Save your money on CDI and coding classes.
Get access to 17+ hours of webinars + answers to your questions by other CDI and coding experts with 24/7 forum discussions.

Check it out for free!

All That Glitters Is Not Gold

“All that glitters is not gold.”

A relative reminds me of that often.
He’s a marriage counselor.
So he has insight into families who appear to have it all together…
but don’t.

He uses that aphorism when I compare myself to others.

But now my inbox has me applying that same idea to health system quality rankings.

It all started with a post on Linkedin.
And then someone in my online community posed a similar question:

On both Linkedin and my online community, it led to a great debate. Most responses fell into two camps:

1. You code the condition.

2. You have an escalation process that allows the code to be removed if it is truly considered clinically invalid by a committee and would result in payment for a condition that was not actually present and did not consume resources.

You can see the debate in the comments on the Linkedin post but to me, point #2 seemed like the obvious more ethical stance - it’s an attempt to prevent fraud.

But then I started receiving private messages.

Coders had seen what happens next.

That same process, created with good intentions, became a slippery slope. It was later used to remove codes that were correct, but negatively impacted a quality metric. The rationale became that certain codes did not capture the “spirit” of the metric or reflected “coding weirdness.”

Personally, I accept those oddities as part of the system. Over time, these quirks tend to wash out in the averages. After all, most quality metrics are relative to peers, not absolute values. Every hospital plays by the same rules.

Or they should be…

I heard rumors that some hospitals were removing codes that “don’t really matter” because they don’t change the DRG. And that it became accepted that a physician advisor’s opinion should override coding conventions.

And over time, those small exceptions created cultures where:

• Codes that “look bad” on metrics became negotiable

• Clinical expertise became a shield against any challenge

• Accuracy gave way to optics

• And most concerning: Coding was left out of the room when coding-policies were created

And then you see reporting that hits you in the face, like Penny Jefferson’s work on admit type inconsistencies and their impact on Patient Safety Indicators:

(For context, many PSIs are only triggered for elective procedures, not urgent or emergent. So, changing the admit type to non-elective leads to less PSIs)

Does that prove fraud?

No.

But it certainly raises questions. Those inconsistencies not-so-subtly suggest that some organizations may be gaming the system.

Then you read about UnitedHealth’s in-home screenings for peripheral vascular disease using devices not used in clinical medicine, which resulted in dramatically higher rates of PVD (which triggers increased payment from CMS) in Medicare Advantage patients compared to Traditional Medicare.

Or health systems that appeared exemplary, later found to be adding unsupported diagnoses like morbid obesity in patients without a qualifying BMI, or “qualitative platelet disorder” simply because a patient was on aspirin.

Those stories force you out of naivety.

They also reframe your own metrics. When you perform well and know you’re playing by the rules… man that feels good.

Years ago, on Brian Murphy’s Off the Record podcast, he asked me what makes a good CDI Physician Advisor. My response focused on documentation.

Today, my answer is different: The top quality is internal integrity, that is, doing the right thing even when no one is looking.

So, do I still believe that having a formal escalation process for truly unsupported diagnoses is ethical?

Yes.

But only in the right organization.
With the right people.
Who hold integrity and compliance in high regard.

And that brings me back to coders.

With very rare exception, non coders should not be telling coders what to code or not code.

Coders are not obstacles to your goals.

They are guardrails.

Working around them is not being strategic.
It suggests that optics matter more than truth.

If your metrics look bad, the answer is almost never “change the codes.”

The answer is one of two things:

• Either it just is what it is, or

• Your clinical care and/or documentation need to improve

You do not fix quality metrics by distorting data.
You fix data by improving care and documentation.

Clinical truth, reflected through accurate coding, is our common ground.

Protect it.
Respect the people who defend it.
And let coders be coders.

Share your thoughts in the comments, or on Linkedin!

That’s all for now.

-Robert

Thank you to Laura Samson, RN BSN CCDS for editing this newsletter!

Take a look inside my online community in the video below!

What other subscribers are enjoying:

1. Check out my CDI and Coding Village online community for free! What you’ll get:

• Answers to questions within 24 hours by other experienced CDI/coding professionals

• Reliable and accurate education via monthly recorded webinars by me and other industry experts

• Comprehensive PDF guides on top CDI topics

• Interactive 24/7 forum discussions with like-minded people who openly share information

• Regular case reviews & discussions

• Access to my exclusive CDI Tip Card

• Direct access to me

2. The Practical Guide to Attending Documentation video course.

• Stop “playing it safe” by billing appropriately (usually higher) with SHORTER notes.

• Minimize note bloat and focus on patient care

• Get 3.5 hours of CME credit! (Use your CME funds!)

3. The Resident Guide to Clinical Documentation video course. The course that interns and residents:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures.

• Gain the confidence to write shorter yet more effective notes.

• Use notes as a checklist to prevent mistakes and provide more complete care.

If you were forwarded this newsletter and would like to subscribe:

Sources:

How UnitedHealth turned a questionable artery-screening program into a gold mine

This is the second in a periodic series from STAT about how UnitedHealth Group wields its unrivaled physician empire to boost profits and expand its influence.

www.statnews.com/2024/08/07/unitedhealth-peripheral-artery-disease-screening-program-medicare-advantage-gold-mine

Senate finds UnitedHealth used ‘aggressive strategies’ in Medicare Advantage

Senate finds UnitedHealth used aggressive strategies to increase Medicare Advantage payments, highlighting billing practices and federal spending concerns.

www.beckershospitalreview.com/financial/senate-finds-unitedhealth-used-aggressive-strategies-in-medicare-advantage/?origin=RCME&utm_source=RCME&utm_medium=email&utm_content=newsletter

Join my online community for free here! Take a look inside 👇️ 

You’ll get:

• Answers to questions within 24 hours by other experienced CDI/coding professionals

• Reliable and accurate education via monthly recorded webinars by me and other industry experts

• Comprehensive PDF guides on top CDI topics

• Interactive 24/7 forum discussions with like-minded people who openly share information

• Regular case reviews & discussions

• Access to my exclusive CDI Tip Card

• Direct access to me

CDI programs send too many queries.

You’ve probably heard this complaint before.

Sometimes it’s a frustrated physician.
Sometimes it’s CDI consultants themselves.
Sometimes it’s someone with something to sell.

It’s easy to get defensive.

But sometimes they’re right.

And instead of getting defensive, I’d use it as a moment to pause and reflect on your own program.

Don’t ask yourself if you’re sending too many queries. You’re always going to say no.

Instead, ask yourself more specific questions:

1. Does every query have a specific impact?

I’ve been told by some CDI specialists that their programs encourage them to send a query when documentation would result in a non-specific code even though the more-specific code would have no impact on DRG assignment, CC/MCC capture, risk adjustment, quality measures, etc.

If it won’t have a practical impact, why send it?

Sometimes the response is, “Well we might get denials for that.”

If that’s the case, then the next question is:

2. Are you actually get denials for that?

Query fatigue is real.
Burnout is real… and dangerous.

You should not be sending queries because of theoretical denials which may or may not be occurring. That’s quite the slippery slope.

(Yes, there are some situations where steps can/should be taken to reduce an otherwise high denial risk, but the solution must be reasonable and not overly burdensome)

To avoid that, you need denials data.

And to have that, you need to reach outside of your silo:

3. Do you have regular meetings / feedback with your Revenue Cycle Integrity Department?

Or whatever department handles denials.

Regular meetings to identify trends and create upstream prevention strategies would be ideal…. but if you’re not, do you even know the names of the people to contact in your revenue cycle integrity department?

I’ve talked to CDI leaders at large health systems who’ve admitted siloing is a major problem.

Get out of your silo.
Make connections
(I don’t like the word connections... make friends.)

Even better, get off virtual!
Have coffee or lunch!
Perhaps… go into the office!
I know, I know. Put down the pitchforks. But I can’t tell you how many “Aha” moments that resulted in real change came from random just-stepping-in-your-doorway-to-chat discussions. Those random unstructured in-person discussions is where the magic often happens.

I digress…

So perhaps you are aware of denial trends.
Are queries your only form of upstream prevention?

4. Are you providing multiple forms of impactful education?

No, I don’t mean monthly 5 page PDFs filled with walls of text that no one will ever read (I’ve seen them. I can guarantee you they don’t work).

I mean real in-person regular departmental meetings. If you do, are they concise (less than 10 minutes), single-topic focused, and devoid of CDI lingo?

Education also needs to come in multiple forms. People often have to hear / see things several times for things to stick.
So not only in-person meetings, but perhaps:

• Weekly concise newsletters (Problem —> solution oriented)

• Posters / pocket cards

• System-wide screen savers to disseminate information

• And maybe even videos! (Yes I know that’s uncomfortable but humans clearly love short form videos - See Tiktok, Youtube, Instagram, Twitter(X), etc.).

Abridge transforms clinical conversations into complete, accurate notes, helping physicians spend less time documenting and focus more on patient care. In a recent ROI analysis with partner health systems, clinicians reported spending up to 23% less time documenting, and observed meaningful improvements in RVUs per encounter—driven by more complete and accurate documentation.

5. Are you going to regular departmental meetings but not getting any traction?

If that’s the case are you using too much CDI lingo?

Have you communicated how CDI may benefit them? (Always take this angle when possible).

Have you listened to the providers to understand why they’re not bought in? Perhaps there’s no alignment (i.e. their census is too high and there’s nothing in it for them).

Perhaps the providers are unhappy with hospital administration and they see CDI as part of that. But if that’s the case, can you take that information to executive leadership? You could be a bridge builder - become a voice for both the providers and administration. Solve a problem for the providers and they’ll be your biggest fan. CDI buy-in will be much easier.

