Rebooting “Observing the Rules for Observation After Outpatient Surgery”

Written in March 2013 by the highly respected Dr. Steven Meyerson, “Observing the Rules for Observation after Outpatient Surgery” evolved over time into one of the most-read articles in the history of RACmonitor. This point became extremely relevant to me earlier this year, when my manager of case management rushed into my office, papers in hand, hot off the printer.

“Read this article!” she exclaimed. “I think we might need to change the way we think about patients who are hospitalized after pre-scheduled procedures.”

Reading through the piece (admittedly, for the first time – forgive me, Dr. Meyerson!), it very quickly became apparent that the content should have already been part of my modus operandi, as a physician advisor for my health system. But it was also very clear why my department manager would have no knowledge of this. Which brought the concern to my mind: “What about RACmonitor’s other followers? How will they know what’s happening here?” 

Happily, the publisher of RACmonitor, Chuck Buck, along with Dr. Meyerson himself, gave me their blessing to update this classic article, which contained excellent direction and advice before the Centers for Medicare & Medicaid Services (CMS) Two-Midnight Rule came into existence just over 10 years ago. In the following paragraphs, you will see original paragraphs from Dr. Meyerson’s article in plain text and my remarks in bold italics.

The rules governing the use of observation for patients undergoing scheduled outpatient procedures are quite different from those that apply to patients coming in from the ED (emergency department) with undiagnosed symptoms or urgent conditions. Placement and billing errors are common, so it is worth reviewing the subject at this time.

This remains true, but the greater point in 2023 is that the finer points of conversion from outpatient to outpatient with observation services for a scheduled postoperative patient require much more deliberate education for physicians, nurses, and case managers. More on that in a bit.

Actually, it’s quite simple: as a rule of thumb, the only time observation can be used for a patient having a scheduled outpatient surgery or procedure is when there is a postoperative complication that prolongs routine recovery. But like most things Medicare, it’s never quite that simple – so read on.

Today, post-Two-Midnight Rule, this is exactly the only time observation can be used for a patient having a scheduled outpatient surgery or procedure – when there is a complication either during the procedure or during the recovery period. While a prolongation of the recovery sometimes comes into play…it’s not a necessity. Anything that requires care above and beyond what was originally expected or anticipated makes a change from outpatient to outpatient with observation services appropriate. This could involve something as simple as nurses performing vitals every two hours instead of every four. The point is that more care or services are being provided to the patient than per the usual routine. Something as subtle as this can sometimes be hard for people to recognize, which is why it is important to be vigilant so observation hours are captured appropriately. 

According to the Medicare Claims Processing Manual (Chapter 4, Section 290.2.2), “hospitals should not report as observation care services that are part of another Part B service, such as postoperative monitoring during a standard recovery period (e.g., 4-6 hours).” While this section mentions “4-6 hours” as an example of a standard recovery period, it is important to note that this time frame only is used as an example and does not set an upper limit on the time permissible for recovery. In other words, it would be inappropriate to place a postoperative patient into recovery just because an arbitrary six-hour time period has elapsed. CMS explains that a patient having an outpatient procedure may be expected to stay up to 24 hours: “When patients with known diagnoses enter a hospital for a specific minor surgical procedure or other treatment that is expected to keep them in the hospital for only a few hours (less than 24),” the agency has indicated, “they are considered outpatients for coverage purposes regardless of the hour they came to the hospital, whether they used a bed, and whether they remained in the hospital past midnight.” (Medicare Benefit Policy Manual, Chapter 1.) In this case, “less than 24” does define a time limit on an expected outpatient stay, and this clarifies that a patient’s stay may be overnight in a hospital bed in outpatient status – and still qualify as recovery.

Chapter 4 of the Medicare Claims Processing Manual was last revised on Dec. 22, 2017, but changes were not made to Section 290.2.2. It continues to include the direction mentioned above about not reporting observation services hours in cases of “postoperative monitoring during a standard recovery period (e.g., 4-6 hours).” Again, like before, the timeframe of “4-6 hours” is not a guide on how long recovery should be considered “routine” following a procedure. If one of your surgeons feels “routine recovery” following a procedure involves hospital care for 48 hours, that patient should remain in outpatient status for 48 hours. But you’ll probably want to sit down and figure out why she considers this prolonged time period routine and necessary. Is it truly a best practice and standard of care? Or is it simply the way she’s practiced for the last 25 years?

