Case Management

Q&A: Can we use DRG averages to comply with 2-midnight standards?

Case Management Insider, October 6, 2015

A lot of people still have questions about the 2-midnight rule and may wonder if their chosen practices comply with the regulation. One such question was posed to speakers, Steven Greenspan, JD, LLM, vice president of regulatory affairs at Executive Health Resources in Newtown Square, Pennsylvania, and Kurt Hopfensperger, MD, JD, vice president of compliance and physician education for Executive Health Resources, during a recent HCPro webcast “Medical Necessity Documentation and Short Stays.”
A summary of the question and answer is as follows.

Q: The average length of stay is available in our electronic system for all our common DRGs and admitting diagnoses. Our physicians are using this information to make inpatient and observation status determinations. For example, if the average length of stay for a DRG or admitting diagnosis is greater than two days they will determine that the patient needs to be inpatient. If the average stay is less than two midnights they assign a patient to observation status.

Do you have any thoughts on practice and is this done at other hospitals?

I actually have seen it done, not very commonly. The problem with doing this is that CMS expects an individual assessment of each and every patient or beneficiary to determine if the physician can reasonably expect him or her to need two midnights of care. It’s the history of the individual patient, his or her comorbidities and individual risk factors that must determine the stay expectation. Using the average stay for a given condition doesn’t take these individual factors into account. In addition, the stays listed in the system are just an average. That means that a number of patients will need a stay less than the amount listed and others more. This number might not also take into account time spent in the ED or when the patient was on observation status. So using this average as an expectation for all patients could lead to problems.

The only time this data could potentially be useful is in the extremely rare situation when a physician is so unfamiliar with a diagnosis that he or she simply has no idea what to expect in terms of stay. But even then, looking at the average stay for the condition would only be one data point out of dozens or hundreds of others that the physician should consider when determining whether a patient will likely need two midnights or longer of acute care.

Got questions about the 2-midnight rule or any other case management topic? Email us and we’ll poll our experts to find you an answer. To submit a question, email Kelly Bilodeau at

For more insights about 2-midnight rule compliance and answers to some common questions, check out the November issue of Case Management Monthly.


Most Popular