If you can’t get into those departmental meetings:

6. What efforts have been made for CDI to have a regular seat at those departmental meetings?

Do you not have executive / administrative support for you to be in those meetings? If that’s the case, does your executive leadership understand the importance? If you have met with them but still do not have their support, could you have done a better job of understanding their motivation / incentives to then know how to communicate the benefits of CDI?

If the departmental meetings are provider only, have you tried lobbying your executive team for a physician advisor position?

Before you go into a meeting with an ask, have you sought to solve one of their issues first (I did this with our cardiologists)?

7. Do you understand documentation workflow and templates?

Perhaps the EMR / tech is problematic and actually hurting CDI / documentation efforts.

If that’s the case, have you worked with EMR / IT to solve some of those issues?

If the tech can’t be fixed, instead of providing generic education, provide actual screenshots of the EMR (as discussed in this newsletter on fractures, the verbiage you’re recommending may not be available if they’re forced to select from drop-down menus. Provide a work-around or, again, work with the EMR to change it)

8. Is your CDI program judged solely on query metrics?

If that’s the case, then you’re incentivized to not do ANY of the above because actively doing those things may result in LESS queries which will look like you’re making LESS of an impact.

We’ve seen this in our program. Upstream efforts have resulted in less financially-impactful queries… but that’s allowed us to focus on quality. Thankfully, we have leadership that understands that importance.

If you are solely query-metric based, then it’s time to have a discussion with your executive leadership to consider other KPIs in an effort to reduce queries with upstream fixes.

I could go on and on, but you get the picture. Keep asking yourself “why?” to each question / accusation. Truly seek to find a root cause.

And yes, all of the above takes someone in CDI who is not doing a substantial amount of day to day chart reviews. You need someone in a position whose role is broader strategic analysis. If you don’t have one, it’s probably time to get one.
It’ll be worth it.

So where does that leave you?

This is how I see it:

• Group 1: Your program is mature, aligned, and data driven. You send the right queries for the right reasons. Keep going.

• Group 2: You are doing your best but you are stuck sending too many queries and you know it. You know your queries are not ideal, but the system forces your hand. You need leadership support, better tools, and clearer alignment. You deserve help and resources.

• Group 3: You have no idea where you stand. Time to use the above as a self audit.

The invitation

This is not blame. It’s an opportunity. Even I need to answer some of the above questions about our own program.

A well run CDI program improves accuracy, strengthens quality performance, reduces denials, supports financial integrity, and builds trust across the clinical enterprise.

Physicians do not necessarily dislike CDI. They dislike unnecessary interruptions and administrative burden with no perceived value. When the work is targeted, relevant, and clearly connected to their world, the relationship changes.

Your next step starts with an honest self audit.
Do not fear the criticism.
Use it as a mirror.
Start fixing problems today, not tomorrow.

That’s all for now.
Cheers,
Robert

Thank you to Laura Samson, RN BSN CCDS for editing this newsletter!

What other subscribers are enjoying:

1. Check out my CDI and Coding Village online community for free! What you’ll get:

• Answers to questions within 24 hours by other experienced CDI/coding professionals

• Reliable and accurate education via monthly recorded webinars by me and other industry experts

• Comprehensive PDF guides on top CDI topics

• Interactive 24/7 forum discussions with like-minded people who openly share information

• Regular case reviews & discussions

• Access to my exclusive CDI Tip Card

• Direct access to me

2. The Practical Guide to Attending Documentation video course.

• Stop “playing it safe” by billing appropriately (usually higher) with SHORTER notes.

• Minimize note bloat and focus on patient care

• Get 3.5 hours of CME credit! (Use your CME funds!)

3. The Resident Guide to Clinical Documentation video course. The course that interns and residents:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures.

• Gain the confidence to write shorter yet more effective notes.

• Use notes as a checklist to prevent mistakes and provide more complete care.

If you were forwarded this newsletter and would like to subscribe:

Take a look inside my online community in the video below!

Check it out for free!

Just a few months ago if you had asked me whether a person who falls from standing and breaks a hip had a traumatic fracture, I would have said yes.

There was an impact.
Therefore it was traumatic.
Most physicians would probably say the same.

As you can guess, that is not the right way to think about it.

And that mindset has real consequences for coding, quality, and reimbursement.

Why this matters

First, credit to Dr. John Kennedy and Brian Murphy whose Linkedin posts brought this to my attention.

When you look at the medical (non-surgical) DRGs around fractures, you will find two major groups based on what ICD-10-CM codes index to which DRGs.

Traumatic
DRGs 533 and 534 for fractures of femur without MCC and with MCC
DRGs 535 and 536 for fractures of hip and pelvis without MCC and with MCC

Pathological
DRGs 542 - 544 for pathological fractures and musculoskeletal and connective tissue malignancy without CC, with CC, and with MCC

MS-DRG v42.1

(Make sure you can see the picture above. If not, scroll to the top of this email and click “read online”).

A few important points jump out. The traumatic DRGs only have two levels: without MCC and with MCC. The pathological group includes the typical three levels: without CC, with CC, and with MCC.

(Vocabulary reminder. The base DRG is the DRG without a CC or MCC.)

Although the traumatic base DRGs have higher weights than the pathological base DRG, the pathological CC DRG is higher than the traumatic base DRG - and CCs are relatively common in this population. The pathological MCC DRG is also higher weighted than the traumatic MCC DRGs.

If you’re unsure why that matters, I covered the reimbursement math in a previous newsletter. Short version: Higher weight means higher reimbursement.

Quality is also affected. The pathological non-base DRGs have longer geometric mean lengths of stay (GMLOS) compared to their traumatic counterparts. If your hospital tracks expected versus observed length of stay, a pathological assignment increases the expected stay (when a CC or MCC is present).

There are more impacts to quality measures which I’ll mention later.

Where most hospitals are missing out

A pathological fracture is a break in a bone that has been weakened by an underlying disease process, occurring with minimal or no trauma. These underlying diseases include metastatic cancer, primary bone tumors, infections such as osteomyelitis, Paget’s disease, hematological malignancies and… osteoporosis.

Osteoporosis-associated pathological fractures are the low hanging fruit as osteoporosis is very common. Bones affected by the other diseases are typically more front-of-mind for clinicians and hopefully are documented as pathological (but likely still a documentation opportunity).

There are two major issues with osteoporosis fractures:

1. Osteoporosis is present but missed on admission, OR the fracture is documented as traumatic rather than being due to / associated with the known osteoporosis.

2. Fragility fractures are under recognized.

Fr- what?
I know.
I had never heard of them either.
Neither had any of my fellow hospitalists I polled.

Abridge transforms clinical conversations into complete, accurate notes, helping physicians spend less time documenting and focus more on patient care. In a recent ROI analysis with partner health systems, clinicians reported spending up to 23% less time documenting, and observed meaningful improvements in RVUs per encounter—driven by more complete and accurate documentation.

What are fragility fractures?

Simplified version of the algorithm from “PostMenopausal Osteoporosis” NEJM (Source below)

Check the algorithm above.
(Again, make sure you can see the graphic. If not, scroll to the top of the email and click “read online”)

Most clinicians are aware of the right side of that algorithm - diagnosing osteoporosis by DEXA scan and FRAX score (fracture risk calculator). However, most are not familiar with the top left: Fragility fracture —> Osteoporosis.

So, what’s a fragility fracture?

A fragility fracture is a fracture which occurs with no associated trauma or with trauma equivalent to falling from a standing height or less. This is important because, as the algorithm suggests above, osteoporosis is diagnosed on the basis of the occurrence of a fragility fracture alone.
Period.
Regardless of DEXA scan.

This definition is supported by (at least) the World Health Organization (WHO), American College of Obstetricians and Gynecologists (ACOG), and American Academy of Orthopaedic Surgeons (AAOS) (sources below).

I also talked to two orthopedic surgeons who confirmed this concept.

This is pertinent to postmenopausal women.
But what about men?

Same.
In fact, mortality is higher in men than women.

The coding of pathological fractures due to osteoporosis will lead to M80- codes. Note that the term “fragility fracture” is included under the M80 header but the specific term is “osteoporosis with current fragility fracture.” So, coders still need the linkage to osteoporosis in documentation.

CDC ICD-10-CM Tabular Index

Why M80 codes are important for Quality Measures

There are no M80- codes included in the list of codes that trigger the CMS Hospital Acquired Condition for Falls and Trauma (HAC-05). Meaning, if a fall with a fracture occurs during a hospitalization and it is due to osteoporosis / fragility fracture and not trauma - then it will not be considered a HAC / the hospital will not get dinged if coded as such.

For Patient Safety Indicator 08 (In-Hospital Fall-associated Fracture Rate) a major update occured in 2025 where a majority of the M80 codes were removed. All upper extremity and lower extremity M80 codes were removed with the exception of femur M80 codes. So, the measure no longer includes non hip M80 fractures / only includes femur fracture subset.

Patient Safety Indicators (PSI) Log of Coding Updates and Revisions Through Version 2025

Takeaways:

Look to see if osteoporosis is present for any patient who presents with a fracture. This can be supported by DEXA, imaging, previous documentation, or therapy such as use of bisphosphonates.

Look to see if the presentation fits the definition of a fragility fracture.

Be mindful of problem lists. Sometimes the chronic problem list includes “osteoporosis without current pathological fracture” which can create conflicting documentation if pulled into the current hospitalization.

If your clinicians pick diagnoses from dropdown menus rather than free text (such as EPIC’s problem-oriented-charting), make sure the correct terminology is available in the system.

For clinicians, if the fracture is associated with osteoporosis, document that link. For CDI specialists, consider querying for pathological fracture associated with osteoporosis if the above clinical indicators are present.