On the other hand, the Florida fiscal intermediary in 2003 explained that “if the physician intends to keep the patient overnight, especially for 24 hours or more of care at an inpatient level of care (prolonged monitoring given co-morbidity, frequent laboratory studies, frequent IV therapy, etc.), then the physician should schedule an inpatient admission. Also, it is important that physicians document the indications for the procedure and the associated co-morbidities since the medical necessity of the procedure as well as the need for the overnight stay can be reviewed by the QIO (Quality Improvement Organization).” (Florida Medicare A Bulletin, third quarter of 2003.)

This is likely the first paragraph that piqued my manager’s attention, and your first clue that something is awry. Reference to “24 hours or more of care” and “inpatient admission” no longer go together, post-Two-Midnight Rule. Also, with the exception of total knee arthroplasty, as described in the much-obsessed-over 2018 Outpatient Prospective Payment System (OPPS) Final Rule, prolonged monitoring given co-morbidities and/or feared complications that might rear their ugly heads do not support starting with inpatient – or even counting observation hours. For example, a surgeon might elect to keep their patient with a history of asthma who is post-laparoscopic appendectomy in the hospital overnight due to a concern of bronchospasm post-intubation. But observation hours should not be counted unless the patient actually develops respiratory issues that require additional assessment or treatment. If the patient’s recovery overnight is unremarkable, the status should remain outpatient without the addition of observation hours.

In the April 7, 2000 Outpatient Prospective Payment System (OPPS) Final Rule (65 FR 18455), in explaining the criteria for selection of procedures for the Inpatient-Only List, CMS noted that “the inpatient list specifies those services that are only paid when provided in an inpatient setting because of the nature of the procedure, the need for at least 24 hours of postoperative recovery time or monitoring before the patient can be safely discharged, or the underlying physical condition of the patient.” Thus, one of the criteria that distinguishes inpatient surgery from outpatient surgery is an expected length of stay of up to 24 hours for outpatient procedures and greater than 24 hours for inpatient. This offers additional evidence indicating that there is no need to change patient status for an overnight stay following uncomplicated outpatient surgery when the stay is expected to be less than 24 hours. Notice that in both cases, whether discussing the expected length of stay for a patient having outpatient surgery or one having an inpatient procedure, it’s the physician’s expectation at the time of admission that determines the proper level of care, not the actual length of stay, as viewed in retrospect.

In the 2018 OPPS Final Rule, CMS refers back to their 2012 OPPS/ASC Final Rule for discussion on how they identify procedures that are “typically provided only in an inpatient setting” and therefore are on the Inpatient-Only List. But, in true CMS fashion, looking at the 2012 Final Rule, you’ll find it references the April 2000 OPPS Final Rule. This is the same discussion Dr. Meyerson references about, “the need for at least 24 hours of postoperative recovery time or monitoring before the patient can be safely discharged.” But it’s important to recognize this is not a direction to place a procedural patient who has an anticipated need for 24 hours of monitored recovery into inpatient status.

So if the patient can stay overnight in a hospital bed following outpatient tests or surgery without observation, when would observation be appropriate? WPS Medicare (LCD L32222) explained that when a patient has a significant adverse reaction (beyond the usual and expected response) as a result of the test that requires further monitoring, outpatient observation or inpatient hospital services may be reasonable and necessary.

Outpatient with observation services? Yes. Inpatient? Only if the patient ends up requiring a second midnight of care, or if you want to test your luck with the vague concept of the physician judgment exception added by CMS into the Two-Midnight Rule on Jan. 1, 2016. (See Dr. Ronald Hirsch’s article at https://www.racmonitor.com/the-new-short-stay-exception-read-before-using to learn more.)

