Interest in ambulatory clinical documentation improvement continues, as does the confusion about what such an approach can and cannot accomplish.
In a previous piece, I discussed what ambulatory CDI was not. As discussions, patterns and themes continue to evolve, now seems like a good time to think constructively about how to chart a pathway moving forward.
First and foremost, when considering an ambulatory CDI program, you should first look to achieve a solid inpatient CDI program. Here's why:
- Unlike ambulatory CDI, the inpatient side is well-defined, focused and has known content and workflows. This is where a new clinical documentation specialist—and a new CDI program—should begin.
- The inpatient arena is a great place to become familiar with effective chart reviews, the proper structure and flow of queries and the proper application of metrics.
- If properly designed, the same CDI software tool can be used both in the inpatient and in the ambulatory setting, but the inpatient setting is the best place to get familiar with it.
Once a solid inpatient CDI has been established as a base, you can then consider extending that into the ambulatory setting. This requires the following issues to be addressed:
- Recognize the areas where ambulatory CDI can help be helpful. The focus will shift from the acute conditions that are such a big part of inpatient CDI and move to more chronic conditions.
- Determine the extent to which the institution is participating in (or plans to participate in) risk-based payer plans such as Medicare Advantage. There are related issues that need to be understood and delineated here as well, including managed populations, bundled payments and “subpopulation health.” OK, I slipped that last one in on you. The term “population health” has been another on a long list of misleading jargon terms tossed around far too often. I maintain that we are more typically involved in managing the health of a defined subpopulation rather than the more grandiose concept of managing the health of the entire population.
- Recognize the many issues that may not be directly related to CDI opportunities (local coverage determination, physician evaluation and management coding, etc.)
- Develop strict definitions of exactly what you want to accomplish and what problem or problems you are trying to solve using ambulatory CDI. You will need to determine in which clinics and setting this will be implemented and what the sequence will be for implementing in those sites.
- Define the metrics you will use to guide the process. To date, programs have mainly reported out their review rates, physician response rates, etc. While interesting, these are probably not the most effective key performance indicators. Remember, a KPI should be a measurable value that demonstrates how effectively an organization is achieving key business objectives, and it is only as valuable as the action it inspires. While it can be difficult to compute manually, the risk-adjustment factor at the heart of risk-adjustment methodologies is an excellent candidate to be used to measure the impact of ambulatory CDI, whether or not the patient is covered by an insurance plan based on risk-adjusted methodologies.
- Finally, identify staffing considerations, and be prepared to be flexible with this. While the CDI industry has a pretty good handle on staffing requirements for an inpatient program, workflows and hands-on needs related to staffing for an ambulatory CDI program are less well understood. Some early ambulatory CDI programs have found the effort to be extremely labor-intensive, though that could change as software tools evolve and pivot to better support the effort.
Ambulatory CDI is an evolving practice. Start by establishing a rock-solid inpatient program and then dip your toes into the ambulatory world. The same principles apply, and if you chose the right tool, the same software can apply. If you’ve set up a successful ambulatory CDI program, I’d love to find out more about what you did.
Jonathan Elion, M.D., FACC, is a practicing board-certified cardiologist in Providence, Rhode Island, and an associate professor of medicine at Brown University.