leveraging combined analytics, the possibilities are limitless with regards to the actionable insight that can be gleaned and subsequent projections. And this can have a positive impact on the quality of care delivered/outcomes achieved, which will eventually drive higher reimbursements under MIPS, ARMs and other new payment structures.

Using the combined data from clinical and financial domains, we should be able to project reimbursements for a given patient group or cohort with some degree of certainty. To then assess the financial impact of an effective clinical or patient service initiative, we could manipulate relevant clinical variables to reflect the “what if” scenario and recalculate reimbursement projections. Giving quality improvement teams and researchers an easy way to assess financial impact is just one more step toward aligning financial interests with good patient care.

The shift to new reimbursement models is well underway, so resistance is futile. Forward thinking providers should consider not just which data to analyze, but in what context and with what goal(s) in mind. When you “begin with the end”, then the value starts to become apparent.