Industry-wide, there is substantial interest in specialty-specific ICD-10 education for physicians. Recognizing that physicians will not, and reasonably cannot, take a broad approach to learning ICD-10 as do coders and clinical documentation specialists, education is increasingly being focused on those specific areas relevant to a given clinical practice. This is coming at a time when more and more hospital care is being coordinated by hospitalists, who typically act as the attending physician or “captain of the ship.”
Recently, a certain clinical documentation specialist asked a very thought-provoking question: What will be the impact of subspecialty physician ICD-10 training on coding practices, including issues of conflicting documentation?
Consider, for example, a hospitalist who admits a patient with a hip fracture and multiple medical problems. The hospitalist documents “hip fracture” and consults the orthopedic surgeon (who, by the way, was well-trained for orthopedic applications of ICD-10) ,and he or she documents “displaced fracture base left femoral neck, initial encounter, closed fracture.” Throughout the admission, both physicians continue to document as they had initially. Does this situation describe conflicting documentation? Clearly, it does not. First, “hip fracture” is not a codeable diagnosis. It is a non-specific clinical description. The coder should look at the entire chart to achieve the most accurate ICD-10-CM coding. Similarly, the orthopedic surgeon also may have documented diabetes somewhere in the chart, but the hospitalist provides the necessary specificity (appropriate to the specialty of internal medicine) of “Type I diabetes mellitus with diabetic neuropathy.” If coding queries were to be generated to address such conflicting documentation, the physician outcry would be predictable and appropriate. It would be unreasonable to expect every physician in every specialty to understand the documentation requirements across the broad spectrum of ICD-10. That is why multiple specialists are often involved in care.
There is, of course, another concern regarding inconsistent documentation and coding queries. Currently, it is common practice to query physicians when situations of apparent conflicting diagnoses arise. The specific example raised by one clinical documentation specialist in a recent inquiry involved a case in which “a hospitalist documented acute hypoxic respiratory failure due to COPD; (and) the pulmonologist documented hypoxia in the setting of asthma.” Current practice would be to utilize the coding query process to arrive at the most accurate definitive diagnosis. My concern is that, as ICD-10 increases pressures on coders, will they have the capacity to issue the appropriate queries, and if not, who will pick up the ball and run with it?
Technology may be able to help. Computer-assisted physician documentation and computer-assisted clinical documentation improvement (CDI) have the potential to offer the tools necessary to identify conflicting diagnoses and resolve such situations early in the patient stay. This becomes absolutely critical for patient care. In the above example, if the hospital had an “acute respiratory pathway,” it would be triggered only by the correct diagnosis identified at the time of care, not through a retrospective query. Active clinical problem lists are also a valid consideration to improve physician communication and documentation.
From a coding perspective, it is important to reiterate the need for coders to review the entire medical record carefully for details necessary to assign the most specific codes possible under ICD-10-CM.
Given the many axes of classification that often are consolidated into a single code, it is unlikely that all the specificity needed to assign a code will be neatly packaged by one provider in one phrase. More often than not, you will need to piece together the details from documentation compiled by more than one provider, contained in different document types and resulting from various interactions with the patient. It also will be important to discern the difference between truly inconsistent documentation and different degrees of specification.
Certainly, inconsistency would warrant a clarification, but would it also be necessary to clarify when multiple providers do not describe the same diagnosis to the same degree of specificity? Would it be necessary to clarify to gain consensus from all providers on the degree of specificity to be assigned? I think not. While concurrence and consistency are ideal, they are not requirements for appropriate code assignment.
About the Authors
Paul Weygandt, MD, JD, MPH, MBA, CCS, CPE, is a Certified Physician Executive (CPE) with more than 20 combined years of experience in medical management, legal counsel, and orthopedic surgery. He has served as a hospital VPMA, improving documentation across all DRG payers. Paul is vice president of physician services for J.A. Thomas & Associates (a Nuance company). He is also an AHIMA-approved ICD-10-CM/PCS trainer.
Angela Carmichael is a registered health information administrator, a clinical documentation improvement practitioner and a certified coding specialist for both hospital and physician services. She is an MBA with extensive experience in the health information management field, specializing in various payment methodologies, coding education, compliance, and management. Angela is an American Health Information Management Association (AHIMA)-approved ICD-10-CM/PCS instructor and ambassador. Angela is the Director of HIM at J.A. Thomas & Associates, a Nuance Company.