In my previous article, Aggressive Tactics by Third Party Auditors Should Make Providers Vigilant, I suggested that one of the actions to prevent unjustified clinical validation denials (CVDs) is to perform concurrent clinical documentation integrity (CDI) reviews and pre-bill audits. Permit me to expound on the role of CDI reviews in preventing DRG CVDs.
When I was a physician advisor for a large multi-hospital health system, a CDI specialist (CDIS) might come to me with concerns regarding a specific provider overdiagnosing some clinical condition. It was usually a pattern, as opposed to an isolated instance of making a single diagnosis which the CDIS felt was not clinically supported.
The conditions which were most often brought to my attention, not surprisingly, were the same conditions which most frequently elicited CVDs. Overzealous diagnosing of acute respiratory failure, sepsis, encephalopathy, and acute kidney injury (AKI) were the most common culprits. Is it acceptable for us to question a physician’s diagnoses?
There is something in the wording of the 2017 edition of the ICD-10-CM Official Guidelines for Coding and Reporting, Convention 19 that unsettles me. It states:
The assignment of a diagnosis code is based on the provider’s diagnostic statement that the condition exists. The provider’s statement that the patient has a particular condition is sufficient. Code assignment is not based on clinical criteria used by the provider to establish the diagnosis.
I think it would have been better stated like this:
The assignment of a diagnosis code is based on the provider’s diagnostic statement that the condition exists, and there should be clinical evidence supporting that conclusion. Coders may not base codes solely on clinical criteria without the provider explicitly documenting that the patient has that particular condition.
The point I am trying to make is that a coder is not permitted to infer a diagnosis from clinical criteria, but I also do not believe that a provider documenting that a condition is present, without supporting evidence, is “sufficient.” I concur with AHIMA’s compliant query recommendations, which state:
When a practitioner documents a diagnosis that does not appear to be supported by the clinical indicators in the health record, it is currently advised that a query be generated to address the conflict or that the conflict be addressed through the facility’s escalation policy.
Often when asked about their diagnoses, confident, proficient physicians do not perceive it as a challenge but as a request for clarification. Personally, I would relish the opportunity to reconsider my diagnoses to be sure I hadn’t missed anything. Documentation is intended to be clinical communication, and if another healthcare provider (HCP) is confused or skeptical, explanation should be forthcoming to ensure quality medical care. The crux of the matter is whether the non-physician reader (nurse, CDIS, auditor) is viewed by the clinician as competent to judge their diagnoses and documentation. Not all clinicians are equally expert, and not all non-provider readers are equally knowledgeable either.
On occasion, the issue is that the provider is neither confident nor proficient, and they feel legitimately threatened. I would posit that if clinical judgement is suspect, the documentation often reflects that. The CDI program is not the venue to address this. Incompetency should be referred to quality and/or medical administration.
For the other cases, the first line of offense is the CDI professional. Most of them are nurses with clinical experience, although I have worked with some extremely capable CDISs who had been certified coders first. The exchange petitioning for improved documentation is facilitated if there is a long-standing respectful working relationship between the CDIS and the provider. One of the main barriers is the prohibition against leading the provider. If the practitioner is unclear as to what is being asked of him, it can be hard for him to answer a query satisfactorily.
My recommendation for this is to have a library of CDI tips explaining clinical criteria or templated queries with non-leading information which give the provider context to make an intelligent decision. A face-to-face encounter between the CDIS and the provider can result in a meaningful dialogue which can serve the same purpose. My experience is that most of the time, similar to a person proofreading his own essay, the author thinks his documentation says something different than what the reader is gleaning.
Once there is escalation, I recommend that the physician advisor do several things:
- Although the practice brief refers to it as a “conflict,” approach the provider in a collegial fashion. In the beginning, I initiated contact itching for a fight, and almost always was shut down by the HCP who was genuinely interested and open to our conversation.
- Try to ascertain why the HCP is making the diagnosis. It is usually one of several issues:
- The diagnosis actually is appropriate and the provider just need to provide additional corroborative data.
- The diagnosis is erroneous and the HCP is willing to clarify that position.
- The HCP has a knowledge gap and you need to educate him or her.
- The HCP is under a misguided impression that he has been instructed to exaggerate how sick a patient is to increase his quality scores.
- The reaction to all of these is to educate, educate, educate.
What do they need to know?
- First and foremost, the primary message must be, “Tell the truth!” I usually follow that with “Tell the story.”
- Providers need to understand that their documentation serves to demonstrate medical necessity. A patient must be represented as sick enough to justify their admission or observation status.
- Documentation should be consistent. Documenting the patient as obtunded and confused in the emergency department (ED), and then noting he is alert, oriented, and cooperative in the history and physical without reference to the encephalopathy in the ED which resolved, sets up a contradiction which auditors will seize upon.
- Coders are permitted to code from a single notation of a diagnosis, and they are encouraged to optimize metrics. Physicians may not understand that entertaining a diagnosis which will increase the weight of a DRG may lead to a coder coding that diagnosis, even if it isn’t actually present, if the provider never documents that the condition was ruled out. Best practice is to see a diagnosis multiple times: upon initial diagnosis, when being treated or on resolution, and then recapped in the discharge summary.
- Some conditions have relatively strict criteria (an increase of serum creatinine ≥ 0.3 mg/dl over a 48-hour period) and some are more loosey-goosey (acute blood loss anemia has no strict decrease in hemoglobin guidelines), but clinical judgement always takes precedent. Coders and auditors cannot read your mind, however; if you are practicing the art of medicine and are making a diagnosis that requires explanation, explain it!
- Best practice is to make diagnoses consistent with those which the average, prudent provider in your community would make. I was requested to have a discussion with a pulmonary-critical care attending who diagnosed every patient on two liters of oxygen as having acute hypoxic respiratory failure. Don’t be intimidated by the fact that you are educating a provider on a condition which should be squarely in his wheelhouse if he is not meeting the standard of care (or the standard of documentation). This was one of those cases where he had misinterpreted previous queries and was generalizing inappropriately. Fortunately, he was receptive to my education and changed his behavior.
In my opinion, the crucial component is to provide feedback for HCPs regarding CVDs. Most folks think no news is good news, and if they aren’t informed of lapses in their documentation, they think they are doing it right. It is tedious to teach an entire hospital staff one provider at a time. Find a mechanism (or multiple processes) to disseminate the lessons you have learned to the entire medical staff. If you have no physician advisor, authorize your CDI staff to do educational interventions, not just individual queries. However, if their education includes risk adjustment or financial impact, it should not be done in conjunction with any given query.
The bottom line is an ounce of prevention is worth a pound of cure. Clinical validation denials are particularly exasperating if your diagnoses are valid, and detailed documentation may ward some of them off. Apprise your CDISs to be on the lookout for the denial diagnoses du jour as they scour the medical record for query opportunities. Empower them to query when they believe the clinical support for a documented diagnosis is absent. Have them escalate if they are meeting excessive resistance. And educate your providers so their documentation is airtight and easily defensible.
As always, the message should be to take excellent care of your patients, document it well, and let the quality metrics and reimbursement reflect it.