News Alert: Startling Misdirection in CMS TKA Memo

Issues abound in the newest edition of MLN Matters

Most of us started our day this past Tuesday by finding the newest edition of the Centers for Medicare & Medicaid Services (CMS) MLN Matters (No. SE19002) in our email inbox.

Titled “Total Knee Arthroplasty (TKA) Removal from the Medicare Inpatient-Only (IPO) List and Application of the Two-Midnight Rule,” my first thought was that some kind of massive server delay had occurred over at CMS. After all, TKA came off the Medicare IPO list over a year ago, on Jan. 1, 2018. Other than MLN Matters No. MM10417, posted Dec. 22, 2017 and titled “January 2018 Update of the Hospital Outpatient Prospective Payment System (OPPS),” which simply announced that TKA was coming off the list, there has not been so much as a shred more clarification published as an MLN Matters (or much else, for that matter) from CMS. But sure enough, this MLN Matters was posted Jan. 8, 2019 – exactly one year and one week after TKA came off the IPO list. CMS certainly does not win the prize for timeliness in this instance.

As I read, it was not long before alarm bells started going off in my head. You too? Let’s compare notes on what confounded us the most, moving from the start of the article to the end:

1. Critical Access Hospitals (CAHs) aren’t subject to the two-midnight rule? That’s news to all of us, not the least of which folks working in those facilities. Can we please get a regulation specialist at CMS to proof these articles before they’re released?
2. In the second paragraph of the “What You Need to Know” section, it reads “…CMS continues its long-standing recognition that the decision to admit a patient as an inpatient is a complex medical decision, based on the physician’s clinical expectation of how long hospital care is anticipated to be necessary, and should consider the individual beneficiary’s unique clinical circumstances.” 

   While I recognize that a few paragraphs later, they discuss the case-by-case exception added to the two-midnight rule in the calendar year (CY) 2016 Hospital OPPS (CMS-1633-F), this initial statement specifically delineates the physician’s “complex medical decision” in regard to inpatient status as being “based on the…expectation of how long hospital care is anticipated to be necessary.” Consideration of the patient’s “unique clinical circumstances” appears here to be taken into account in relation to the expectation of hospital length of stay (i.e., at least two midnights), not, per the case-by-case exception of 2016, that “the admitting physician/practitioner’s judgment that the beneficiary required hospital inpatient care despite lack of a two-midnight expectation (is) based on complex medical factors, including but not limited to patient history, co-morbidities, and current medical needs; severity of signs and/or symptoms; (and) risk of adverse events.” The manner in which the length of expected hospitalization time is emphasized (and, indeed, the only consideration taken into account) at the start of this MLN Matters and not the points described in the case-by-case exception, when it comes to physician determination of inpatient status, makes me even more reluctant to advise my providers to utilize the case-by-case exception.

3. What on earth is present in this first cited case to support starting the patient in inpatient status from the get-go? The patient is listed as being 75 years old with a history of knee osteoarthritis with pain and swelling, and no surgical complications. How is this patient’s condition complicated or complex enough to support inpatient status without a two-midnight expectation? Or, why would this patient be anticipated to require at least two midnights of care in the hospital? Simply stating that this patient started as an inpatient without providing any supporting data to justify the decision is detrimental in an educational article such as this one. 
4. For the second case, a “65-year-old female presented to the facility…for elective TKA surgery and was placed in observation post-operatively.” OK, there must have been a complication during the procedure. Let’s read on and see what it was. It’s not in the case summary… let’s move to the rationale for the approval. “No intraoperative complications were noted.”  What in the what? Why was placing the patient in observation post-operatively appropriate, then? The answer is that it wasn’t. A patient who has been determined to be appropriate for inpatient status would start in inpatient status. But a patient not appropriate for inpatient status would start in – you guessed it – outpatient status.  

   While years ago, surgeons and other medical staff felt that “observing” a patient overnight for routine recovery following a scheduled procedure fit the bill for adding observation services, it should be clear to everyone today that it does not. Observation services involve care provided to the patient above and beyond what is expected or routine in the normal recovery process following a procedure. So if a surgeon’s routine plan of care for recovery involves an overnight stay, a morphine PCA for the first 12 hours post-op, and IV antibiotics Q6, all of it should be wrapped up into the single outpatient payment for the procedure, which includes the routine post-op care. A patient appropriate for outpatient status should have observation services added to the claim (because remember, there is technically no such thing as observation status) only in the event of a complication that takes place during the procedure itself, or during recovery following said procedure. In this CMS example, there was no procedural complication. So placing the patient into observation post-operatively was not supported. Not pointing this error out in the article is, again, detrimental to the educational nature of the piece, and even may serve for some as a green-light from CMS to endorse this practice. Reading over the U.S. Department of Health and Human Services (HHS) Office of Inspector General (OIG) audit of Northwestern Memorial Hospital posted in March 2015 (https://oig.hhs.gov/oas/reports/region5/51300051.pdf), I feel the OIG would agree with me on this one.

