Health Information Management

Q&A: Assessing CDI program success metrics

CDI Strategies, August 15, 2013

Want to receive articles like this one in your inbox? Subscribe to CDI Strategies!

Q: I have come across an ethical dilemma. We have a small CDI program and a “home grown” application we use to report metrics to the chief financial officer (CFO). In this, we track whether a CDI specialist’s query captured a CC/MCC. If it is the first and/or only CC/MCC it gets counted for the financial impact of their job even if other CC/MCC later factor in.

I don’t feel this is appropriate, so, for the last few months, I have not counted those as financial impacts in favor of only counting those which truly show a difference. I predict there may be a drop in financial impact and that I will need to explain the reasoning to our CFO. Any advice you could offer would be greatly appreciated.

A: I would start by asking your CFO how he/she wants to measure the success of the CDI department and share with him/her the various potential metrics available and the strengths and weaknesses of each.

Many organizations want to see a return on investment (ROI) to “justify” the CDI department and its staff, but the value of CDI extends beyond direct revenue capture. For example, the data used to determine a Hospital’s Value Based Purchasing (HVBP) score is based the assigned principal diagnosis. A CDI specialist can therefore potentially impact revenue through the organization’s HVBP scores when they chose CHF as the principal diagnosis rather than acute respiratory failure if both meet the definition of a principal diagnosis. Personally, I think it can be tedious to count the money associated with every query, which can negatively impact your productivity, but if that is how your CFO wants to measure the success of the department, so be it.  

That said there should be uniformity to the data you collect, so information regarding financial impact of queries should be captured whenever a query is issued. Perhaps a better metric would be “potential” financial impact vs. “actual” financial impact. I think this metric is sensitive to the situation you describe because the query initially appeared to have a financial impact, but did not after the physician (without a query) documented another diagnosis valued as a CC or MCC which therefore discounted the value of the initial query. 

At some point, hopefully, providers will become educated and no longer require querying on particular diagnosis. For example, providers used to write the diagnosis “urosepsis,” but now specifically document a UTI or sepsis from a urinary source after repeated education and query efforts. Does the lack of queries in this situation mean the CDI team is less successful since they no longer get to capture the change in reimbursement based on querying for “urosepsis?” Hardly. Regardless of whether the CDI department reports to the CFO or the HIM director, or the director of case management, administration needs to understand how query rates and metrics will fluctuate over the life and growth of an effective CDI department.   

Some CDI departments use CC/MCC capture rates rather than the financial impact tied to a specific query. PEPPER data provides an organization’s CC/MCC capture rate for both medical and surgical DRGs with each quarter as a separate data point over the course of a rolling two years. This can be a particularly helpful metric, especially if the query results in a diagnosis classified as a CC or MCC being added to the health record. Additional diagnoses can influence severity of illness (SOI) and risk of mortality (ROM), which can affect the facility’s mortality index and profiling even if the claim is not paid under the APR-DRG grouper so there is value in capturing all CCs/MCCs. 

Additionally, querying is a time consuming task. I like to give the CDI credit for the time they spend on the query process, so I would want to capture the query effort even if it doesn’t result in a financial impact as measured by the capture of a CC/MCC or a change in the principal diagnosis.  In my opinion, the value of CDI is impacting the documentation behaviors of a provider, which is most effectively accomplished through interaction with the provider (e.g., queries and educational efforts) rather than just “pre-coding” records.

 If you want a true measure of how CDI is changing provider documentation through any possible means (not just through querying) look for occurrence rates of problematic or incomplete diagnoses and make a concerted effort to decrease those occurrence rates. For example, if 40% of your patients are discharged with congestive heart failure (CHF) without further specificity as systolic or diastolic or combined, then set a goal to decrease the occurrence rate of CHF to 20% in six months and then measure again in six months to see if you have reached that goal. 

You may find the recent release of the 2013 Physician Query Benchmarking Report interesting. It lists a number of additional items worth tracking and illustrates what items other CDI programs are analyzing to show their program success.

Additionally you may want to take a look at these related articles:

Editor’s Note: Cheryl Ericson, MS, RN, CCDS, CDIP, AHIMA Approved ICD-10-CM/PCS Trainer, CDI Education Director for HCPro Inc., answered this question which was originally published on the ACDIS Blog. Contact her at For information regarding CDI Boot Camps offered by HCPro visit

Want to receive articles like this one in your inbox? Subscribe to CDI Strategies!

Most Popular