Coding Questions Arising From NPAC, Answered

By Erica Remer, MD, FACEP, CCDS, ACPA-C

I attended the 2024 National Physician Advisor Conference (NPAC) hosted by the American College of Physician Advisors (ACPA) in the middle of April. Its theme was “Safeguarding Patients Amidst Shifting Currents of Healthcare.” I love this conference because it is so rich with information and expands my knowledge base beyond clinical documentation integrity (CDI).

I gave an intro to CDI talk at the Essentials and Fundamental course on the first day, where we trained new physician advisors (PAs) and prepared veterans to sit for the ACPA-C. This is a new certification developed by the premier physician advisor organization to signify that an individual has the working knowledge to serve as a PA.

I also had the honor of kicking off the second day with a plenary session called Physician Advisors: Lifeguards Innovating Solutions to Diverse CDI Problems, in which I shared some details about projects I implemented while I was a physician advisor. I received some questions after the fact, and thought this might be a good venue in which to answer them.

How can you get organizations to move beyond only using MS-DRGs as the tool of risk-adjusting metrics?

Originally, CDI was meant to seek out and capture comorbid conditions and complications (CCs) and major CCs (MCCs), thus optimizing each MS-DRG and tier and improving reimbursement and length of stay. When the All-Patient Refined (APR)-DRG methodology was introduced, optimizing the severity of illness (SOI) and risk of mortality (ROM) metrics was a companion goal.

Then came value-based purchasing and other related quality metrics, again establishing other risk-adjustment-type models. And we can’t forget the Hierarchical Condition Categories (HCC) model for Medicare Advantage (MA) and other Accountable Care Organizations (ACOs).

If an organization is only looking at MS-DRGs, they are missing opportunities in multiple other risk-adjusted models. It’s a good place to start for burgeoning CDI programs, but it is only a jumping-off point.

My philosophy always has been to encourage providers to deliver excellent medical care, foster excellent documentation, have CDI specialists (CDISs) pick up the slack when needed, and ensure that coders accurately and compliantly code – and your reimbursement and quality metrics will fall where they belong in every model.

This segues into another question asking if I have any comment on documentation for getting out of PSIs (patient safety indicators). I attended two great talks on the topic of PSIs and quality metrics, by my Vice Chairs (Waldo Herrera-Novey and Adriane Martin) and two other members of my committee (Neelima Divakaran and Joey Cristiano). We all agree: We want folks not to get dinged when the problem is not really a post-procedural condition, and to address medical quality issues if it is. I never condone doing excessive contortions to avoid legitimate post-operative complications. Providers need to be taught that post-op is not a temporal reference, and to ensure that proper linkage is present. If the acute hypoxic respiratory failure is due to an exacerbation of underlying chronic obstructive pulmonary disease (COPD), say that, not “post-operative respiratory failure.”

Another PA asked if I could suggest resources through which you can access how CC/MCCs have financial impact on reimbursement. MS-DRGs are tiered by whether there are CCs or MCCs. Having a single CC is equal to having five of them. The relative weight is a measure of the average resource consumption, severity, and complexity of the patient. When you multiply the relative weight by the hospital’s blended base rate, you determine the reimbursement. If you refer to Table 5 of the current fiscal year’s Inpatient Prospective Payment System (IPPS) Final Rule (https://www.cms.gov/medicare/payment/prospective-payment-systems/acute-inpatient-pps/fy-2024-ipps-final-rule-home-page), you can find the specific DRGs’ relative weights. You can see the difference between a base DRG without a CC or MCC versus with one. You have to ask your administration for your hospital’s base rate.

Our health system has chosen to use Sepsis-2 as our criteria. Are we wrong? and Why not use severe sepsis if it’s on the list of CMS-approved codes and there’s a recognized definition?

First, let me get the terminology straight. SEP-1 is the terminology that indicates The Severe Sepsis and Septic Shock Management Bundle, which is the National Quality Forum sepsis quality measure that the Centers for Medicare & Medicaid Services (CMS) use. Its aim is to standardize sepsis treatment to reduce morbidity and mortality from the condition.

(To find the most up-to-date version, search for Specifications Manual for National Hospital Inpatient Quality Measures and the year of interest. It is listed by quarter. For Q3 and Q4 2024: https://qualitynet.cms.gov/files/65972a96d4b704001df0ae89?filename=HIQR_SpecsMan_v5.16.zip)

In the 1990s, esteemed medical groups came together to try to define the condition of sepsis and operationalize its treatment. Multiple iterations of the literature came out every few years, but Sepsis-2 (not SEP-2) was essentially defined as a presumed or confirmed infection with some systemic inflammatory response syndrome (SIRS) criteria. Although there were always other clinical indicators that could be seen in sepsis, the medical community gravitated towards using the general SIRS criteria, which were ubiquitously collected in all patients in whom the diagnosis was being considered. These were:

• Fever or hypothermia;

• Tachycardia;

• Tachypnea; and

• Elevated or low white blood cell count.