Here’s another sneaky confounder the MLN Matters article fails to shed light upon: while this patient was eventually appropriate for inpatient status, given her development and need for hospital management of hypoglycemia spanning two midnights, the patient claim might then not fall into DRG 469-470, since the hypoglycemia was not a direct complication of the procedure. If this is the case, the inpatient admission would not be part of the Bundled Payments for Care Improvement (BPCI) or Comprehensive Care for Joint Replacement (CJR) programs, rendering the patient disqualified for the three-day skilled nursing facility (SNF) waiver. Yikes!

5. In the third case, here is the moment we have been waiting for one year and one week: CMS clarification on what co-morbidities or other clinical factors support inpatient care without a two-midnight anticipation. Alas…maybe not. This patient arrives for his scheduled TKA with quite the active medical history: coronary artery disease, atrial fibrillation, complete heart block with a pacemaker, diabetes, osteoarthritis, and hypertension. In the rationale for approval section, we’re told that “…due to the patient’s extensive cardiac history, it is reasonable to approve this case based upon presence of risk factors for an adverse event.” But what does this really mean? Is it the combination of three cardiac conditions that fit the bill? What if there were only two? How about the heart block? Would it count as equally risky if the patient did not have a pacemaker? We’re just not sure, and in the end, we still get no clear guidance. Back to relying on your own educated guesses, people!

Even if we did have more clarification, I believe there is another point of education lacking here. There is no mention of what the physician documented as to why this patient’s co-morbidities put him at higher risk, making inpatient designation appropriate. Is CMS telling us that physicians simply have to rattle off a list of co-morbidities and not offer any insight into why they are risky, in relation to the upcoming procedure? I can’t imagine this is the case, given how many other missives have emphasized the need for elucidation of the physician’s medical decision-making, based on the patient’s clinical condition and supporting factors. I feel that CMS really missed a terrific chance to show us just what kind of insight is needed from the physician in the documentation when it comes to equating co-morbidities with risk.

6. Can I take another moment to soothe the part of my psyche that relishes the act of proofreading (weird, I know) and complain that the change in the third case (which continues in the fourth) of adding a zero in front of the month was not the case for the first two? Also, sometimes a zero is placed before a single-digit day of the month, and other times it’s not.

The rationale given in the fourth case regarding why it is not appropriate for inpatient status should be applied to the first case. They have it right here, noting that “the procedure was performed without any intraoperative complications. Patient comorbidities were minor, and no adverse concerns were documented.” Agreed. So why is there no mention of the inappropriateness of placing the 75-year-old patient with no co-morbidities into inpatient status in the first case? There sure should be.

7. Lastly, let’s finish with the frequently asked questions. Once again, the response to the third question emphasizes that the “complex medical decision” by the physician about making a patient inpatient or not involves “…how long hospital care is anticipated to be necessary.” It is really striking to me that they repeatedly place such an emphasis on the two-midnight anticipation and barely mention consideration of risk involving co-morbidities or patient condition.

Also, didn’t the 2018 OPPS final rule mention taking the patient’s anticipated need for post-surgical services into account when making a decision on status? In the CMS response to the comment about the ability of patients to access post-acute care at a SNF, they said “we agree that the physician should take the beneficiaries’ need for post-surgical services into account when selecting the site of care to perform the surgery. We would expect that Medicare beneficiaries who are selected for outpatient TKA would be less medically complex cases with few comorbidities and would not be expected to require SNF care following surgery.” To many, including me, this seemed to indicate that for TKAs alone, CMS was allowing anticipation of skilled care in a post-acute facility to play into the decision-making for inpatient status. But without this mentioned in any of the scenarios in the most recent MLN Matters, does it really? This may have been one of the greatest bones of contention in the whole 2018 OPPS Final Rule, and now we still don’t have any clear direction? Frequently asked questions, indeed.

While I am aware that CMS has already been contacted about these issues, it’s hard to say when corrections and clarifications will be available. In the meantime, I recommend formulating a plan to ensure that everyone in your health system who might pay attention to this MLN Matters knows about its limitations, and frankly, errors. And perhaps consider reaching out to CMS yourself to encourage a rapid amendment to the article.  

Comment on this article