Use coupon BF2025 at checkout to get 30% off my resident and attending video courses on billing and documentation! Coupon expires in 72 hours (Tuesday Dec 9).

Testimonial for my video course for attendings

Share your thoughts in the comments, or on Linkedin!

That’s all for now. Cheers,

Robert

Thank you to Laura Samson, RN BSN CCDS for editing this newsletter!

What other subscribers are enjoying:

1. Check out my CDI and Coding Village online community for free! What you’ll get:

• Answers to questions within 24 hours by other experienced CDI/coding professionals

• Reliable and accurate education via monthly recorded webinars by me and other industry experts

• Comprehensive PDF guides on top CDI topics

• Interactive 24/7 forum discussions with like-minded people who openly share information

• Regular case reviews & discussions

• Access to my exclusive CDI Tip Card

• Direct access to me

2. The Practical Guide to Attending Documentation video course.

• Stop “playing it safe” by billing appropriately (usually higher) with SHORTER notes.

• Minimize note bloat and focus on patient care

• Get 3.5 hours of CME credit! (Use your CME funds!)

3. The Resident Guide to Clinical Documentation video course. The course that interns and residents:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures.

• Gain the confidence to write shorter yet more effective notes.

• Use notes as a checklist to prevent mistakes and provide more complete care.

If you were forwarded this newsletter and would like to subscribe:

Sources:

Postmenopausal Osteoporosis - PubMed

Postmenopausal Osteoporosis

pubmed.ncbi.nlm.nih.gov/37991856

Male osteoporosis-what are the causes, diagnostic challenges, and management - PubMed

Osteoporosis is underrecognized and undertreated in men, even though up to 25% of fractures in patients over the age of 50 years occur in men. Men develop osteoporosis with normal aging and accumulation of comorbidities that cause bone loss. Secondary causes of bone loss may be found in up to 60% of …

pubmed.ncbi.nlm.nih.gov/35961836

Patient Safety Indicators (PSI) Log of Coding Updates and ...

Fragility fractures

WHO fact sheet on fragility fractures, including key facts, risk factors, prevention and WHO response.

www.who.int/news-room/fact-sheets/detail/fragility-fractures

Osteoporosis Treatment: Updated Guidelines From ACOG

More than two-thirds of osteoporotic fractures occur in women, and one-half of postmenopausal women will experience an osteoporotic fracture. The American College of Obstetricians and Gynecologists (ACOG) has published new recommendations for managing this undertreated condition, including guidance on new medications and targeted treatments.

www.aafp.org/pubs/afp/issues/2023/0700/practice-guidelines-osteoporosis-treatment.html

Osteoporosis and Fragility Fractures: Reinvigorating a National Public Health Priority

On May 15, the Senate Special Committee on Aging issued a Request for Information (RFI) on reducing the risk of falls and fall-related injuries and prominently cited the impact of hip fractures following falls.

www.aaos.org/aaosnow/2019/sep/advocacy/advocacy04

I tried keeping this newsletter as concise as possible. For a complete review of the SOFA-2 update (with a surprise guest), check out my webinar in my online community. If you’re confused by anything below, I likely addressed it in the webinar.

After 30 years, the long awaited update to SOFA is here (as was suggested in the 2016 Sepsis-3 definition publication).

Published October 29, 2025 in JAMA, it’s already the hot topic amongst clinicians, CDI specialists, and revenue cycle professionals.

But with that comes misunderstandings and miscommunications. So, let me address the biggest one off the bat:

This is not a new sepsis definition

The sepsis-3 definition is “Life-threatening organ dysfunction due to a dysregulated host response to infection.”

That has not changed.

In the sepsis-3 publication, the SOFA score was used to describe “life threatening organ dysfunction” given it’s ease of use, familiarity to the clinical community, and it’s validated association with mortality.

However, the sepsis-3 article also says, “Neither qSOFA nor SOFA is intended to be a stand-alone definition of sepsis.”

But given their association, it’s understandable why some consider this a “new sepsis definition.” We must caution ourselves, both clinically and in the revenue cycle, from completely marrying the two as there are certainly evidence-based non-SOFA sepsis-related organ dysfunction - especially as health plans use additional restrictions on top of SOFA to issue clinical validation denials. To that end, based on comments from the lead author Dr. Mervyn Singer (who was also an author of the Sepsis-3 definition), it seems the authors are aware of how health plans have been weaponizing SOFA in denials (more on this later).

SOFA Score Changes:

Brain (renamed from “central nervous system”)

• 1 point:

  • GCS score of 13 – 14 (unchanged from SOFA-1)

  • If drugs are used to treat delirium even if GCS score is 15. They reference drugs included in the “International management of pain, agitation, and delirium in adult patients in the ICU guidelines” but I cannot find any publication with that name (nor in their references). Perhaps it is the Clinical Practice Guidelines for the Prevention and Management of Pain, Agitation/Sedation, Delirium, Immobility, and Sleep Disruption in Adult Patients in the ICU which is published in a Q/A format. The main medications mentioned are Precedex and Haloperidol. In my opinion, that seems incomplete as other medications such as risperidone, olanzapine and quetiapine are often used… especially as the caption includes “long term use” of medications for delirium. I’d like to see some clarification on that.

  • They do now include additional descriptors such as localizing pain (see picture) which I like because health plans have started to deny if no GCS is documented. Internists and intensivists are much more likely to document these descriptors versus GCS.

• 2 points

  • GCS 9 - 12 (or withdrawal to pain) (SOFA-1 was 10 - 12)

• 3 points

  • GCS 6 - 8 (or flexion to pain) (SOFA-1 was 6 - 9)

• 4 points

  • GCS 3 - 5 (or extension to pain, no response to pain, generalized myoclonus) (same as SOFA-1 with addition of pain/myoclonus descriptors)

Respiratory

1st column is 0 points

The thresholds are generally higher (meaning lower numbers) for P:F ratio compared to SOFA-1. However, they have now included an alternative SpO2:FiO2 ratio when P:F ratio (i.e. an ABG) is not available. Caveat: SpO2 must be less than 98%. This should be a win for denial appeals as health plans give no credence to S:F ratio despite evidence suggesting it’s adequate.

• 1 point

  • P:F ratio <300 (previously 400);

  • OR S:F ratio <300

• 2 points

  • P:F ratio <225 (previously 300);

  • OR S:F ratio <250

• 3 points

  • P:F ratio <150 (previously 200);

  • OR S:F ratio <200 (must be receiving advanced ventilatory support)

• 4 points

  • P:F ratio <75 (previously 100);

  • OR S:F ratio <120 (must be receiving advanced ventilatory support or ECMO)

The max amount of points someone NOT receiving advanced ventilatory support is 2 unless not using because the devices are not available or due to “ceiling of treatment” (which was a phrase I was not familiar with - this is essentially when higher forms of treatment are not used due to provider/family decision such as goals of care).

Advanced ventilatory support is defined as receiving high-flow nasal cannula, CPAP, BiPAP, noninvasive ventilation, invasive mechanical ventilation, or long-term home ventilation.

The use of ECMO should be assigned 4 points for the respiratory component only when being used for respiratory indications. However, when ECMO is being used for a cardiovascular indication, 4 points should be assigned for both the cardiovascular and respiratory components.

Abridge transforms clinical conversations into complete, accurate notes, helping physicians spend less time documenting and focus more on patient care. In a recent ROI analysis with partner health systems, clinicians reported spending up to 23% less time documenting, and observed meaningful improvements in RVUs per encounter—driven by more complete and accurate documentation.

Cardiovascular

1st column is 0 points

The cardiovascular component still centers on mean arterial pressure and vasopressor use, but SOFA-2 offers more nuance and was updated for modern vasopressor use.

• 1 point 

  • MAP <70 aka 60 - 69 (unchanged)

• 2 points

  • Low dose vasopressor: Sum of norepinephrine and epinephrine <0.2 ug/kg/min OR any dose of other vasopressor or inotrope;

  • OR dopamine <20 ug/kg/min;

  • OR MAP 50 – 59 (if vasopressors are not being used due to “treatment ceiling”)

• 3 points

  • Medium dose vasopressor: Sum of norepinephrine and epinephrine >0.2 to <0.4 ug/kg/min;

  • OR low-dose vasopressor (sum of norepinephrine and epinephrine <0.2 ug/kg/min) with any other vasopressor or inotrope;

  • OR dopamine >20 - <40 ug/kg/min;

  • OR MAP 40 – 49 (if vasopressors are not being used due to “treatment ceiling”)

• 4 points

  • High-dose vasopressor: sum of norepinephrine and epinephrine >0.4 ug/kg/min;

  • OR medium-dose vasopressor (as detailed above) wth any other vasopressor or inotrope or mechanical support;

  • OR Dopamine >40 ug/kg/min;

  • OR MAP <40 (if vasopressors are not being used due to “treatment ceiling”)

Liver

1st column is 0 points

The liver thresholds are tweaked slightly but the basic pattern is unchanged.

