When treating a stroke or heart attack, it is said that every minute counts. But when it comes to the Centers for Medicare and Medicaid Services (CMS), it is hours that must be counted. Throughout the regulations, the issue of “How many hours?” gets raised. And despite what seems to be a simple concept, counting hours for CMS is anything but simple.
The 2023 Outpatient Prospective Payment System Proposed Rule brought us the latest in CMS adventures into counting hours. The issue of hour counting is included in CMS’ proposal to establish a new hospital type, the rural emergency hospital (REH). These hospitals, which are actually not hospitals according to the Social Security Act, section 1866(a)(1)(Y), will be required to not exceed an annual per patient average of 24 hours. A REH will be required to provide emergency services and observation services but may also provide other outpatient care, including laboratory, radiology, and surgical services.
But in the Proposed Rule, and confirmed on a CMS open door forum call, CMS has not yet determined which patients count in calculating the annual 24-hour average. Counting only patients who receive observation services will be relatively easy as each claim will have the HCPCS code G0378 indicating the number of hours of observation services received. But aiming for a 24-hour goal for such patients would be quite difficult as outpatients receiving observation services often require more than 24 hours, and REHs may be forced to transfer patients who only need a few more hours of care simply to avoid hitting that annual 24-hour mark. Including every patient, including the patients who come for outpatient services and spend 15 minutes, would be optimal since their sheer volume will dilute the effect of the observation patients who exceed 24 hours. CMS is asking for suggestions so if you have an opinion, tell CMS in the comment section.
One of the most well-known instances of counting hours applies to critical access hospitals who are required to have an average annual inpatient length of stay of under 96 hours. While the COVID-19 public health emergency (PHE) has led to that requirement being temporarily waived, it will be back once the PHE is allowed to expire.
But it is not that simple.
In order to calculate the average annual length of stay in hours, CMS must be able to ascertain the date and time of inpatient admission and discharge for every inpatient admission. And while the date and time of admission can be found on the claim form, there is no location on that form to indicate the time the discharge was effectuated. While the hospital’s internal data systems probably indicate the time of the patient’s departure, CMS does not have access to this data. To this day, I have no idea how CMS validates that a critical access hospital’s annual average inpatient length of stay is under 96 hours. Interestingly, a large group of members of Congress have asked CMS to do away with the 96-hour rule completely, arguing that the limitation is no longer appropriate.
Although CMS adopted the Two-Midnight rule in 2013 leading inpatient admissions to be viewed based on the number of midnights needed and spent in the hospital, the rule actually still relies on counting hours. That’s right. For determining admission status, once still uses the benchmark of 24 hours. But not just any 24 hours. If one goes back to the 2014 IPPS Final Rule, CMS tells us “Our previous guidance also provided for a 24-hour benchmark, instructing physicians that, in general, beneficiaries who need to stay at the hospital less than 24 hours should be treated as outpatients, while those requiring care greater than 24 hours may usually be treated as inpatients. Our proposed two-midnight benchmark, which we now finalize, simply modifies our previous guidance to specify that the relevant 24 hours are those encompassed by two midnights.” Now of course this can be best viewed as CMS sleight of hand, a way to adopt the two-midnight measurement while minimizing the need to change the regulations, but they are right that it is still counting hours.
Another interesting nuance of hour counting applies to observation services. Billing for observation services begins with the order from a qualified practitioner and ends when the necessary observation services end, but the coding team must carve out of that total any hours where the patient received a service that includes active monitoring, such as a stress test, a colonoscopy, or administration of chemotherapy. On the other hand, CMS requires that patients who receive more than 24 hours of observation services to receive the Medicare Outpatient Observation Notice (MOON). But in this instance, CMS instructs hospitals to start counting, as with the billing, when the observation service is ordered. But here CMS instructs hour counting to continue irrespective of any carved-out services. On a practical basis, this is much easier than calculating carve-out time and adjusting the time at which the MOON is due. Nonetheless, what CMS is asking hospitals to do is count as observation time that is truly not observation.
The last instance of hour counting is often the most vexing. In previous versions of the Medicare Benefit Policy Manual, Chapter 6, section 220, CMS used to state, “Services that are covered under Part A, such as a medically appropriate inpatient admission, or services that are part of another Part B service, such as postoperative monitoring during a standard recovery period, (e.g., four to six hours), which should be billed as recovery room services.” It is clear on careful reading that CMS’ inclusion of four to six hours was simply to illustrate an example of a recovery time and not to set the recovery period for every surgery as four to six hours. Yet that was how it was viewed, with hospitals obtaining orders for observation services at hour five or hour seven even if the patient was still within their expected recovery time. It is clear that the standard recovery period is what the surgeon determines prior to surgery as the amount of recovery services that will be needed by the patient, be it an hour, four hours, six hours, or overnight. The fact that CMS removed that “four to six hours” from the manual supports the removal of that time frame from our vocabulary.
In addition, the same surgery can have different recovery periods depending on the surgeon, the type of surgery, and the patient’s unique characteristics. I have heard it said that “If a surgery is not inpatient only, the standard of care is that the patient can be discharged the day of the surgery.” That is not correct. What is correct is that the payment for the surgery includes the standard recovery period so the most fiscally sound process would be to discharge the patient as soon as it is clinically appropriate. But to suggest to a surgeon that CMS says they must discharge all outpatient surgery patients on the day of surgery is not only incorrect, it is also dangerous.
Medical care has become much more efficient and hospital stays have become much shorter. What used to be measured in days is now measured in hours.
But ensuring we are counting the right hours may be the key to getting paid for that care.