In Sepsis-2, there were three buckets of sepsis: sepsis as defined above, severe sepsis, which was sepsis with organ dysfunction, and septic shock, which was severe sepsis with hypotension not reversed with fluid resuscitation. ICD-10-CM has codes to indicate all three gradations.

Eventually, it was recognized that this was too broad a definition – there were many patients with infections who exhibited SIRS criteria but did not have sepsis, and some patients who had sepsis but didn’t demonstrate these particular signs.

Hence Sepsis-3. In 2016, The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) were rolled out. The definition was established as “life-threatening organ dysfunction caused by a dysregulated host response to infection.” It was recognized that SIRS was sometimes an appropriate response to infection, and therefore, might not indicate sepsis. Furthermore, the presence of organ dysfunction is the critical ingredient for the diagnosis of sepsis.

This contracted the buckets to two: sepsis (the condition formerly known as severe sepsis, which includes organ dysfunction) and septic shock. ICD-10-CM did not change the coding schema. As a result, all sepsis now should have a code for the sepsis, another code indicating the presence of organ dysfunction, with or without shock, and additional codes for the causative infection and the resultant organ dysfunction.

In early 2017, Surviving Sepsis Campaign adopted the Sepsis-3 definition. They concurred with the International Consensus folks that “The current use of two or more SIRS criteria to identify sepsis was unanimously considered by the task force to be unhelpful” because “SIRS criteria do not necessarily indicate a dysregulated, life-threatening response.”

During the reign of Sepsis-2, many organizations put in place an alert system, and they utilized those basic SIRS components as the trigger. It is still useful as a warning signal, but SIRS does not establish sepsis. Many providers are resistant to change and cling to SIRS. I am supportive of using SIRS to identify patients who might have sepsis, but I strongly caution against using it as the diagnostic criteria.

The question I would ask the first organization is: do they want to catch in their net patients who do not end up having sepsis? Identify potential sepsis patients preliminarily and then have the clinicians rule it out over the course of the encounter. Then, the diagnosis should be removed before being coded and final-billed.

If you don’t, you frontload the system with patients who are erroneously assigned into a sepsis DRG, and you will be inundated with clinical validation denials on the back end. Not only that, but the U.S. Department of Health and Human Services (HHS) Office of Inspector General (OIG) has added Medicare inpatient hospital billing for sepsis to their workplan.

Is it wrong to use Sepsis-2 as your criteria? I would not recommend it. The current definition has been around for seven years. It’s time to transition.

Why shouldn’t providers use “severe sepsis” if it is on the list of CMS-approved codes and there’s a recognized definition? I don’t believe clinicians should alter their practice of medicine to satisfy coding. They may need to include some antiquated terminology, however, for the coders to capture legitimate conditions, due to the peculiarities of coding rules. Fortunately, there is coding guidance instructing coders that if providers diagnose sepsis with related organ dysfunction, the coder may compliantly pick up R65.20, Severe sepsis without septic shock, without the doctor using the obsolete qualifier “severe.” The condition formerly known as “severe sepsis” is now just “sepsis.”

Consider inserting my macro into your electronic health record (EHR):

Sepsis due to (infection) with acute sepsis-related organ dysfunction, as evidenced by (organ dysfunction/s).

This allows the clinician to avoid having to use old terminology, with the bonus of helping them ensure that the condition is indeed present and clinically valid.

The last question was, “How do you message the importance of thorough and complete documentation when physicians/providers are thinking the E&M 2023 changes have now relieved them of ‘note bloat?’”

I don’t endorse conveying the message that documentation is for billing. I believe thorough and complete documentation is important to tell the story of each patient encounter.

The professional fee Evaluation and Management (E&M) changes made in 2023, when Inpatient and Observation Hospital Care merged, are only based on medical decision-making (MDM) or time, which now should reduce note bloat. The history and physical examination should be performed and documented “as medically appropriate.” No more gratuitous review of systems just to check a billing box. There is no reason to embed a radiology report in the progress note at all, let alone every day of the admission. The provider only gets credit for it once. Rather than copying and pasting the assessment and plan daily, maybe spend a few moments thinking about what needs to be in the note today and how it informed your decisions.

Reducing note bloat means streamlining the documentation to make it really count. Does “no acute events overnight, no acute distress this morning, reportedly at baseline” explain why the patient in Room 204 is still here today? Does a list of signs and symptoms relay what your impression is? Does a hodgepodge of copied and pasted paragraphs really constitute a discharge summary? Remember that your “copying and pasting saves me time” is someone else’s “I hate slogging through other people’s copying and pasting to find the essential points.”

It is crucial to not buy into the “documentation is an unwanted burden on the provider” nonsense. Documentation is part of the process of delivering quality medical care to the patient and enabling others to recognize it. If providers put “mentation” into their documentation, the medical record and the patient will be better off for it.