• 1 point

  • Bilirubin 1.21 - <3.0 (SOFA-1 range was 1.2 - 1.9)

• 2 points

  • Bilirubin 3.1 - <6.0 (SOFA-1 range was 2.0 - 5.9)

• 3 points

  • Bilirubin 6.1 - <12.0 (similar to SOFA-1)

• 4 points

  • Bilirubin >12 (similar to SOFA-1)

Kidney (renamed from “renal”)

1st column is 0 points

• 1 point

  • Cr 1.21 - <2.0

  • OR UOP <0.5 ml/kg/hr for 6 - 12 hours

• 2 points

  • Cr 2.1 - <3.5

  • OR UOP <0.5 mk/kg/hr for >12 hours

• 3 points

  • Cr >3.5

  • OR UOP <0.3 ml/kg/hr for >24 hr

  • OR anuria (0 mL UOP) for >12 hr

• 4 points

  • Receiving or fulfils criteria for renal replacement therapy (RRT aka dialysis) (includes chronic use);

  • OR if RRT is not used due to "ceiling of treatment" and they otherwise meet criteria: Cr >1.2 or <0.3 ml/kg/hr UOP for >6 hr and at least one of the following:

    • Potassium >6.0

    • OR pH <7.2 with bicarb <12

The article did not expand on what they mean by "includes chronic use" for renal replacement therapy. Does that mean patients on chronic dialysis automatically get 4 points? I’m not entirely sure. If someone has clarity on that, please let me know.

Note that receiving RRT for non-renal causes (such as for intoxications, hyperkalemia for other reasons, acidosis for other reasons, Uremia for other reasons, etc.) does not get you 4 points in the kidney category.

If receiving intermittent RRT, score 4 points on days not receiving RRT until RRT is terminated.

Hemostasis (renamed from “coagulation”)

1st column is 0 points

The platelet thresholds changed slightly:

• 1 point

  • Platelets <150 (same as SOFA-1)

• 2 points

  • Platelets <100 (same as SOFA-1)

• 3 points

  • Platelets <80 (SOFA-1 was 50)

• 4 points

  • Platelets <50 (SOFA-1 was 20)

What the paper says about fluids

Nothing.

Literally, the words “fluids,” “bolus,” and “hydration” are not mentioned anywhere in the publication.

I point this out as health plans are routinely requiring that SOFA criteria be met “after IVF bolus” or “after adequate hydration.” The authors could easily have endorsed that approach if they agreed with it.

They did not.

Again, based on comments from the lead author, it seems clear they knew how health plans have been misusing SOFA, and I suspect the wording in this update reflects that.

How the score is actually calculated

The 1st bullet point under the SOFA-2 score table is this: “The final score is obtained by summing the maximum points from each of the 6 organ systems individually within a 24 hour period.” This is an important clarification as health plans commonly issue denials based on:

• Taking the lessor of the scores

• Only those that were obtained at the same time

• Only those that were obtained on the same day (but the article says that for missing data after day 1 of ICU care, carry forward the last observation - the rationale being that non measurement suggests stability)

• And, again, only those which were obtained after fluids. This strategy directly contradicts SOFA-2.

Closing thoughts

Is SOFA-2 more restrictive than SOFA-1? Yes, slightly. For total SOFA score, more patients had lower values in SOFA-2. When the authors reclassified patients from SOFA-1 to SOFA-2, they found that 49% of the time, their SOFA scores remained the same. 11% of the time, their SOFA-2 score was higher. 40% of the time, their SOFA score was lower.

With that said, I believe the new SOFA-2 score helps us on the denial appeals side with added nuance.

As health plans react to SOFA 2, expect to see new denial language that tries to spin this paper in their favor and/or combines requirements from Sepsis-3, SOFA-1, and SOFA-2. Make sure to highlight the points I’ve mentioned above, and read the paper in its entirety to make sure you understand it yourself (only about a 15 minute read). Print it out. Bring a pen and highlighter.

If you want even more information about SOFA-2, check out my webinar in our online community with a 7-day free trial - you’ll also get 14 hours of recorded webinars hosted by me and other experts, complete PDF summaries on top CDI topics, weekly case reviews, my exclusive CDI tip card, direct access to me, and access to our forums (with >400 topics and growing) where you can share ideas & solutions to new problems with peers in real-time discussions!

Share your thoughts in the comments, or on Linkedin!

That’s all for now. Cheers,

Robert

Thank you to Laura Samson, RN BSN CCDS for editing this newsletter!

Check out what other subscribers are enjoying:

1. Our CDI and Coding Village online community. Join with a 7-day free trial! It’s impossible to know it all, it takes a village! What you’ll get:

• >14 hours of recorded webinars hosted by me and other experts

• Complete PDF summaries of top CDI topics

• Weekly case reviews

• My exclusive CDI Tip card

• Access to our forums (with >400 topics and growing) where you can share ideas & solutions to new problems with peers in real-time discussions!

• Direct access to me

• Cover your cost (and then some) with our affiliate and village advisor programs!

2. The Practical Guide to Attending Documentation video course.

• Write shorter notes and get home faster with 100% confidence you’re billing appropriately.

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Use notes to protect yourself from lawsuits

3. The Resident Guide to Clinical Documentation video course. The course that interns and residents:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures.

• Gain the confidence to write shorter yet more effective notes.

• Use notes as a checklist to prevent mistakes and provide more complete care.

If you were forwarded this newsletter and would like to subscribe:

If you were forwarded this newsletter and would like to subscribe:

Close revenue cycle gaps at the point of conversation by transforming every clinician-patient encounter into audit-ready, billable documentation in real time—with zero back-and-forth. Learn more.

Questions about critical care billing are some of the most common I get.

I’ve mostly avoided writing about it until a recent MLN from CMS caught my attention. It highlighted some nuances I hadn’t fully appreciated. Then, after my Linkedin post about it took off, I figured it was time for a newsletter.

So, let’s dig in.

The Overview

Critical care billing is fairly straightforward: it’s time-based.

99291 covers 30 to 74 minutes of critical care time.

If you don’t meet the minimum 30-minute threshold, you’ll need to bill the appropriate E&M charge (initial or subsequent hospital visit, as appropriate).

99292 is an add-on code used when you exceed the 74 minutes covered by 99291. This is where things start to get tricky… we’ll come back to that.

But let’s start with the foundation.

The Basics

First, as always, a prerequisite to any charge is medical necessity. Sure, you may have spent 35 minutes taking care of a patient who is physically located in the ICU, but is a critical care charge medically necessary? And is your involvement in the case medically necessary in the first place (this must be stated, unfortunately). Medical necessity can be a bit subjective, but it’s actually laid out pretty clearly by CMS:

The above quotation and the rest below are from this CMS document.

To that end, many develop dot phrases / macros that insert this language to ensure this medical necessity is documented as well as some of the requirements detailed below. But as I’ve said before, don’t just document something because it’s “required for billing.” Document it because it’s true and accurate.

So what about beyond that 99291 code?

Well, this is where CMS and the AMA/CPT differ.

Say what now?

CMS & AMA/CPT: A quick refresher

The CPT code set (the codes that capture the specific services/procedures performed by physicians and “qualified healthcare providers” (QHPs) or “non-physician practitioners” (NPPs) as the AMA/CMS calls them) is developed by the American Medical Association. It is then up to CMS to decide if they agree/accept the guidelines laid out by the AMA regarding those CPT codes. Sometimes they do not and CMS will publish their own criteria.

It’s then up to payers (I don’t love that term) to decide which guidelines they’ll follow and for what codes they’ll pay. That’s why there are some CPT codes that exist… but no insurance company pays anything for them. So, it’s not worthwhile to use them.

CMS is obviously a “big payer” (and sets guidelines / regulations themselves, of course) so many other payers follow their lead. This is all complicated, so that’s why coders are crucial to making sure your submitted codes are accurate because no one expects you to change your coding (or documentation) based on payer. Coders typically use software that alerts them to these nuances.

AMA/CPT vs CMS 99292 time differences

CMS will pay out a 99292 “add-on code” when the full additional 30 minutes beyond the initial 74 minutes has been met. So: 99291 for the first 30 - 74 minutes, 99292 for >104 minutes (74 + 30).

However, the AMA/CPT says 99292 can be coded for time “up to” 30 minutes beyond the first 74 minutes using the “midpoint” threshold… but don’t get lost in the lingo. For the AMA/CPT, you can code 99292 (in addition to 99291) with >75 minutes (up to 104 minutes) of critical care time. Then, the AMA/CPT allows an additional 99292 with >105 minutes (up to 134 minutes) of critical care time.

This differs from CMS who says 99292 should only be reported when >104 minutes (74 + 30) have been met, then >134 minutes (103 +30) for the second 99292.

(Big thank you to Betsy Nicoletti for helping me understand these differences. Check out this coding intel post for a nice little chart on the above. She has a membership if you’re interested).

Anyhow… document what you did. Leave the rest to the coders.

No double dipping.

For anyone familiar with E&M / professional billing, this should be a familiar concept. You cannot include something in your E&M / critical care time that you are also billing for separately. Example: A cardiologist who charged for an EKG read can’t then use that EKG read as part of their MDM points for their E&M charge. No double dipping. This is what CMS says is included:

Procedures such as resuscitation, endotracheal intubation, and central line insertion can be billed separately but make sure your note clearly states that time spent doing these procedures was not included in your critical care time.

Other details about 99291 and 99292:

Does the time have to be continuous?

No.

What about when the critical care services crosses midnight?

If it’s a continuous service, report the total time. But, if the care is disrupted (such as having to care for another patient), then that starts a new initial service on that second calendar date past midnight:

What about two providers in the same group providing critical care for the same patient on the same date?

This was a new one for me that I learned after the recent MLN (and is new as of 2022). Essentially, add up the time as an aggregate, following the same time-based guidelines as above. See the additional caveats below:

What about split/shared services with a physician and NPP?

Whoever provided >50% of the time, they should drop the code. But, again, no double dipping. If you and the NPP are seeing the patient/providing services at the same time, that time can only be counted once. Each provider should document what they did and the time they spent providing critical care. Oh and don’t forget the -FS modifier (again, coders can/should do this for you).

