Across the nation I am finding more and more medical school graduates practicing clinical documentation improvement (CDI) especially as the need for clinical accuracy grows.

Some of them I have had the pleasure of working with are hospitalists or other hospital-employed physicians practicing as physician advisors or clinical documentation specialists (CDSs). Some of them are retired physicians moonlighting as advisors. A good deal more I have worked with have become full-time consultants performing a myriad of tasks, from physician education, CDS consulting, and even working as expert witnesses and performing claim appeals. And last but certainly not least, there is a group of medical school graduates working full-time as CDSs in lieu of having completed their residency. Often that last group includes immigrants who got hung up in the red tape of making the transition to the U.S. and found practicing as a CDS the next-best way to utilize their skills and education in the healthcare environment.  

I have trained CDSs from all sorts of backgrounds, including coders-turned-CDSs, nurses-turned-CDSs, and others. I recently had the pleasure of working some medical school graduates practicing as CDSs. Comparing and contrasting the groups makes for an interesting discussion, but more importantly, it requires a nearly completely different training strategy and approach for each group.  

Let’s start with the dedicated coders. A couple years ago I did a two-week boot camp in which nearly 100 percent of the trainees had backgrounds in coding. While nurses often consider CDI to be the domain of the clinically trained and may bemoan the thought of having dedicated coders take on the task, I say not so fast. I’ll skip to the conclusion and divulge right way that the coder group was absolutely one of the most successful, teachable, and enjoyable groups with which I have ever worked. You don’t need to educate coders on what a principal diagnosis is, what a secondary diagnosis is, or really any coding rule or guideline there is – for the most part, they already have it. In a word, they are 80 percent of the way there right out of the gate. What they really needed was to have some of the dots connected for them in regard to how to interpret certain clinical indicators. When approached like a miniature medical school with a focus on how to craft the most relevant queries, you will find a very high rate of success with this bunch.

The second group to discuss is the mainstay of CDI training: nurses. The nurses I have worked with come from varied backgrounds. Some are case managers, some are unit managers, some are in education, and some came straight from clinical practice. In a mixed group of nurses, one challenge that has to be overcome is created by the nature of nursing practice itself. Often a nurse is a “med-surge nurse” or an “OB nurse” or a “surgical nurse,” etc. and as such, he or she lacks the clinical knowledge for which we just credited them. They are wizards as it pertains to the clinical issues surrounding very specific patient populations, but often will need help in other body systems not familiar to their area of practice. Contrast this to coders, who often already are used to all of the body systems and all of the common diagnoses; one could argue that the nurse has a clear disadvantage here.  Add to that the fact that you have to teach the nurse all of the basic coding definitions and all of the guidelines, making for an uphill battle the first week. The situation is more challenging than even that, however. A great deal of “unlearning” has to be done with nurses to try to prevent them from  jumping to a diagnostic conclusion from clinical indicators without the physician clearly stating it in a way that can be coded. Lastly, nurses can tend to have a bad habit of trying to recreate the entire medical record in their CDS review. In spite of all this, nurses remain the staffing pool of choice when recruiting new CDSs. Once trained, they tend to have a better grasp of the clinical indicators than the coder population, and they retain much of the advanced concepts better. Nurses also tend to be more adept at having clinical discussions with challenging physicians and getting buy-in. Just remember that teaching nurses requires a mixed approach. Many of the same pathophysiological concepts that would be standard training for the coders have to be gone over again with nurses, but the focus will be on teaching them coding rules so they can identify when a query may be needed for coding purposes.

The last group is the medical school graduates. To state the obvious, teaching pathophysiological concepts to this group is not necessary. Only with the most controversial concepts does one usually find themselves delving into clinical indicators and diagnostic criteria when working with physicians. If the physicians are new, a review of basic coding definitions and guidelines will be in order, the same as it would be for the nurses. It is noteworthy that I found the medical graduates so self-motivated that they had already mastered a fairly large chunk of coding rules on their own. So what, then, is left to teach? It is those controversial and advanced topics that will occupy most of the discussion. It’s no surprise to me that I sometimes find myself picking their brains for insight on difficult-to-solve problems, even though I am there as the instructor. Physicians are also particularly interested in the specific wording required to translate in one code set versus another, as that piece is most often what they find lacking in their knowledge base. It is a word game, of course, and when you look at some of the asinine coding rules, I will admit that I am embarrassed to be the one who has to explain the minutiae to a group of physician advisors. It is worth the effort, however. The result will always be an employee/advisor/consultant who is highly knowledgeable in the clinical area, understands enough coding to get the job done, and now has been  armed with the specific verbiage required to capture the appropriate severity of the patients being reviewed.

So, what of working with the foreign medical school graduates? One might anticipate language barriers, cultural misunderstandings, and knowledge gaps as possible challenges to be overcome. Such was not my experience, however. This group holds a distinct advantage with regard to having peer-to-peer conversations with medical staff, as is required as part of an effective CDI program. Many of the coding rules that are inconsistent and ridiculous tend to uncomfortably come to the forefront of these conversations, but I wouldn’t say that is the biggest challenge. My biggest challenge was resisting the temptation to ignore non-competition agreements and trying to recruit them.  

Now, to answer the question: “what is the hardest group to train”? Without a doubt, the hardest classroom of trainees to properly address would be a mixed group. As you can see from above, the strategies for instruction have to be tailor-made to suit the audience in order to reap the best results. When you have a mixed group, you are always going to have some occupants of the room who are getting what they need while others are bored and falling into a semi-comatose state. The group’s state at any given time will fluctuate depending on the type of material you are covering relative to each individual’s background.

The most effective approach here is to give a constant mix of coding rules and pathophysiological concepts at the same time, which is much easier to type than it is to actually do. 

About the Author

Allen R. Frady is a senior consultant for Optum360. His experience includes areas in management, implementation, education and clinical practice.  With 20 years in healthcare, he provides clients assistance in the areas of documentation, program implementation and compliance.  His background includes critical care nursing, coding, auditing, utilization review, and documentation improvement.

Contact the Author

Comment on this Article

Share This Article