TMF Health Quality Institute, the Texas Medicare QIO, in its Medicare Outpatient Observation Physician Guidelines Q&A, answered the question, “Can a same-day surgery patient with no postoperative complications be admitted to observation?” TMF responded “No. There must be medical necessity of observation services documented in the medical record. Observation is not to be used as a substitute for recovery room services.” Another question asked, “Can a patient be placed in observation status prior to outpatient surgery?” TMF’s answer: “No. The need for observation care should be determined by the patient’s condition during the postoperative recovery period, not prior to surgery.”

TMF further described other situations in which observation would not be appropriate for surgical patients, such as:

• Routine stays following late surgery;
• Outpatient therapy/procedures (unless there is documentation that the patient’s condition is unstable);
• Normal postoperative recovery time following surgery;
• Stays for the convenience of the patient, family, or doctor; and
• Stays prior to an outpatient surgery procedure.

TMF who? TMF Health Quality Institute evolved into a different kind of QIO since the spring of 2013, when Dr. Meyerson’s original article was published, as currently, the only Beneficiary and Family Centered Care (BFCC) QIOs in action are Livanta and Kepro when it comes to hospitals. The points given above still ring true, though, when it comes to situations that would not be appropriate for observation hours.

TMF even supplied a list of typical postoperative problems that warrant observation:

• Persistent nausea/vomiting;
• Fluid/electrolyte imbalance;
• Uncontrolled pain;
• Dysrhythmias;
• Excessive/uncontrolled bleeding;
• Psychotic behavior;
• Unstable level of consciousness; and
• Deficit in mobility/coordination.

This is helpful, too, and still applicable. Remember that anything that requires additional assessment, care, or treatment beyond routine recovery means an observation order should be placed by the provider. The list above only scratches the surface of reasons why a patient might be appropriate for observation services.

So, what should a surgeon or proceduralist do if no adverse event has occurred, but the physician wants to extend monitoring because the patient is at risk for complications or may not recover as expected due to age, frailty, or comorbidities? The proper approach would be to use overnight recovery in an outpatient bed. There would be no need to order observation because an adverse event did not occur; the physician may order observation only after such an event, and the medical record must indicate clearly the reason that observation was medically necessary. A surgeon concerned about the risk of complications would order extended recovery, monitor the patient overnight (as an outpatient in a bed), and either release the patient the next day or order observation (or admission) if a complication does occur.

This paragraph still rings true, but I’d like to point out that supporting the term “extended recovery” can be a slippery slope. Many institutions used this designator interchangeably with “observation” or “23-hour observation” back in the days when needing 24 hours or more of hospitalization equaled inpatient status. I have found that sticking with “inpatient,” “observation,” and “outpatient in a bed” makes things much clearer.

It’s ironic, and somewhat illogical, that a patient may be admitted to the hospital as an inpatient prior to surgery for what ordinarily would be an outpatient procedure if the surgeon is concerned about a high risk of complications due to the patient’s clinical condition or past history, but the surgeon would not order observation based on a similar risk assessment after the operation or procedure.

This is exactly why, almost five years later, when the 2018 OPPS Final Rule came out about including direction about total knee arthroplasties, which were taken off the Inpatient-Only List, there was much angst and gnashing of teeth.

Some surgeons are uncomfortable placing a patient in an outpatient bed overnight following surgery without ordering observation. They should be reassured that this is not a quality or safety issue. The same monitoring and treatment may be ordered for a patient in overnight recovery as for a similar patient who has observation services ordered. Some also are concerned about the financial implications, and feel that if they order observation, “at least the hospital gets paid.” Unfortunately, this is not always the case. Contrary to CMS policy, which provides for payment to the hospital for observation services for a patient placed into observation from the ED or from a physician’s office, “if a hospital provides a service with status indicator ‘T’ on the same date of service, or one day earlier than the date of service associated with HCPCS code G0378 (used to denote observation hours), the composite APC 8003 (used to bill the observation stay for patients placed into observation from the ED) would not apply … HCPCS code G0378 will continue to be assigned status indicator ‘N,’ signifying that its payment is always packaged.” In other words, if a patient has an outpatient procedure (status indicator “T” or “J1” on CMS Addendum B), the hospital does not receive any additional payment for observation. It is “packaged” into the Ambulatory Payment Classification (APC) payment for the procedure whether observation is ordered or not. (OPPS Final Rule, Nov. 1, 2007, CMS-1392-FC.)