By the way, I run an online community for CDI specialists and coders. This week, one of or members made a post just to say how much she’s loving the community 👇️ . We have monthly webinars (15 recorded - immediately available to members, each with a PDF summary of that topic), 400+ forum discussions, and you’ll get exclusive access to my one-page CDI tip card. Come check us out with a 7 day free trial!

Do you have to itemize your time or provide a start/stop time?

No. But do provide a total time, do not provide a range of time and do not use “greater than” language. See below.

Can you drop an E&M code and a critical care code on the same day on the same patient?

Yes, but only if the E&M was separate from the critical care, and the patient didn’t need critical care at the time of the E&M service. Make sure documentation (including two separate notes) shows why each was medically necessary. You’ll need the 25 modifier to indicate these are two seperate encounters.

Can you include time spent doing activities off the unit or off the floor of the patient?

Although CMS and the AMA/CPT differ in their language here, they essentially say the same thing and the answer is no. You must be on the unit with the patient. The AMA/CPT says it outright that the patient must be “immediately available” to the practitioner. CMS says the patient must have your “full attention.”

Lastly, can multiple providers from different specialties submit critical care codes?

Yes!

Below is a banger of a chart CMS released in that MLN summarizing much of the above.

Wrapping it up

Critical care coding is one of those topics that looks simple at first glance but quickly turns into a maze of time thresholds, documentation nuances, and payer-specific quirks. The key is to stay grounded in what matters most: medical necessity, accuracy, and clear documentation.

If you document what you actually did, capture the total time, and make sure your note reflects the patient’s true level of illness (including the life-threatening aspect of their impairment of one or more organ systems), you’ll be fine. Use EMR documentation technologies to help you out - but make sure they’re true and accurate.

Coders can handle the rest.

If you found this breakdown helpful, share it with your colleagues! Don’t forget to subscribe to this newsletter if you’re not already, and follow me on Linkedin!

That’s all for now. Cheers,

Robert

Thank you to Laura Samson, RN BSN CCDS and Kristi Knight, RN CPC CPPM CCDS-O for reviewing this newsletter!

Check out what other subscribers are enjoying:

1. Our CDI and Coding Village online community. Join with a 7-day free trial! It’s impossible to know it all, it takes a village! What you’ll get:

• >15 hours of recorded webinars hosted by me and other experts!

• My exclusive CDI Tip card!

• Access to our forums (with >400 topics and growing) where you can share ideas & solutions to new problems with peers in real-time discussions!

• Cover your cost (and then some) with our affiliate and village advisor programs!

2. The Practical Guide to Attending Documentation video course.

• Master the new 2021 / 2023 E&M Guidelines

• Write shorter notes and get home faster with 100% confidence you’re billing appropriately.

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Use notes to protect yourself from lawsuits

3. The Resident Guide to Clinical Documentation video course. The course that interns and residents:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures.

• Gain the confidence to write shorter yet more effective notes.

• Use notes as a checklist to prevent mistakes and provide more complete care.

If you were forwarded this newsletter and would like to subscribe:

Close revenue cycle gaps at the point of conversation by transforming every clinician-patient encounter into audit-ready, billable documentation in real time—with zero back-and-forth. Learn more.

When I first got involved with our denials program, it was a bit like flying blind. Each coding or clinical validation denial was handled one at a time. We’d write an appeal, send it off, and hope for the best. There was little tracking of the data. If I helped with an appeal, it was tough to even find out if we won.

With some effort, that’s changed. And the results have been powerful.

The Data We Now Capture

Today, every coding and clinical validation denial is logged with:

• Denial letter date

• Payer and the vendor they use

• Original billing DRG and revised DRG

• Specific audited codes

• Denial dollar amount

• Appeal strength: weak, weak-moderate, moderate-strong, strong (and “no appeal” if we don’t pursue it)

• Dates of first, second, third level appeals and the final outcome

• Discharging physician

This level of detail has been game changing.

It’s revealed hidden trends of which we weren’t aware, and given us objective numbers to match our gut instincts (and sometimes to challenge them).

What the Data Revealed

• Denials are climbing: Year over year, we’ve seen a clear increase in clinical validation denials, enough that they’ve now surpassed coding denials.

• Big dollar numbers: Before we tracked them, we had zero idea the total impact of coding and clinical validation denials. Now, we can slice and dice the impact by diagnosis, by payer, or by timeframe. This has not only been eye-opening for me, but also for other stakeholders. Although most physicians probably don’t care much about impacts on hospital billing, showing dollar amounts that start with M is still impactful.

  • We’ve used this data to get buy-in for CDI-focused hospitalist metrics, hiring of additional staff to fight denials, and get better tech solutions.

  • When I attempted to get the proposal of a new sepsis code in front of a board member of the American Hospital Association, the first request was to provide the dollar impact of sepsis denials by MA plans. I produced that data within minutes and with that - the scale of the problem was immediately understood.

• Payer patterns: One payer wanted to discuss clinical definitions (yet again) of sepsis and respiratory failure (with the assumption we should be fully aligned with theirs). However, after looking at the previous year’s data, we realized that that payer had issued ~3 times the number of clinical validation denials than every other payer combined.

  • Additionally, our appeal success rate is 15% with them, whereas our success rate with others is closer to 60%. This data suggested they are the outlier (which helps inform upstream contracting considerations and working toward agreed-upon clinical definitions rather than accepting theirs outright).

  • Plus, we showed them the exact percentage of denials that we did not appeal the year prior (objectively suggesting we are a good-faithed appeals program - that that we do not appeal every denial as I know some programs do). We admit when we’re wrong and only appeal when we feel the chart genuinely supports the documented diagnosis (and that is certainly the case >15% of the time).

But wait, data isn’t everything.

Before I go on, this data and trending is retrospective - and it isn’t everything.

As a physician advisor, there is still value in being involved with the denials themselves. For example, we started seeing denials for encephalopathy in patients with baseline dementia. As soon as I saw a few of those (before the appeals processes had even concluded), I immediately deployed education to our hospitalists to:

1. Document the patient’s baseline mental status.

2. Describe exactly how they’re off baseline (and ideally use GCS).

3. Chart their return to baseline.

4. Separate any hospital-acquired delirium clearly.

When Data Shows the Game Has Changed

We also use our numbers to check our gut instincts.

This year, we felt that our denials were not being given the same good-faith consideration that they had been before. To check that, we looked at the data. In previous years, we had never lost an appeal that we ranked as strong. This year? We had already lost several.

Nothing changed about us - but it suggested the game had changed and we needed to change our strategy.

Focusing on “better appeal letters” clearly wasn’t going to be a winning strategy.
“Better documentation,” while remaining a central strategy, wasn’t the end-all-be-all either. In fact, we’d lost a denial where I used the documentation in that case to highlight an example of near-perfect documentation.

We had to go further upstream.

Upstream beyond Upstream: Contracting

Fighting denials one at a time is exhausting and expensive. The better play is preventing them.

With the help of Richelle Marting, JD, we have a multi-step plan to limit or even eliminate clinical validation denials through contract language (again, getting her on-board only happened after I showed our executive team the cost of these denials).

But our data also shows with which payers to push hard and where to take a softer approach - it’s a delicate dance after all.

We’ve already utilized the power of the CMS provider complaint process (where MA plans were not following CMS guidelines) to eliminate Sepsis-3 denials, and to have previous Sepsis-3 denials repaid, from one MA plan.

That’s the power of good data and upstream strategy.

Investing in the Process

Yes, this level of tracking does take work. We’ve had to utilize a portion of an FTE to assist with data entry. But, we’re working on new technology to streamline tracking and communication. It’s time-intensive, but worth it.

Defining Success

The last I looked, our overall appeal success rate was about 38 percent which was not a thrilling number the first time I saw it. However, after discussing with Dr. Kendall Smith with PayerWatch who informed me that the industry average is about 30-40 percent, I felt a bit better. But honestly, my goal isn’t just to win more appeals.

True success is fewer denials to begin with.

That means continuing to track the data to help influence not only more complete documentation but also better tech, smarter payer conversations, stronger leverage-strategy, and optimized contract language to limit or eliminate denials in the first place.

The takeaway: If you’re only fighting denials one letter at a time, you’re playing defense forever. Track your data, follow the patterns, and move upstream. The best defense is a good offense.

Share your thoughts in the comments, or on Linkedin!

That’s all for now. Cheers,

Robert

Thank you to Laura Samson, RN BSN CCDS for editing this newsletter!

Join us this Monday October 13 at noon CST for our 15th recorded webinar - this one by Revka Stearns on spinal fusion coding! No commitment needed - sign up with a 7 day free trial!

(Revka’s taking advantage of our village advisor program where members get 60% of their annual membership covered when they give a 1-hour recorded webinar.)

Check out what other subscribers are enjoying:

1. Our CDI and Coding Village online community. Join with a 7-day free trial! It’s impossible to know it all, it takes a village! What you’ll get:

• >14 hours of recorded webinars hosted by me and other experts!

• My exclusive CDI Tip card!

• Access to our forums (with >400 topics and growing) where you can share ideas & solutions to new problems with peers in real-time discussions!

• Cover your cost (and then some) with our affiliate and village advisor programs!

2. The Practical Guide to Attending Documentation video course.

• Write shorter notes and get home faster with 100% confidence you’re billing appropriately.

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Use notes to protect yourself from lawsuits

3. The Resident Guide to Clinical Documentation video course. The course that interns and residents:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures.

• Gain the confidence to write shorter yet more effective notes.

• Use notes as a checklist to prevent mistakes and provide more complete care.

If you were forwarded this newsletter and would like to subscribe:

Close revenue cycle gaps at the point of conversation by transforming every clinician-patient encounter into audit-ready, billable documentation in real time—with zero back-and-forth. Learn more.

