Why Retrospective Code Review Shouldn’t Be Your Primary Risk Adjustment Strategy

By Dr. Seth Flam, CEO, ForeSee Medical
Twitter: @ForeSeeMedical

The risk adjustment rules insist that each new calendar year, diseases and their correlated ICD codes from the previous year, be properly recorded in the current year, when clinically suitable. You may think it common practice that providers would just redocument clinically valid diagnoses by looking back at their encounter notes and update the disease burden of a patient. But there’s a few important things to consider. One is that providers don’t always see patients in the framework of a typical calendar year. Any sensible provider would ask themselves, why would I complete the same work again a month later if my patient was seen on December 31st, and had all of their diseases properly recorded? The truth is, most providers view the typical calendar year as just a discretionary time frame along the continuum of care.

The coding method known as “code recapture” analyzes ICD codes used in the previous year and displays them to the user for reapproval in the present year. Since the typical calendar year assessment periods don’t always align with standard medical decision-making processes, more often than not, the recapture process never occurs. Recapture is important, and all risk adjustment tools should include that functionality, but not exclusively. It’s easy for most EHR developers to build applications that simply recapture ICD codes. Simple functionality is always the easiest to develop, but depending on recapture alone, for an entire population of complex patients, often equates to under-documentation missed RAF score possibilities.

During a new patient’s first encounter they typically don’t show up with detailed documentation and a thorough list of their ICD codes from the prior year. Most times, they bring a pile of printed records, which the medical staff just doesn’t have the resources to process through right away. Also remember, seniors produce multiple visits per year and the strain of comprehending old records into a new chart can be overpowering. For most new patients, the recapture strategy just isn’t realistic.

The natural evolution of many chronic diseases like diabetes is to worsen, even with the best care. If, in the previous year, diabetes without complication was documented, it is possible that in the present year, the patient has progressed. It is also possible they may have a complication of diabetes, for example retinopathy or neuropathy. So for patients with progressive diseases, recapture alone won’t work either.

On the other hand, disease discovery uses the structured clinical data already present in your electronic health record system in the form of medication and problem lists, lab test results and vital signs. It can also decipher the data buried in the unstructured data like text and PDF clinical notes and simplify how software can understand and process it. We all want to understand and properly document the disease burden of our patient populations accurately. Complicated coding models make that job difficult for both physicians and coders. Today, with more computing power than ever before, we now have the ability to do what recapture alone will simply will never achieve, and that’s to accurately discover disease. If you engage in value-based reimbursement models, disease discovery for risk adjustment coding is the key to improving care, more complete documentation and sustainable profitability.