Is Knowing Clinical Criteria Sufficient to be a CDIS?

The practice of medicine takes clinical criteria into consideration, but clinical judgment can override criteria.

As I was deleting emails one morning, I saw a picture of Cynthia Tang, as in Pinson & Tang, authors of the annually updated CDI Pocket Guide.

I had never met her, so I was surprised. It used to happen to me a lot in the emergency department – you talk to a doctor almost every day and formulate a picture in your head of what the person behind the voice would look like, and then the first time you meet him at a work event, you are shocked to see that he has a bushy red beard and is toweringly tall.

I then noticed the title of her article, which really turned out to be more of a plug for their CDI Pocket Guide, “Positioning Coders for CDI Success.” Let me preface this by stating that some of the best clinical documentation improvement specialists (CDISs) I have known came from the health information management (HIM) world and were not nurses. Ms. Tang said, “coding specialists, not just nurses, can and should perform the CDI function.” She further stated that “just as CDI specialists benefit from learning coding guidelines and principles, coding specialists can and should learn the clinical criteria.”

I agree that the best CDI specialists understand and stay current on coding guidelines and principles. They also have excellent relationships with their coding colleagues and are not reluctant to ask for input or explanations of the coding rules. I also believe that coding specialists can and should learn clinical criteria. Again, it makes them better at their job.

But clinical criteria are only one facet of clinical medicine. If we were really able to distill medicine down to clinical decision tools, we wouldn’t need clinicians. Watson, the supercomputer, really could replace doctors.

However, that is not the case. There are some diagnoses that are made strictly by a positive test or by checking boxes in a set of clinical criteria. An X-ray with a fracture line indicates a broken bone. A urinary tract infection (UTI) that grows out E. coli >100,000 CFU signifies an E. coli UTI. A creatinine reading of 5.0 indicates renal failure (be it acute or chronic). Rheumatoid arthritis requires inflammatory arthritis of three or more joints, positive rheumatoid factor, duration of more than six weeks, elevated CRP or ESR indicating an inflammatory process, and inability to attribute symptoms to a different condition. Even with distinct decision points, there is often room for exception. And even with distinct diagnostic criteria, there is often room for clinical judgment.

The best example I can offer is sepsis. Sepsis is not a disease process that can be diagnosed by a set of abnormal vital signs and laboratory tests. The Sequential Organ Failure Assessment (SOFA) score is not the definitive means to make the diagnosis. Sepsis is a clinical syndrome recognized by competent clinicians trying to prevent out-of-control infections from resulting in death. It has a multitude of presentations, the commonality being organ dysfunction due to a dysregulated host response to a localized infection. Learning and scouring for clinical criteria may capture some of the cases, but it will miss others that require a keen clinical eye.

Can coders have a clinical eye? Absolutely. Ask Lisa Lorenzi and Kathy Murchland, ex-colleagues who are shining examples of coders-turned-savvy-CDISs. It takes the ability to read charts and recognize patterns, an inquisitiveness to encourage clinicians to share their thought processes, and it takes experience. I do not believe that a newly trained coder without previous medical exposure (e.g., was a paramedic in a previous lifetime) could serve as a CDIS. I believe it takes some years of scrutinizing medical records, reading literature, and trying to glean the story from the documentation. I think the best way to learn is from reviewing cases and having a clinician explain what they often don’t document. Why did you think that? How did you know that the condition was present or ruled out? What were you looking for when you did this test? Why didn’t you think they had that disease?

CDI is really about reading between the lines, finding the diagnoses that are intimated, but not asserted in the record. It is about following the narrative and timeline, and clinically validating whether the documentation and diagnoses make sense. When I do a chart review, my brain is trying to piece together the clinical indicators to make hypotheses, to see if I can deduce what is going on, and then see if the documentation confirms it, refutes it, or just misses the boat. This is not easily taught, nor can it be written down in a 250-page “CDI Bible.”

Coders who want to be CDISs can do several things to achieve their goal:

• Use clinical criteria as guidelines, not indisputable rules. Recognize that there are exceptions, and try to understand why, in this case, the providers didn’t think the indicators were consistent with that clinical condition, even though the clinical criteria would have suggested it was present.
• Talk to providers and nurse CDISs to have them explain their thought processes. Have them present clinical topics at your staff meetings. Attend society conferences where clinicians are presenting. Read articles that have case studies or case reports. Establish a relationship with a provider or physician advisor, and bounce questions off them. See if you can join elbow rounds on occasion.
• Examine the chart like a reader, not like an editor or proofreader. Try to understand the story; don’t just be scanning for missing CCs or MCCs. Don’t just accept or reject computer-assisted codes.

The practice of medicine takes clinical criteria into consideration, but clinical judgment can override criteria. There are many useful reference books and websites, but they are not going to make anyone a clinician; it takes understanding of the pathophysiology and keeping the patient front and center at all times. Clinical documentation integrity specialists, be they nurses or coders, would do well to remember this.

Programming Note: Dr. Remer co-hosts Talk Ten Tuesdays with Chuck Buck every Tuesday at 10 a.m. EST.