For most physicians, our NPI number is something we barely think about. It was assigned back in residency, and outside of calling in a prescription here and there, it rarely comes up. In fact, many of us don’t even know our number off the top of our heads and have to Google it every time.

But here’s the problem: when misunderstanding its importance collides with the complex bureaucracy of billing and information transfer, it creates inconsistencies. And those inconsistencies can lead to sneaky insurance denials.

If your hospital departments are siloed and not identifying trends, those denials can quietly drain significant revenue.

More Than Just a Number

It’s not the NPI number itself that’s the issue. It’s the information tied to it. CMS and health insurers use NPI records to uniquely identify physicians for billing. Alongside the NPI number and other information, the record contains details like:

• Entity type (individual vs group)

• Business address

• Credentials

• Primary taxonomy classification

Any incorrect or outdated detail can trigger a denial.

One field in particular causes trouble: Primary Taxonomy Code.

Why Taxonomy Matters

Your primary taxonomy code is your official “specialty of record:” hospitalist, pulmonologist, infectious disease, and so on. If it’s wrong or outdated, insurers may see a mismatch between what you bill and what actually happened, leading to denials.

Here’s a real-world example:
An internal medicine resident is assigned the taxonomy of “internal medicine physician.” Years later, they complete a pulmonology fellowship, but their taxonomy is never updated. When they start billing as a pulmonologist, the system still shows them as internal medicine.

When that pulmonologist consults on a patient who is also being seen by a hospitalist, to the health insurance company, both appear to be internal medicine physicians. Since two physicians of the same specialty generally cannot bill for E&M on the same patient, same day, the claim for the consult gets denied.

As I’ve discussed in a previous newsletter about E&M billing, billing is impacted by your specialty and sub-specialty. A general cardiologist and an EP cardiologist can each bill separately if they both see the patient on the same day. But if the EP cardiologist never updated their primary taxonomy code, both appear as general cardiologists, and once again, the payer denies the second claim. If you’re RVU-based, that’s money you never see.

The issue gets even trickier when fellows moonlight as hospitalists, then transition into specialty roles, or when they split time between hospitalist and specialty shifts.

Why This Matters Outside the Hospital

Even in independent practice, your NPI profile carries weight. Many online directories pull from the NPPES system (ya know, those online resources that know your business address, phone number, training information and you have no idea how - yeah that’s from the NPPES system).

Anyway, if you’re an independent physician, make sure that business address and phone number are accurate, so patients know how to find you. Also, make sure your personal address and phone number are not listed!

How to Fix It

To ensure your information is correct, do this:

1. Go to the NPPES website.

2. Select the blue button on the right: “create or manage an account.” This takes you to CMS’s login page.

3. Set up your login credentials, then return to the NPPES site.

4. Update your information, especially your primary taxonomy specialty and your business address.

If you find errors that need to be corrected, contact your medical staff office and push them to fix onboarding gaps so this doesn’t keep happening. CMS even allows surrogates to be assigned to handle updates on your behalf.

A Word to Credentialing and Medical Staff Offices

Please don’t put the full burden of this on physicians. Many of us have never heard of NPPES (or did once and have simply forgotten). Simply sending an email that says “update your info” isn’t enough.

Most don’t understand the “why.” We don’t live in the world of administrative bureaucracy. We live in the world of patient care and keeping up with the latest medical updates. If you want this fixed, make it easy. Build systematic onboarding processes, help physicians navigate the updates, and confirm the data gets corrected.

I once saw a denial issue drag on for a full year because everyone was pointing fingers and no one took ownership. When it finally landed on my desk, I called the physician, walked them through the update, and the entire problem was solved in 10 minutes.

That’s how simple the solution can be, but you should have processes in place to ensure it never happens in the first place.

Liked this newsletter? Share your thoughts in the comments, or on Linkedin!

That’s all for now. Cheers,

Robert

Thank you to Laura Samson, RN BSN CCDS and Kristi Knight, RN CPC CPPM CCDS-O for reviewing this newsletter!

Check out what Revka Stearns, one of our online community members, posted on LinkedIn! Click on the picture or the link in the caption and she gets credit if you join!

The CDI & Coding Village

Check out what other subscribers are enjoying:

1. Our CDI and Coding Village online community. Join with a 7-day free trial! It’s impossible to know it all, it takes a village! What you’ll get:

• >13 hours of recorded webinars hosted by me and other experts!

• My exclusive CDI Tip card!

• Access to our forums (with ~400 topics and growing) where you can share ideas & solutions to new problems with peers in real-time discussions!

• Cover your cost (and then some) with our affiliate and village advisor programs!

2. The Practical Guide to Attending Documentation video course.

• Write shorter notes and get home faster with 100% confidence you’re billing appropriately.

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Use notes to protect yourself from lawsuits

3. The Resident Guide to Clinical Documentation video course. The course that interns and residents:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures.

• Gain the confidence to write shorter yet more effective notes.

• Use notes as a checklist to prevent mistakes and provide more complete care.

If you were forwarded this newsletter and would like to subscribe:

Close revenue cycle gaps at the point of conversation by transforming every clinician-patient encounter into audit-ready, billable documentation in real time—with zero back-and-forth. Learn more.

Last week, I had my septic tank routinely serviced. The company made it clear: no service would be provided until they received payment.

I met them in my driveway with a check.

I get it. Chasing down payments after the work is done is costly. I could delay, dispute, or even ignore the bill. I could request documentation of every little thing they did. They’d have to spend time and money sending reminders, possibly arguing over whether the job was done “as expected,” maybe even taking me to court. Those costs would eventually be passed back to customers. By collecting payment up front, they avoid all that, keep prices lower, and can actually plan their business around predictable revenue.

…you picking up what I’m putting down?

Speaking of poop, let’s talk about health insurance companies.

On July 22, 2025, Humana announced it was “accelerating efforts to eliminate prior authorization requirements to ensure a faster, more seamless process.”

Call me cynical, but announcements like this rarely mean what they seem. The press release emphasized that they will “reduce the administrative burden for physicians associated with prior authorizations,” and approving care requests faster.

Why my emphasis on “physician burden?” You’ll see later.

Their entire press release is focused on reducing prior authorizations but says nothing about payment for those services.

The devil is in the details…

But let’s take a step back.

Why the focus on Prior Authorizations?

While most of the revenue cycle is hidden, prior authorizations have become a household term. They are one of the few revenue cycle hurdles that both patients and physicians actually see.

• Patients experience delays in care or canceled procedures.

• Physicians burn hours on peer-to-peer calls or hire staff just to manage paperwork, and they’re using social media to amplify their voices (see Dr. Elisabeth Potter on X)

• After the UHC CEO assassination, the frustration fueled viral social media posts like, “I’m sorry, prior authorization is required before thoughts and prayers” (where “delay,” “deny,” and “depose,” were written on the bullet casings).

Insurers know this is a massive PR problem.

The Shift to the Back End

Here’s what’s already happening and I suspect will increase with the reduction of prior auths: instead of denying care up front with prior auths, insurers are moving denials to the back end (learn what that means in my newsletter issue on the revenue cycle.)

They’ve been laying the groundwork for years with ever-changing medical policies. These aren’t labeled as prior authorizations requirements, but they function the same way. Services are performed, but payment is withheld unless documentation or patient criteria exactly match those policies.

And because policies vary across each and every payer, and can (and do) change multiple times a year, they are nearly impossible to track, even for large health systems. But that’s the strategy - deny via obscurity and complexity.

This is nothing new. Even CMS participates in this practice via RACs (Recovery Audit Contractors) with a look-back period of 3 years, which is longer than most private payer contracts.

Even worse, some payers have been signaling this shift for some time, stating in denials that prior authorization is “not a guarantee of payment.”

Why It’s a Brilliant PR Move

From the insurer’s perspective, it’s win-win:

• Patients get their care without delay.

• Physicians don’t have to do peer to peers or deal with paperwork.

• Furthermore, physicians will likely still get paid, so they’ll stop complaining, especially online.

  (We often see payers will pay the professional fee to physicians, but deny the facility payment. I suspect that will continue with this move)

• The denials fall disproportionately on facilities (health systems, surgical centers, imaging centers, etc.), which bear the larger costs.

The facilities are then forced to go back to physicians after the fact for additional documentation, and request documentation requirements for future cases. Heck, health systems may even have to have their own “prior authorization” processes in place. Suddenly, the health system looks like the bad guy, not the payer

I spoke to Eric Fontana, an industry researcher at Union Healthcare Insight, who confirmed that they’re seeing this broader strategy shift as well.

What This Means for Providers

This is not really a reduction of barriers. It’s a relocation - shifting to a less visible target.

So while headlines promise “fewer prior authorizations,” what’s more likely coming is:

• More retrospective denials of payment.

• More obscure requirements.

• More pressure on hospitals, surgical centers, etc. to absorb costs, chase down clinicians for paperwork, or put their own “checks and balances” policies in place (aka more administrative costs).

The Bottom Line

Payers are trading one form of friction for another. It’s a clever PR strategy, but the administrative burden and cost-shifting will continue to grow.
Patients may stop complaining about delays.
Physicians may feel relief.
But the facilities where care is actually provided will carry the brunt of denials and fight the battles quietly, behind the scenes.

And health systems, I hope you’re prepared.

Share your thoughts in the comments, or on Linkedin!

That’s all for now.
Cheers,
Robert

Thanks to Laura Samson, RN BSN CCDS for editing this newsletter!

