by Trey La Charité, MD
When providers use different definitions for the same disease, confusion and chaos result. Inaccurate definitions allow Recovery Auditors (formerly called RACs) ample opportunities for denials, depriving your facility of valuable dollars/patient care resources. Unfortunately, providers learn disease definitions during medical training. When they begin practice, adopting new disease definitions or accepting terminology changes proves difficult.
As clinical documentation improvement (CDI) and coding professionals, guiding your facility’s providers towards more accurate disease definition usage is challenging, frustrating, and an on-going process. And yet, maintaining accurate and consistent definitions of medical diagnoses is an important aspect of the CDI specialists’ role.
Definitions of medical diagnoses are rarely permanent; they change over time as our understanding of disease pathophysiology, recognition, and management evolves. This creates a significant challenge for CDI and coding professionals and the providers with which they work. Since there is no uniform method for distributing medical information, adoption and implementation of changes vary tremendously between providers. For the CDI or coding professional, this poses two problems:
- Providers fail to learn and apply new definitions (and continue to use outdated terminology)
- Providers adopt new terminology faster than the ICD system can create corresponding codes
Outdated terminology may lead to unnecessary downcoding; using terminology with no corresponding ICD code may lead to a diagnosis not being reported. Both problems can lead to providers not receiving appropriate credit for the severity of illness of their patients.
Changes in heart failure definitions
Let’s look at heart failure, for example. Starting in the late 1980s, the pathophysiology of congestive heart failure (CHF) shifted from the classic left-sided versus right-sided designations taught in basic physiology courses to the concept of systolic function versus diastolic function.
Documenting CHF as either systolic or diastolic (in addition to being described as acute or chronic) allows coders to assign the most specific and highest-weighted codes possible in our current ICD-9-CM coding system. However, if a clinician uses non-specific, outdated terminology such as “CHF exacerbation” or “acute left-sided CHF,” the corresponding ICD-9-CM code(s) are lower-weighted, the patient doesn’t appear to be as ill, and the physician appears to not have done as much work to provide care to that individual.
Additionally, the financial ramifications for a healthcare institution can be significant as “acute systolic CHF” is an MCC, “acute left-sided CHF” is a CC, and “CHF exacerbation” is neither.
On the other end of the spectrum, consider a clinician that has already adopted the new heart failure terminology proposed by both the American College of Cardiology and the American Heart Association in 2013 (Read the report, “ACCF/AHA Guideline for the Management of Heart Failure: A Report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.”).
In this new taxonomy, heart failure has been subdivided into “heart failure with reduced ejection fraction” and “heart failure with preserved ejection fraction.” (These new phrases are invariably abbreviated in the medical record simply as HFrEF and HFpEF.)
Currently, there are no corresponding ICD-9-CM or ICD-10-CM codes for these descriptors. Therefore, if these terms are used by providers, coders may only code “congestive heart failure, unspecified” (ICD-9-CM code 428.0) which does neither the clinician nor their host hospital any good. In fact, Coding Clinic for ICD-10-CM/PCS, First Quarter 2014, p. 25, stated that coders may not assume that “HFrEF” and “HFpEF” mean “systolic” or “diastolic” heart failure respectively. For clinicians that practice in an academic institution, this can be a very difficult situation.
Changes in renal failure definitions
As a second example, let’s consider acute renal failure. For years, the definitions for this diagnosis varied widely. Finally, in 2004, the RIFLE criteria—Risk, Injury, Failure, Loss, End-Stage Renal Disease—was proposed by the Acute Dialysis Quality Initiative Group, yielding one uniform definition for diagnostic accuracy.
However, by 2012, the Kidney Disease Improving Global Outcomes (KDIGO) work group rewrote the definition and proposed the new terminology of acute kidney injury (AKI). While this change was instituted to facilitate provider recognition that even small changes in kidney function have significant clinical consequences, this taxonomy created a significant coding and reporting problem for our hospitals.
Acute renal failure, as defined by the RIFLE criteria, referred to creatinine levels three times higher than a patient’s baseline level, or to a 75% decrease in a patient’s glomerular filtration rate. In contrast, the entire spectrum of acute renal dysfunction, including increases in creatinine levels of only 0.3 mg/dl up to the acute need for continuous renal replacement therapy, is now covered by the single acute kidney injury term. (For the simplicity of my arguments, I have intentionally omitted the urine output components of these definitions.)
In the current ICD system, however, when a provider writes AKI in the medical record, the corresponding code remains acute kidney failure (ICD-9-CM code 584.9). Therefore, for a relatively small rise in a patient’s creatinine, the hospital submits a code designated as a CC. If a patient’s creatinine only increased from 1.1 to 1.4 and was only treated with two bags of IV fluids, is this really an accurate portrayal of a patient’s severity of illness? Should the hospital receive the associated increased reimbursement?
As the RIFLE criteria are now clinically obsolete, AKI should have a corresponding ICD code. In fact, AKI as defined by the KDIGO includes three stages of severity based on increasing serum creatinine levels. Therefore, we should have an appropriately corresponding ICD code for each stage. My suggestion is to create three new codes labeled AKI, Stage 1 (being neither a CC or an MCC), AKI, Stage 2 (being a CC), and AKI, Stage 3 (being an MCC). This would ensure an accurate description of a patient’s clinical situation and appropriately compensate hospitals for the varying degrees of resource use required to treat each stage. Code changes such as these fall under the auspices of the ICD Coordination and Maintenance Committee; CDI professionals and clinicians must wait for them to take action on this subject.
CDI and coder leverage
A CDI or coding professional recognizes that his or her institution’s providers attended different training programs. Therefore, they expect a certain degree of variability in how each provider establishes the presence of a certain diagnosis. However, the CDI or coding professional also needs to be charged with standardizing (as much as possible) the definitional criteria used by the organization’s providers across the entire institution. When disease definition or nomenclature changes occur, the CDI team needs to quickly implement an efficient and effective communication plan.
Additionally, many facilities work with their coders and physicians within a given specialty to regularly review definitions for common diagnoses such as those mentioned here. In such a collaborative forum, all sides are able to discuss the needs and problems associated with definition variances and establish mutually agreeable terms for the facility to use.
Ideally, the ICD system would have the involvement and agility to modify or create code sets that correspond to these changes soon after they occur. However, until that happens, CDI and coding professionals must navigate a utilitarian yet as compliant a course as possible within their facility.
Editor’s note:La Charité is a hospitalist with the University of Tennessee Hospitalists at the University of Tennessee Medical Center at Knoxville (UTMCK) and a former ACDIS advisory board member. He is also a clinical assistant professor with the Department of Internal Medicine and serves as the physician advisor for UTMCK’s Clinical Documentation Integrity Program, Coding, and RAC response. His comments do not necessarily reflect those of UTMCK. Contact him at clachari@utmck.edu. This article originally appeared in the March issue of the CDI Journal.