As the healthcare industry evolves, so, too, have the metrics we use to represent and monitor performance. In CDI, there seems to be one particular metric that keeps rearing its ugly head: Physician query response rates. They are still being used across the industry as a metric that is “supposedly” representative of how well your CDI program is doing and/or assessing the quality of clinical documentation.
I’m sorry to be the bearer of bad news, but if you are using this metric to represent anything other than what I mention at the end of this paragraph, then please stop fooling yourself. Save yourself the time, energy, and hassle, and remove it from your reporting bank of metrics, unless you want to gauge or measure how well leadership is doing in engaging physicians in your CDI efforts.
That’s it. Nothing more, nothing less. In short, it represents physician engagement in your CDI program.
A higher physician query response rate (I’m going to limit it to responses that are actually supporting your efforts and not the ones that go into the “swear jar”) is purely an indication that a physician views the CDI program’s efforts as an important and valuable contribution to delivering high-quality, efficient healthcare. It means that they will actually take the time to have their workflow disrupted to give you a response that may help paint a more accurate picture of their patient’s severity of illness.
First, it is not an indicator of the quality of your physician’s documentation. Second, it is not an indicator of your CDS’s performance. And finally, it is definitely not an indicator of long-term improvement in documentation habits overall. Let’s take a moment to expand on each of these points.
A Measurement of the Physician’s Documentation Performance
Using the query response rate as an indicator of a physician’s or a group of physician’s performance may make you feel better (especially if you have a 90+ percent rate), but it really has little, if anything, to do with their documentation performance.
For starters, the response rate is dependent on the generation of a query to begin with, so we can only hope that a sufficient sample size of that physician’s charts are being reviewed in order for a sufficient number of queries to be generated. Case in point: Answering one query when there is only one query is a 100% query response rate.
In addition, you may have a physician who is, overall, a very good documenter of severity in their patient mix and may ignore the only two queries that were generated for him/her in a month due to time constraints. This would make for a query response rate of 0%.
Furthermore, the physicians that are the least receptive/responsive to queries are likely to get fewer or none of them. These physicians tend to have lower quality documentation and there could be limited query response rate data on them. There are too many variables outside of a physician’s control for us to use this metric effectively to measure their individual or group’s documentation performance.
A Measurement of the CDS’s performance
I was recently shocked to learn that there are programs out there that still measure the productivity and performance of their clinical documentation specialists based on the response rate they get from the queries they generate. Once again, there are too many influences outside of the CDS’s control that could bias CDS behavior and make this metric less effective.
Would a CDS be less likely to generate a query on a physician who, historically, has been a very poor responder to their queries? There can be disincentive to submit queries to a physician whose poor response rate will end up being a bad reflection on the CDS. Not only is this approach unfair, it is protective of the physicians who actually do need help.
A Measurement of Long-Term Improvement in Physician Documentation Practices
This one seems obvious, but I will elaborate nonetheless. Actual improvement of documentation practices over time should, theoretically, see a decrease in the overall number of queries being generated (unless the reduction in queries enables you to expand the mandate of your CDI program into quality scores, etc.).
While you may be fortunate enough to see a reduction in query volume, it is not clear what effect these improvements in physician documentation practices would have on your query response rates. Perhaps, it would even reduce the query response rates as physicians feel more empowered in their documentation and are more likely to leave confusing queries unanswered.
There are few other metrics over and above the query response rates, like the number of queries for example, that are also not a good representation of documentation quality. MCC/CC capture rates are also popular and this, indeed, can be a good representation of trends in severity documentation quality, but I will provide a word of caution in a future blog.
If your response rates are high for ALL specialties and physicians covered in the scope of your CDI program, then congratulations to you, your team, and your leadership for being successful in engaging your docs.
I propose you decrease your overall query rate as a goal. If your response rates are low, your organization has some work to do in engaging your clinicians effectively. This, in my opinion, is something that can be bolstered by your clinical documentation specialists, but should not be their responsibility alone.
“Not everything that can be counted counts, and not everything that counts can be counted.” - William Bruce Cameron