Q&A: Prior encounter information in queries

Q: Regarding the role of prior encounters in queries, is there a recommended timeframe for how long ago the prior encounter should occur to be used in a query? If a patient’s last ECHO was three years ago, should that information from a prior encounter still be pulled into a query? Is there a general point in time when a prior encounter should no longer be referenced? Is there any consensus on if a prior encounter is too old to reference as information may be outdated?

Response #1: We follow the guidelines laid out in the ACDIS/AHIMA Guidelines for Achieving a Compliant Query Practice—2022 Update, which does not indicate a timeframe, but instead says that the information can be used in a query as long as the condition is relevant and meets the MEAT (monitored, evaluated, assessed, treated) criteria.

Response #2: I find that depends on the query. To determine if, for example, a lab finding is acute or chronic, or even if the patient is at their baseline, you may have to review labs from a year ago or longer to identify changes. For other conditions, however, I would not go back more than six months. My rational for this (and it is not written down as policy) is that if a patient is being seen routinely for an issue, you should not have documentation older than that. Older documentation more than likely indicates a chronic, stable condition with routine follow-up.

Response #3: At my previous facility, we would look at labs within 30 days. Diagnostic exams such as ECHOs would be valid for one year. 

Our leadership team worked with the providers to define the timeframe of what was acceptable. We also had to have justification for looking back. For example, if a patient had an unspecified history of congestive heart failure (CHF), then we could look back to construct a query. 

Response #5: We do not have a time limit on prior encounters. In the example you gave, however, the ECHO may have changed over that period of time. In the CDI query, when using information from prior encounters, it should be clear to the provider when and where the clinical indicators were found. You may also want to consider discussing the relevance of older information with your cardiology physicians. There may be some situations where this information is relevant and others where it is not. This would allow you to develop some guidelines.   

Response #6: In our health system, it would depend on the information you’re looking at to add support to a current CDI query. For example:

• If it’s for kidney function/acute kidney injury: It would be appropriate to look back one year for a baseline in a patient without chronic kidney disease.
• If it’s for an ECHO for CHF specificity: It would be appropriate to look back one year, possibly longer depending on the scenario, for that last ECHO EF result.
• If it’s for HIV status: There would be no limit to how far back would be appropriate to find that documented information for serology or conversion to the B20 code.

Editor’s note: This question was answered by members of the ACDIS CDI Leadership Council and was originally published in the CDI Leadership Insider, a monthly newsletter exclusive to Council members. For the purposes of this article, all Council member answers have been deidentified.