Let’s get granular…Getting down to the nitty gritty for quality improvement

So I typically have a tune stuck in my head and when thinking about my recent quality projects, the “song” that keeps popping up is to “get granular” (flashback to my 80s teen days and Olivia Newton John).

All kidding aside, the best way to design or improve quality activities and performance is to get granular. Many times when a hospital’s quality scores look bad, there is often a muttering throughout the organization that “it’s Coding’s fault.” That may be partially true in certain instances, but the real problem is a lack of systematic coordination of quality improvement efforts and communication about those efforts.

So, let’s get granular. As an example, who looks at a catheter-associated UTI? It is probably a minimum of three or four different people, including an infection control nurse, a quality improvement specialist, a CDI specialist and a coder. However, it can also be seen in second level efforts by an additional two or three people. And let us not forget the hands-on providers who are caring for this catheter-associated infection, and documenting it in the medical record. Here is where the trouble lies. Many of these staff have different interpretations of:

• What consists of a catheter-associated UTI?
• What causes it?
• What must/should be documented and especially abstracted and reported to the CDC NHSN data base?
• What is “codeable”

I recently worked with an organization that reported 44 annual cases to the NHSN but had only coded 14 instances. That was quite a gap and was related to the difference between abstracting definitions and requirements and coding requirements. This gap is concerning as we should be able to be as close as possible to the real rate in all manners that we report data. This gap would not have come to light without a granular look at all data collection processes and staff involved.

The next granular step is to determine the “how.” How do we get alignment of data collection, definitions and review processes? It takes a painfully detailed look at many elements such as:

• What types of complications or conditions will be reviewed? By whom? Concurrently or retrospectively?
• Will there be a second level review process? When (concurrent or retrospective)? By whom?
• Where will the review activity and thoughts on the case be documented? How will it be shared? When?
• How will the organization ensure all needed parties are aware of an existing complication and by what method? Concurrent or retrospective?
• Who will seek clarification if the documentation or clinical case is not clear? How? When?
• How will the provider response be communicated to all participants?

There are many more questions related to the above process, but this is a sampling to share how granular we need to think in order to improve quality.

Next, granularity is needed when there is a lack of consensus regarding a complication.

• What sort of process occurs at that point?
• Who will take the second (or third) look?
• Will we discuss our differences via email, phone, group conference call or meeting?
• What if there is still a lack of consensus? Is a third party external review needed?
• If there is still a lack of consensus, who makes the final call? And again, how is that communicated?

I have three more thoughts on the need for granularity and all relate to communication and transparency.

• Who will communicate back to the provider(s) regarding reported complications? How? When?
• How will organizational quality improvement efforts, successes and challenges be communicated? To whom? How? How often?
• How will the organization’s quality scorecard be communicated throughout the organization? By whom? How often?

As you can see from the numerous questions I have asked, getting granular is much more than mere data collection and monitoring. It’s essential to for effective quality improvement efforts and workflow. Let’s get granular!

Cheryl Manchenton is a Senior Inpatient Consultant and Project Manager for 3M Health Information Systems.