Don’t be Uncertain about Uncertain Diagnoses

Document an uncertain diagnosis early if consistent with clinical indicators, not wild speculation

A physician advisor recently asked me a question regarding Section II. H. from the ICD-10-CM Official Guidelines for Coding and Reporting. The Guidelines state: “if the diagnosis documented at the time of discharge is qualified as ‘probable,’ ‘suspected,’ ‘likely,’ ‘questionable,’ ‘possible,’ or ‘still to be ruled out,’ ‘compatible with,’ ‘consistent with,’ or other similar terms indicating uncertainty, code the condition as if it existed or was established.”

This is only for inpatient admissions, on the technical side of billing. Professional billing and outpatient services do not follow this rule.

Dr. Jack Sublett wanted to know” “if the uncertain diagnosis were not present in the discharge summary or the progress note on the day of discharge, is it against the Guidelines to code it?” He included a caveat that the condition under question is valid, and backed up with solid clinical indicators.

This is a rich topic. Let’s unpack it together.

I remember when I was a physician advisor, we had a specific list of what we considered to be bona fide uncertain diagnoses. “Concern for” was not included, and I was glad when Coding Clinic came out with advice in 2018 (first quarter) indicating that “concern for” was acceptable as an uncertain diagnosis phrase. I try to get providers to use as strong uncertain diagnosis words as their incertitude allows. Here is a set of uncertain modifiers that is not exhaustive: possible, probable, suspected, suspicion of, perhaps, likely, more likely than not, ruled out, yet to be ruled out, question of, questionable, concerning for, concern of, appears to be, compatible with, indicative of, suggestive of, potential/potentially.

Coders are permitted to use their discretion as to whether another “similar term” indicates uncertainty. The Coding Clinic published for the first quarter of 2014 established that “evidence of” does not telegraph uncertainty, and can be coded as definitive.

The condition must be present as an uncertain diagnosis at the time of discharge or demise. What constitutes “documented at the time of discharge?” Obviously, it is optimal for it to be found in the discharge summary in an uncertain form.

Since providers often have a window during which they can generate said discharge summary, and coders frequently get a jump on coding the encounter, tradition dictates that if the uncertain diagnosis appears in the last progress note, the condition may be picked up.

So, that means the condition would be listed in the discharge summary or in the final progress note, preferably on the day of discharge. What happens if the final progress note has the condition documented in an uncertain format, and then the formerly pending discharge summary completely sidesteps it? In such a scenario, it wasn’t ruled out, it wasn’t ruled in, and it doesn’t appear at all.

If the coding has been done and the bill has already been dropped, I doubt folks are even going to double-check the errant discharge summary to see if the condition was maintained in the documentation. If an auditor happens to review that record, they would be within their right to remove the uncertain diagnosis, since it did not materialize in the discharge summary. If this is a regular occurrence, I would advise the facility to encourage more prompt discharge summary completion and/or to withhold coding of records until the discharge summary is available.

As previously noted, Dr. Sublett also submitted that the condition was valid, and backed up with solid clinical indicators. If that is the case, the ball should be in the clinical documentation integrity (CDI) team’s court to query. Perhaps the diagnosis isn’t even uncertain anymore.

I recommend getting the provider to “evolve, resolve, remove, recap.” This means document an uncertain diagnosis early if consistent with clinical indicators, not wild speculation. Then “evolve” the diagnosis, whenever possible, to a definitive one. If the condition “resolves,” document that.

Patients generally do not have acute hypoxic respiratory failure or sepsis for their entire stay. If providers are hesitant to “remove” the diagnosis entirely from their assessment and plan list, they need to telegraph that the condition is no longer active. And then, the most important aspect is to “recap” all pertinent and relevant diagnoses in the discharge summary.

I also suggest that providers marry uncertain diagnoses with definitive signs or symptoms. For instance, one might document “chest pain, likely acute gastritis” or “anemia, possibly chronic iron deficiency.” Since on the professional side, one can only code to the highest degree of certainty, this structure gives the professional coder something to code. The inpatient technical coder can pick up the uncertain diagnosis (if it is still present at the time of discharge or demise).

Finally, remember that there are exceptions to the uncertain diagnosis rule that prohibit the coding of a condition from an uncertain format. These include HIV, Zika, novel influenza, and COVID-19. The coder would be obligated to pick up the definitive symptoms of cough and fever for the “rule out COVID-19” case.

I hope this clears up your uncertainty about uncertain diagnoses!

Programming Note: Listen to Dr. Erica Remer live when she cohosts Talk Ten Tuesdays with Chuck Buck, Tuesdays at 10 Eastern.