Join us on September 22 for our 14th webinar on Anemias & Bone Marrow Disorders including CDI and coding concerns of:
👉 GI bleeds
👉 PSI9 and exclusion criteria many are missing
👉 Coagulopathies
👉 Anemia-related risk-adjustment
👉 Afib and hypercoagulable state
👉 CC/MCC/HCC opportunities
👉 And much more!
And as always, I'll start with a quick anatomy and physiology review, so you TRULY understand what you're reading in charts.
Check out our online community with a 7-day free trial!

Check out what other subscribers are enjoying:

1. Our CDI and Coding Village online community. Join with a 7-day free trial! It’s impossible to know it all, it takes a village! What you’ll get:

• >13 hours of recorded webinars hosted by me and other experts!

• My exclusive CDI Tip card!

• Access to our forums (with ~400 topics and growing) where you can share ideas & solutions to new problems with peers in real-time discussions!

• Cover your cost (and then some) with our affiliate and village advisor programs!

2. The Practical Guide to Attending Documentation video course.

• Write shorter notes and get home faster with 100% confidence you’re billing appropriately.

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Use notes to protect yourself from lawsuits

3. The Resident Guide to Clinical Documentation video course. The course that interns and residents:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures.

• Gain the confidence to write shorter yet more effective notes.

• Use notes as a checklist to prevent mistakes and provide more complete care.

If you were forwarded this newsletter and would like to subscribe:

Close revenue cycle gaps at the point of conversation by transforming every clinician-patient encounter into audit-ready, billable documentation in real time—with zero back-and-forth. Learn more.

“Chronic microvascular ischemic changes.”
“White matter hyperintensities.”

If you’ve reviewed enough inpatient charts (or care for elderly patients), you’ve seen these findings on CT heads or MRI brains.

But here’s the thing:

• I rarely see it in provider notes.

• I never hear anyone mention it.

• I don’t think I ever learned about it in residency.

So I went digging.

Searching for Answers

UpToDate? Nothing.
A PubMed search for “chronic microvascular ischemic changes”? One lonely result.

So I hit the 2025 “easy button” (OpenEvidence) and asked: “What are chronic microvascular ischemic changes, and how do they impact patients?”

That led me to the real term: Cerebral Small Vessel Disease (CSVD).
That search gave me 636 PubMed results. Jackpot.

What is cerebral small vessel disease (CSVD)?

After discussing with a neurologist and performing a literature search, in short:

• Long-standing injury and dysfunction of small blood vessels in the brain

• On MRI: shows up as white matter hyperintensities, lacunar infarcts, old microinfarcts

• The most common, chronic, and progressive vascular disease (per this review)

• Management: the usual suspects (control BP, diabetes, cholesterol, smoking, etc.)

• Outcomes: linked to progressive cognitive decline (45% of dementias), gait disturbance, increased risk of stroke (25% of ischemic strokes), and functional impairment.

That’s… much of the average inpatient population.

Why This Matters for Inpatient Care

If you care for hospitalized patients, you see this daily:

• Small metabolic shifts (UTI, dehydration, low sodium) tip these patients into encephalopathy

• They’re more prone to delirium (which results in longer length of stay and increased staffing needs)

• Falls are more likely

• Many need post-acute PT/OT

This condition is quietly driving resource use across hospitals - it’s not just an incidental finding. Yet it’s rarely documented. So, the question is: Does it matter (is it a CC/MCC)?

Coding of CSVD

It’s never easy, and there is no code for cerebral small vessel disease.
Nor for chronic microvascular ischemic changes.

So, that means a trip down encoder lane.

• “Cerebral small vessel disease” → I67.89, Other cerebrovascular disease

• “Microvascular ischemic changes” → I67.82, Cerebral ischemia (includes “chronic cerebral ischemia”)

And both of these ICD-10-CM codes? They’re CCs!

ICD-10-CM/PCS MS-DRG v42.0 Definitions Manual

So, the good news: DRG methodology will reimburse for the resources needed to manage these patients.
Bad news: since it’s almost never documented, we aren’t giving ourselves credit.

The MEAT of CSVD

As many of you know, inpatient coders can’t code directly from radiology reports. It needs provider documentation. So, where’s the MEAT (Monitor, Evaluate, Assess, Treat)?

• Monitoring and treating blood pressure, diabetes, cholesterol, smoking

• Placing on delirium precautions

• Giving antiplatelets

• PT/OT consults for associated gait and functional decline

• Counseling families on cognitive and functional expectations as a result of this condition

  • Which sometimes may lead to palliative care or hospice discussions / consultations.

Since becoming more aware of this disease process and intentionally including it in my documentation, it has stayed front-of-mind and helped shape my patient conversations and influence my goals of care conversations. That has sometimes resulted in consulting palliative care, and in some cases even hospice. And with MyChart providing immediate access to notes, I’ve had family members directly ask about this new term (CSVD or chronic cerebral ischemia) which they’ve never seen before. Those questions naturally open the door to deeper counseling, expectation-setting, and care planning. All of that, of course, is part of managing this condition

But the key is that the condition, and how it’s being managed (including counseling), has to be documented. That’s the only way to show its impact on the current admission and to make sure we get credit for the work we’re doing.

The Takeaway

CSVD is under-recognized but has a huge impact on our inpatient population. It contributes to encephalopathies, delirium, falls, strokes, dementia, and functional decline… and yes, it codes as a CC.

But if it’s not documented, we don’t get credit for managing it.

So:

• Providers: watch for it on brain imaging, document it when present, and include the MEAT to show how it’s affecting the patient and your care. It will make your care more complete, and it’s certainly changed the way I approach my patients.

• CDI programs: make sure your providers know what CSVD is and why it matters. Use education, tech, whatever tools you’ve got to bring it front-of-mind.

That’s all for now. Got thoughts or comments? Share them on Linkedin!

Cheers,
Robert

Thank you to Neurologist Dr. Zachary Goodwin for reviewing, and Laura Samson, RN BSN CCDS for editing this newsletter!

When you’re ready, there are 3 ways I can help you:

1. Join our CDI and Coding Village online community with a 7-day free trial! It’s impossible to know it all, it takes a village! What you’ll get:

• Watch >13 hours of recorded webinars, and participate in monthly interactive webinars hosted by me and other experts (perfect for new coders / CDIs and those pivoting their career)

• Get access to my exclusive CDI Tip card.

• Share ideas & solutions to new problems with peers in real-time discussions

• Get your cost covered (+ more) with our affiliate program and village advisor program!

2. Check out The Practical Guide to Attending Documentation video course.

• Stop underbilling your hard work & MASTER the new billing rules

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. The course that helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe:

Close revenue cycle gaps at the point of conversation by transforming every clinician-patient encounter into audit-ready, billable documentation in real time—with zero back-and-forth. Learn more.

Sigh. Sepsis…

Denials are rampant.
Hospitals are losing money.
Patients are stuck in a coding no man’s land.
But as time goes on, the problem and its fix is becoming fairly obvious.

It comes down to this: our current coding system (and their clinical definitions) forces an either-or choice between “localized infection” and “sepsis with life-threatening organ dysfunction.” But there is no way to capture the very real middle ground.

That missing middle is creating a clinical headache, and a financial one.

Almost a Decade Since Sepsis-3

It has been nearly ten years since the publication of Sepsis-3 in February 2016.

For those who need a refresher, Sepsis-3 defined sepsis as:

This was a big shift from the long-standing definition based on “Systemic Inflammatory Response Syndrome” (SIRS) criteria, used in some form since 1991 (and updated in 2001 as Sepsis-2). I’ve written a previous newsletter on this topic here.

The Debate Is Far From Over

Despite many declaring the issue settled, the “sepsis debate” is alive and well.

CMS still uses Sepsis-2 and has doubled down on it. And there is certainly not universal adoption of Sepsis-3. Many bedside physicians still use Sepsis-2 and have not rushed to adopt Sepsis-3, even when educated on it. I contrast this to the 4th universal definition of myocardial infarction. Even if physicians have never heard of it, after being educated most adopt it quickly because it makes clinical sense.

Sepsis-3 has legitimate problems, and ignoring them means ignoring questions about the impacts on patients and the very real effect on hospital reimbursement.

The American Hospital Association said it well in a letter to CMS last year:

Yes, Sepsis-3 predicts mortality better than Sepsis-2. That’s great for research, but not so great for early identification - a key to sepsis treatment and survival. The Surviving Sepsis campaign has adopted the Sepsis-3 definition, but not the qSOFA screening tool (preferring SIRS as a screening tool).

But don’t confuse my lack of loving Sepsis-3 clinically with an endorsement of Sepsis-2. I see Sepsis-2 used improperly frequently and it’s a major problem. Sepsis-2 is certainly not the answer, either.

But This Newsletter Is Not About That Debate

That is the last I will say about Sepsis-2 versus Sepsis-3. This is not about winning that argument. Sepsis-3 is not going back into the toothpaste tube and with being almost a decade in, I doubt a new clinical definition will be developed to solve this issue.

So, we need a different fix - one that addresses both clinical concerns and reimbursement realities.

(But for the rest of this issue, when I say “sepsis,” I am implying by Sepsis-3 definition)

The Missing Middle

Right now, coding forces an either-or choice:

• Localized infection

• Sepsis with life-threatening end-organ dysfunction

Does that sound right to you?
Do patients go from having pneumonia to suddenly BAM! full-blown sepsis with life-threatening organ dysfunction?
Of course not.
There’s a progression.
A spectrum.

And can we prevent that progression? Absolutely. It’s the whole point in early recognition and treatment.

And do all patients with localized infection require the aggressive resources to prevent that progression? No.

This spectrum used to be captured by:

• The localized infection code

• The Sepsis code

• The additional code for Severe Sepsis

• The additional code for Septic Shock

Now, “severe sepsis” is redundant and unnecessary, so we’ve lost the middle ground of that spectrum.

