A New Year, A New Approach To CDI

  • Dr. Terrance Govender
  • January 10, 2018

As we move full speed ahead into 2018, I want to take this opportunity to wish everyone a prosperous New Year! While I am personally not a fan of New Year’s resolutions, I do take time to reflect on my previous year, as many do, and set goals for the upcoming year.

Upon reflection of the CDI industry, it is apparent to me that while we have made huge strides over the years, many of them are not self-sustaining. I believe that this is largely due to the fact that we have not changed physician behavior in a sustainable manner, and as you have seen me write many times before, many clinicians have come to rely on the CDI program as a crutch rather than a safety net.

In my previous post, I outlined why engaging physicians in CDI is crucial now more than ever. In summary, this is due to the increased exposure and accountability at play with pay for performance metrics, value-based purchasing, and other risk-adjustment assessments. The Clinical Documentation Specialist will not be able to query a physician on every single diagnosis that their patient may have for which further clarification/documentation is needed. CDSs must not jeopardize their relationships with physicians by subjecting them to “query fatigue”.

What happens, then, to the diagnoses that are clinically relevant and need further clarification, but are not queried on because the documentation already supports the appropriate DRG assignment? Many of these diagnoses have HCC (Hierarchical Condition Category), SOI (severity of illness), and ROM (risk of mortality) value, and may impact other quality assessments and even reimbursement models.

You would be correct in assuming that these diagnoses are frequently never captured and get lost in the ether. Yes, many CDI programs go beyond the DRG shift, but with the sheer volume of relevant diagnoses to the whole constellation of risk-adjustment models, significant opportunity still exists, even at hospitals with the most effective programs.

At ClinIntell, our advanced analytics confirm that there is significant under-documentation of many diagnoses across the country, especially beyond the MCCs that are commonly focused on. The concurrent and retrospective query process has significant limitations to it, one of them being that our querying capabilities are limited by information on a diagnosis that already exists in the chart. If a provider fails to provide any information or clinical indicators that suggest a diagnosis, then not only do we not know that the patient even had that diagnosis to begin with, but we surely cannot query for it.

This is also where computer-assisted physician documentation (CAPD) platforms fall short. While they can help get more specificity on certain diagnoses, prompts can only be initiated based on information already in the chart. I have seen varying results and clinician feedback with these platforms.

What the New Approach Calls For

  1. Physician Participation:

    The physician can no longer be protected from practicing the core fundamentals of high-quality clinical documentation. The organizational costs are just becoming too high. In the same breath, there is great competition for a physician’s time and attention. If you consider that the query process will not disappear and is here to stay, then it behooves a physician to familiarize themselves with not only the core fundamentals of high-quality clinical documentation, but also with the most common diagnoses that drive severity of illness in their patient mix. The accurate documentation of acute blood loss anemia for an Orthopedic Surgeon comes to mind…

  2. Patient Mix Insight:

    It is not feasible for physicians to familiarize themselves with the requirements of all the high severity diagnoses that may occur in their patient mix. This is where advanced analytics, like the ClinIntell model, becomes so valuable. Gaining insight into documentation practices in your patient mix using advanced analytics not only focuses your approach, it also allows you to measure improvement and sustain performance over time. Furthermore, patient mix insights allow you to identify opportunities per physician based on their unique set of patients, where benchmarking typically falls short.

  3. Performance Reporting:

    Monitoring physician performance and comparing them to their peers internally has a significant role to play in engaging physicians. Whether it be the results of the Hawthorne Effect, or a mere appeal to a physician’s competitive nature; it works. As long as the reports are easy to read, not cluttered with metrics that mean nothing to physicians, and able to provide actionable feedback to improve documentation practices, performance will, also, almost certainly improve.

  4. True Performance Metrics:

    Historically, CDI has been peppered with various process metrics (e.g., query response/agreement rates) and metrics that lack an objective goal (e.g., CC/MCC capture rates). At the end of the day, you want to know how good your program is at delivering results, but how you achieve these results are less of a concern for hospital leadership. And CMI won’t cut it, for reasons explained here.

These are four main areas that will need more attention and focus if we are to, indeed, make a lasting difference in the CDI profession moving forward. The industry has been effective at fine-tuning the traditional query-based approach, but there is still plenty of juice left in the orange.

My next post will outline how the ClinIntell approach effectively addresses these areas.

“Insanity is doing the same thing over and over again and expecting different results.”

–Albert Einstein


Dr. Terrance Govender
VP of Medical Affairs, ClinIntell, Inc.