Medical Necessity: Realities and Hypotheticals

While everyone is familiar with the notion that doctors practice medicine, did you know it extends to administrative decision-making as well? Hospital utilization and billing concepts frequently boil down to two often conflicting questions: what can be done, per Centers for Medicare & Medicaid Services (CMS) rules, regulations, and insurance contracts, and what should be done to provide the most effective care for the patient? These questions represent the proverbial shades of grey we all dread, but must be prepared for. At the recent National Physician Advisor Conference hosted by the American College of Physician Advisors (ACPA), two such cases were highlighted during a panel discussion involving the ACPA’s Observation Committee. 

The first scenario involved an elderly patient covered by Medicare with profound co-morbidities and significant functional decline over the last year. He presented to an emergency department in the evening from his assisted living facility with end-stage organ failure related to sepsis. Discussion between the patient and admitting hospitalist revealed that the patient had been offered hospice care in the past, but declined. Now, he thought otherwise, but wished to say goodbye to multiple family members who were in the process of traveling into town from around the country. The patient requested continued treatment for his ailments, including the sepsis, until then. A midnight had passed while the patient was in the emergency department, and family members were not expected to arrive until the following day, after a second midnight passed. 

Per the Two-Midnight Rule, a patient who requires hospital care for two midnights or longer is appropriate for inpatient billing. But does it apply in this case? Is hospital care waiting for family members to arrive to say goodbye before transferring to inpatient hospice necessary hospital care? On the one hand, hospital care is hospital care, and active treatment for sepsis DOES fit the bill. But if the treatment is simply performed to buy a day or two and is not intended to be curative, does that count? Does the reasoning for the care always have to be considered? While it’s clear that a patient who is ready for discharge on oral antibiotics no longer needs IV antibiotics, even if the patient requests it, this scenario is more nuanced. 

The second case involved a patient covered by Medicare who had received a lung transplant a few years before and presented to the GI lab for a planned colonoscopy. During check-in, it was noted that her pre-procedure testing for COVID-19 was positive. While the patient was completely asymptomatic, her pulmonologist advised IV Remdesivir treatment for three days, given the significant risk of infection due to her immunosuppression associated with the transplant. However, the hospital was not equipped to provide this daily treatment in the outpatient setting, so the patient was transferred from the GI lab to an inpatient unit for hospitalization. 

Once again, at face value, the Two-Midnight Rule could be used to support inpatient status here. The patient was anticipated to require two midnights of care, receiving an IV therapy while being monitored for clinical deterioration. But two questions need to be considered. First, does the hospital’s inability to provide Remdesivir in the outpatient setting support inpatient care? What if the hospital was also unable to perform stress tests over the weekend? Would chest pain patients be appropriate for inpatient care if they presented Friday or Saturday? Few would agree with that conclusion. Second, remember, the patient was asymptomatic. The COVID-19 infection could ultimately manifest into significant, even life-threatening symptoms, which would require prompt medical attention, but was hospital monitoring necessary, given the patient’s current condition? A post-op patient who is monitored in the hospital for recovery complications following an outpatient procedure remains in outpatient status unless the clinical condition actually does deteriorate, meaning the patient then requires at least two midnights of hospital care. What if the transplant patient never manifested symptoms and discharged from the hospital after three uneventful days? 

From my standpoint, the first case is appropriate for inpatient billing because the patient is indeed receiving medical care to treat his acute condition (sepsis), and that care can only be provided in the hospital setting. It does not matter what his intended treatment plan is in the future, after his family members arrive; the current plan is treatment that will surpass two midnights. Also, who’s to say that the patient won’t change his mind and decline hospice, electing to continue treatment, after further discussion with family? On the other hand, I believe the second case is appropriate for outpatient care without observation services. Yes, there is a legitimate concern by the pulmonologist for the patient, but two points in the case point to a lack of medical necessity for hospitalization. One, the only reason the treatment is taking place in the hospital is because they have not established a process to provide it in the outpatient setting, as many other hospitals around the country have done. Two, the patient is asymptomatic, and her infection was found incidentally. What about every other transplant patient in the community? Are they being randomly checked for COVID-19 and hospitalized for treatment even if clinically, they are well? If not, then this case falls into the “what if” and “just in case” category often seen with post-op patients who are hospitalized after an uneventful outpatient case to monitor for potential recovery complications. 

My challenge to you lies in considering the varying shades of grey. In a world where process standardization is imperative to maintain health system viability, but the practice of medicine remains an art, your utilization management teams need broad understanding of medical necessity and billing rules, a nimble core of decision-makers, and established methods to track the results. Without them, your medical and utilization staff will be trapped in perpetual loops of indecision and uncertainty as each challenging scenario arises. While no health system can prepare for every unique case, having a standardized process of escalation, a framework of decision trees based on regulations, and a record of case studies addressed in the past will serve you well.