Today, instead of APC 8003, we have C-APC 8011 to bill observation stays. However they are still separated from an outpatient procedure with status indicator “T” or “J1” on CMS Addendum B, and there is no additional payment given to the hospital for the observation services. This does not mean there is no reason to pursue an order for observation services, when appropriate. Remember that for quality tracking purposes, including assessment of length of stay, it’s important to identify which patients followed a regular and routine recovery pathway, and which did not. Also, while Fee-for-Service Medicare does not provide additional payment for observation services in these cases, your other payors likely do. 

Notice that the TMF list of situations in which observation would not be appropriate includes “stays prior to an outpatient surgery procedure.” It adds that observation cannot be used for a prep of any kind, including, for example, preoperative hydration or cardiac assessment, bowel prep, or “renal protection protocol.” Nor can observation be based on time spent in recovery. Observation is properly used only if an intra- or post-op procedure complication actually occurs.

I’d like to interject here with an example of a rare instance where pre-hospitalization for observation services before a procedure would be justified.  Mind you, this is the only case I saw in almost nine years of serving as a health system physician advisor, but it demonstrates that it is a possibility. The patient, a brittle Type 1 diabetic, was scheduled for a colonoscopy. The gastroenterologist was concerned because when the patient was prepped at home for the same procedure just a year before, she developed profound hypoglycemia, to the extent that she required emergency care and subsequent hospitalization for stabilization. In this case, the patient truly required pre-hospitalization, not simply to carry out the GI prep, but also to administer IV fluids with dextrose and closely monitor her blood glucose levels for active treatment, either with adjustments to the fluids or adjustments to her insulin administration. 

If surgery or a procedure interrupts observation and the patient returns to the observation bed for continued evaluation or short-term treatment, and there is still a question of whether the patient will have to be admitted, observation would continue – but the time the patient was under “active monitoring” in the operative suite (including routine recovery in the post-anesthesia care unit) would be carved out for billing purposes.

This remains true.

So far, Recovery Audit Contractor (RAC) auditors have not paid a lot of attention to the use of observation. With respect to postoperative observation, since there is no additional payment, there would be no incentive for an auditor paid on a contingency fee basis to bother auditing these records. However, there is reason to be concerned about proper use of observation for surgical patients, since Medicare requires accurate billing even if there is no payment rendered. In its 2012 and 2013 Work Plans, the U.S. Department of Health and Human Services (HHS) Office of Inspector General (OIG) outlines a focus on “observation services during outpatient visits.”

“We will review Medicare payments for observation services provided by hospital outpatient departments to assess the appropriateness of the services and their effect on Medicare beneficiaries’ out-of-pocket expenses for healthcare services,” the precise language reads. So even though there is no significant effect on Medicare beneficiaries’ out-of-pocket expenses for postoperative observation, if the OIG investigates a hospital’s use of observation due to an excessive number of such claims filed as compared to peers, the hospital could face a compliance challenge if the review reveals inappropriate and/or excessive use of observation.

I did not investigate the outcome of the OIG Work Plans in 2012 and 2013 mentioned above. Looking at the active Work-Plan items on the OIG website involving “observation,” there is only one involving hospital care, dated November 2016 and titled Medicare Payments for Transitional Care Management. However, this involves care management services provided to patients moving from a hospital, partial hospital, or skilled nursing facility to the community setting, and not specifically observation services provided in the hospital setting.

Considering the limited circumstances under which postoperative observation is appropriate, and the lack of reimbursement for the service, hospitals would be wise to monitor the use of observation among these patients and ensure that when observation is billed, there are documented postoperative complications – and that observation is not used for preoperative preparation or for routine postoperative recovery.

Agreed! If your hospital’s surgeons and proceduralists continue to place patients into outpatient status with observation services for routine postoperative care, that’s a situation that needs to be addressed. Conversely, education is also imperative to ensure justified observation hours are captured and billed.  From your providers to the bedside nurses to the case and utilization managers, do they know when an observation order should be entered? Make it a point to ask around…you might be surprised what you find.