What Should My CMI Be?

  • Dr. Terrance Govender
  • August 09, 2018

While the CDI industry has started to see a progression towards understanding that an actual improvement in the quality of clinical documentation is not necessarily measurable by CMI alone, especially in the short term, there are still many CDI programs that are held accountable by even a long-term decrease in CMI. Many corporate leaders to whom CDI frequently reports to not only lack a firm understanding about the relationship between clinical documentation quality and CMI, but may sometimes incorrectly hold CDI accountable for unfavorable changes in CMI.

What Is Beyond Our Control In CDI?

The CDI function does not have control over the types of patients that happen to walk through the door and that are admitted to the hospital for care. This “patient mix” phenomenon is frequently not controlled for and/or considered when the benchmarking methodology for determining CDI goals and performance is adopted. CDI also has little to no control of the service lines and/or types of services provided by an organization. Seasonal changes in patient mix and volume are in addition to, not under, the control of the CDI team. To the knowledge of many, these factors have a significant influence on the reportable CMI of a health system or facility.

What Can CDI Control?

CDI has the capability to ensure that, based on the patient mix admitted during a specific time of year and to a specific service line, the full patient severity is coded through concise, clear, and accurate clinical documentation. This is traditionally measured through MCC and CC capture rates/trends, as it affects the reported CMI. It stands to reason, then, that you could have a situation where CMI goes down, but the capture of MCCs and CCs was actually sustained or increased during that same time period. This type of scenario should be readily recognized by the program and clearly communicated with leaders to whom CDI reports to.

Benchmarking CMI

There is a place for using benchmarking to determine what your CMI should be, but when it comes to CDI opportunity and performance, it can be very misleading. Consider the following CMI trend:

CMI Trend Analysis 1

Note that the “benchmark CMI”, however determined by the organization, whether as a benchmark of itself from the previous year or compared to other organizations considered to be “high performers”, can portray CDI in a poor light for many of the months during this chosen time period. May 2017 to June 2017 particularly stands out, when CMI sharply declined. When taking a look at what your MCC and CC capture trends did during this time, this is the type of scenario that may help make a case for a declining CMI that is not due to poor CDI performance.

ClinIntell’s advanced analytics control for your patient mix and other factors monthly, allowing clients to know what the true opportunity or CMI gap is for the health system, hospital, specialty group, or even individual physician, based on the analysis of claims data. Unlike benchmarking, it does not assume a static patient mix, which can be very misleading when trying to determine what your CMI should be. Our clients can be confident that if there is a widening of the “gap”, it implies a decline in CDI performance through “missed” MCCs and CCs.

CMI Trend Analysis 2

In this view, one notes that even though CMI sharply declined from May to June, documentation quality, as measured by the capture of MCCs and CCs on patients who had them, held the line, and the gap remained at 0.10. Using that same methodology, the client is also able to observe a decline in CDI performance in the setting of a CMI that increased from March 2018 to April 2018, due to a widening of the gap from 0.10 to 0.12 during that same time period.

ClinIntell’s clients have an added advantage with our headline metric: Documentation score – a percentage metric that represents the CMI gap. A closure of the gap indicates full severity reporting of the patient mix for that month and corresponds to a documentation score of 100 percent. In the trend graph above, while it may be difficult to pick up a trend in the CMI gap, documentation score gives the organization immediate feedback on the output or CDI performance during the time period in question.

As mentioned earlier, even though CDI has significantly evolved to appreciate that a true improvement in documentation quality goes beyond the measurement of CMI performance alone, we are frequently held accountable for fluctuations in CMI. Through advanced analytics, ClinIntell’s clients can quickly determine whether those fluctuations were due to CDI performance or a change in the patient mix.

Next time, we will talk about how, through advanced analytics, you can determine whether the breadth and depth of documentation is also being captured, even if your organization is performing well with a high documentation score (small CMI gap).

“A point of view can be a dangerous luxury when substituted for insight and understanding.”

-Marshall McLuhan, Canadian Communications Professor


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