Knowledge Base Article

It is hard to believe it is 2017. Time flies when you are having fun and when you are not. It is also hard to believe it has been over three years since Medicare changed the definition of what supports an inpatient admission to the two-midnight rule. This occurred in October, 2013 because CMS was concerned about the number of inappropriate inpatient admissions being denied by review contractors, by the large number of extended outpatient/observation stays that had potential financial impacts for the Medicare beneficiary (co-pays and liability for skilled nursing home stays), and the inconsistent practices between hospitals for inpatient and outpatient status. The policy establishes that inpatient payment is generally appropriate if physicians expect patients’ care to last at least 2 midnights; otherwise, outpatient payment would generally be appropriate.

Unfortunately, the two-midnight policy was not the magic bullet Medicare thought it would be and a recent report by the Office of Inspector General (OIG) finds that there are still inconsistencies and issues with the application of the rule. So let’s examine what might be “right” and what might be “wrong” related to the two-midnight rule. Here I must apologize ahead of time – determining and getting a patient in the correct status is not as simple and straight-forward as this discussion may make it sound. I have the utmost respect and admiration for the physicians and utilization review staff that work very hard daily to interpret and apply Medicare’s guidelines.

Applying the rule

One thing hospital staff has struggled with since implementation of the two-midnight rule is where does admission criteria (such as InterQual and Milliman) fit in this model? The first question that has to be asked related to patients presenting to the hospital is whether they need extended care (such as beyond an ER visit) in a hospital setting and this is a good place to utilize commercial criteria. These criteria can help determine if care in a hospital setting is appropriate.

Once it is determined that care in a hospital setting is necessary, the next task is to determine if the physician believes the patient will need such care beyond a second midnight. If yes, then an inpatient admission is appropriate; if no or if unsure, outpatient with observation is likely the correct status. For inpatient admissions, the medical record should reflect that care beyond a second midnight is expected – for example, the admission orders and plan of care should support that the patient will be receiving tests and/or treatments beyond a second midnight.

And, as a hospital, if you want to be paid for your services and avoid a technical denial, make sure there is an inpatient admission order signed by a practitioner with admitting privileges prior to the patient’s discharge.

Inappropriate inpatient short stays

The OIG reported that overall inpatient admissions have decreased since the implementation of the two-midnight rule by 2.8% and short inpatient stays have decreased by 9.9%. Although this is good news, the OIG also reported that 39% of short inpatient stays “were potentially inappropriate for payment under the 2-midnight policy because the claims did not appear to meet any of CMS’s criteria for an appropriate short inpatient stay.” These accounted for $2.9 billion in payments. We must consider however that the OIG estimated the number of inappropriate inpatient short stays based on claims’ data without actually reviewing the medical records. This assessment was based on inpatient stays with inpatient-only procedures; mechanical ventilation; an unforeseen circumstance such as the beneficiary’s death, transfer to another hospital, or departure against medical advice; or a duration of 2 midnights or longer in the hospital when outpatient time prior to admission is added to inpatient time. Using only claims data, the OIG would be unable to identify appropriate inpatient admissions where the patient experienced clinical improvement after the physician documented an expectation of a 2-midnight stay. This could explain some of the volume of potentially inappropriate short inpatient stays but I understand the OIG’s concern.

Also of concern are the most common reasons for short inpatient stays cited by the OIG report: coronary stent insertion, fainting, digestive disorders, and chest pain. Again the decision to admit is complex and the admitting physician must consider several clinical factors including the beneficiary’s medical history, the severity of the beneficiary’s symptoms, and the expected care. There are patients that will require longer stays, say for coronary stent insertion, due to co-morbidities and overall risk, but most Medicare patients are able to have this procedure and be discharged after one midnight.

This is where it is critical to apply the rule correctly – at the time of admission, did the physician expect the patient to require hospital care beyond a second midnight? Does the patient’s condition and the expected treatments as evidenced in the admission orders and plan of care in the medical record support that expectation? If it is a condition or procedure that can usually be treated in less than two-midnights, does the medical record explain what is different for this patient or for this case?

Inappropriate long outpatient stays

The OIG report did find a slight decrease in the number of long outpatient stays (2.8%) but there were still almost 750,000 long outpatient stays. At MMP, Inc., we also notice that some of our clients continue to have long observation stays going beyond a second midnight. If a Medicare outpatient needs medically necessary care beyond a second midnight, then it is appropriate to admit the patient as an inpatient. This means that as an outpatient receiving observation services is approaching a second midnight, it is time to get an inpatient order or evaluate the need for continued medically necessary care (see the next section for valid reasons for long outpatient stays). These patients do not have to meet any commercial inpatient criteria to be admitted – they only have to continue to need medically necessary care in a hospital setting beyond that second midnight.

Valid reasons for long outpatient stays

But what if after evaluation it is determined that the patient doesn’t continue to need medically necessary care in a hospital setting? What if there are other reasons the patient cannot be sent home at this time that have to do with the convenience of the patient, physician or facility? This is much more common than one might think – certain diagnostic testing is not offered on weekends; testing is not completed until late in the day and the physician will not round until the next morning to discharge the patient; the patient has to wait until the next day to get a ride home from the hospital; etc. In these cases, it is acceptable to keep the patient in the hospital one more midnight as an outpatient.

However, observation services are likely not medically necessary in these cases anymore than inpatient services would be. If continued medically necessary care was appropriate past a second midnight, an inpatient admission would be correct. Therefore, there may be valid reasons for a long outpatient stay, but not really for observation services beyond a second midnight. When medically necessary care in a hospital setting is no longer needed and the patient remains due to convenience factors, the hospital should no longer report covered observation hours on the claim. At this point, observation hours should not be charged or should be reported on the claim as not medically necessary with a GZ modifier. If the hospital is ready for the patient to be discharged, but the patient refuses to leave or the patient’s physician refuses to discharge the patient, it is acceptable to issue an advanced beneficiary notice (ABN) to the patient making them financially responsible for the continued hospital care.

The last things of concern to the OIG are the continued variation in use of inpatient and outpatient status among hospitals and ultimately the financial impact on Medicare and Medicare beneficiaries. Short inpatient stays ranged from around 1% to above 5% and long outpatient stays were from 2% to above 11% between different hospitals. It is not surprising that all hospitals are not applying the rules the same, as Medicare reviewers have even struggled to get it right. This is evidenced by the starts, stops, delays, and transitions of short-stay reviews within Medicare.

Good luck to all the utilization reviewers out there. Maybe a crystal ball or Ouija board would help…