How This Hurts Hospitals

While the lack of clinical consensus should be, and is, a major influencer of proposing a potential fix, the reality is that the hospital reimbursement issue is forcing this issue to the top.

CMS calculates DRG reimbursement by tracking and averaging the cost of treating each condition. Because CMS still uses Sepsis-2, their dataset includes patients without organ dysfunction. Those cases use fewer resources, which drives the average cost (and thus payment) for sepsis DRGs down (In RCM lingo, this is “artificially lowering the relative weight of sepsis DRGs”).

This means that hospitals treating true Sepsis-3 patients are under reimbursed. If you’re keeping track, that means it’s a doubly wammy when payers layer on extra criteria like “after hydration” or “after ED treatment” before they will accept a Sepsis-3 diagnosis (i.e. issuing clinical validation denials).

The Proposed Fix: A New Code

A group of industry leaders, including Dr. James Kennedy, Dr. Cesar Limjoco, Dr. R. Kendall Smith, Penny Jefferson, and Shirlivia Parker, have proposed creating a new ICD-10-CM code for “early sepsis” or “pre-sepsis.”

I support this initiative.

These terms are intuitive to physicians. When explaining this issue to other physicians (the lack of a code that captures this spectrum) they will often interject and say “like early sepsis?” before I even mention it. So, I suspect this new term will be easily adopted by physicians.

Additionally:

• CMS could use the code to track it for appropriate reimbursement (possibly as a comorbid condition (CC/MCC), rather than mapping to a sepsis DRG).

• Over time, these pre-Sepsis patients would be removed from Sepsis DRG calculations, driving reimbursement for true Sepsis-3 cases upward appropriately.

If Sepsis is simply too loaded of a term and results in push back, we could alternatively consider something like “Infection with systemic signs or symptoms.” A code for “SIRS” could make sense here, as there are codes for “SIRS due to non-infectious origin.” However, I would prefer to avoid the term “SIRS” to future-proof the code against evolving criteria changes (and prevent us from having to do a coding change debate all over again).

But, I’ll leave that debate to the decision-makers. I just want to raise awareness and start a conversation about the issue and this potential fix.

But What About a Clinical Definition?

Yes, there is no clear universal definition for this pre-sepsis condition. But not every ICD-10-CM code has one.

What is the universally accepted definition of chronic hepatic failure? Chronic cerebral ischemia? Acute respiratory failure with hypoxia? None exist, yet the codes do, and they have major impacts on quality reporting and reimbursement.

In fact, the code creation could lead to the development of an accepted definition as it would allow tracking of this pre-sepsis state.

Moving This Forward

To make this happen, we need help from, among others,:

• The American Hospital Association

• The Sepsis Alliance

• The Surviving Sepsis Campaign

• AHIMA

• ACDIS

• CDC NCHS

• CMS

Certainly endorsement from clinical professional organizations such as the IDSA would also be helpful.

If you believe in this cause and know someone in these organizations, please share this newsletter with them. Even if they disagree with this proposed fix, we need to start a conversation.

But in my view, this solution makes a lot of sense, both clinical and financial.

Share your thoughts in the comments, or on Linkedin!

That’s all for now. Cheers,

Robert

Thanks to Dr. James Kennedy for his input, and thank you to Laura Samson, RN BSN CCDS for editing this newsletter!

When you’re ready, there are 3 ways I can help you:

1. Join our CDI and Coding Village online community! It’s impossible to know it all, it takes a village! What you’ll get:

• Get access to my exclusive CDI Tip card.

• Watch >13 hours of recorded webinars, and participate in monthly interactive webinars hosted by me and other experts (perfect for new coders / CDIs and those pivoting their career)

• Share ideas & solutions to new problems with peers in real-time discussions

2. Check out The Practical Guide to Attending Documentation video course.

• Stop underbilling your hard work & MASTER the new billing rules

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. The course that helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe:

Massive CMS audit news keeping you up at night?
Norwood puts your fears to bed with code audit services by credentialed experts, provider education, risk adjustment and outpatient CDI optimization, and on-demand talent. Contact us for a zero-pressure discussion.

Support this newsletter by checking out today’s sponsor! My comment: The people at Norwood are good people.

/

By now, you’ve probably heard the buzz about CMS’s big announcement on May 21, 2025. If not, here’s the headline: they’re launching full-scale RADV (Risk Adjustment Data Validation) audits on every single Medicare Advantage (MA) plan. Yes. Every single one.

They’re hiring 2,000 coders (up from a mere 40) and rolling out “advanced systems” to search medical records, find unsupported HCC diagnoses, and recoup overpayments. If you’re not sure what HCCs are, check out my previous newsletter for a quick refresher.

Why Now?

Because they’re waaaaay behind. The last big clawback happened after they audited payment year 2007. But CMS estimates that MA plans may be overbilling the government by $17 billion a year. MedPAC thinks it might be worse at $43 billion.

And with the current administration making a big push against waste, fraud, and abuse, they’re coming for that money.

But Wait… They’re Targeting MA Plans, Not Health Systems… Right?

Yes. And while MA plans are responsible for the accuracy of the codes that they submit to CMS (because that impacts how much CMS pays them), MA plans rely on provider documentation to justify the HCC codes they submit. So if CMS tightens the squeeze on MA plans, I suspect those MA plans could tighten the squeeze on us.

So this is an opportunity. We can audit ourselves, find our weak spots, and develop solutions to “protect our payer partners.” Ya know, extend the olive branch and all that.

So What Are They Focusing On?

Glad you asked, Robert. The latest OIG report targeting Coventry Health is a great roadmap. They grouped “high-risk diagnoses” into these categories:

• Acute stroke

• Acute myocardial infarction (AMI)

• Acute Embolism

• Sepsis

• Pressure ulcers

• Cancers: lung, breast, colon, prostate, and ovarian

Their approach to identifying unsupported diagnoses is fairly straightforward (and helpful for building your own audit strategy). For example:

• Sepsis: If it only shows up on one outpatient claim, but not on an inpatient claim, that’s a red flag. Sepsis almost always lands a patient in the hospital.

• Acute MI or stroke: Same logic. If it's coded in an outpatient note without an associated hospitalization, it’s suspect.

• Embolism: If it’s coded without prescribed anticoagulation, that suggests an inappropriate claim.

• Pressure ulcers: If the code only shows up once in the entire year, CMS sees that as unlikely for a significant chronic issue like this.

• Cancers: If they appear on a single claim during the year and don’t have associated chemotherapy drug treatment or surgical therapy in a 6 month period before or after the claim - again, high risk for being inappropriate.

Building Our Own Audit

We ran our own review, modeled after the OIG’s methods, with some differences because of some lack of complexity in our reports.

We focused our audit on this year’s data. CMS is looking back to 2018–2024, but we want to protect ourselves moving forward. Note that if you perform retrospective audits of previous years and find unsupported HCCs for a patient in a capitated payment model, you’ll have to reconcile that and possibly return money.

One challenge we ran into quickly: figuring out which patients are actually attributed to our ACO (Accountable Care Organization) or MSSP (Medicare Shared Savings Plan). That turned out to be harder than expected. So we shifted gears and focused on our Traditional Medicare population. Why? Because it mirrors MA patients, and this audit wasn’t about finger-pointing, but rather identifying system-level issues and designing fixes.

By the way, these chart audits were time consuming and human capital intensive, and not something we’d done before. A great partner, like today’s sponsor Norwood, can help you get it done right and make you look like the hero of your hospital.

What We Found

Honestly, it wasn’t as bad as I feared (kudos to our providers and coding team). For sepsis, every case we flagged came from hospital follow-ups. While those codes shouldn’t have been used in outpatient visits, they didn’t impact HCC scoring since the inpatient claim already captured it for that year. Still, the new code Z51.A, “Encounter for Sepsis Aftercare,” may have been more appropriate in some of those cases.

For acute MIs, strokes, and PEs, we saw a mix of hospital follow-ups and diagnoses that were copy-forwarded from previous hospitalizations outside of the current calendar year. In a few cases, we saw December hospitalizations with January hospital follow ups that could improperly impact HCC risk-adjustment.

What we’re doing about it

Here’s what we’re doing to clean it up:

• Coding software edits: We’re flagging these acute diagnoses that appear in non-inpatient settings. It’s not impossible they’re appropriate, but it’s rare. As Dr. Erica Remer says, “If you’re typing ‘acute stroke’ with one hand in clinic, you better be calling 911 with the other.”

• Discharge documentation update: Most hospitalists were letting acute conditions default to “active” status at discharge (many didn’t even realize they were supposed to review them). We’re changing that. Moving forward, hospitalists will click the “resolved” button for acute conditions at discharge. This simple step prevents those diagnoses from being automatically pulled into outpatient follow-up visits.

• Coder education: We’re reinforcing that acute conditions must have clear, supported documentation in outpatient settings. Otherwise, they shouldn’t be coded.

• Provider education: We’re helping outpatient clinicians understand the difference between acute codes versus “personal history of” codes and why it matters for risk adjustment.

• Ongoing monitoring: We’ll keep an eye on the OIG work plan and run regular audits to stay ahead.

Remember: CDI isn’t about maximizing revenue, it’s about revenue integrity, and even more importantly, documentation integrity. When you document accurately, the rest takes care of itself.

Now go forth and audit. And don’t say I didn’t warn you.

 👇️ BTW, come GEEK OUT with us 👇️ with loads of forum topics, >12 hours of webinars, and my hot-off-the-press CDI Tip card found exclusively in the village!

That’s all for now. Don’t hesitate to ask questions as they help inspire future issues!

Cheers,

Robert

When you’re ready, there are 3 ways I can help you:

1. Join our CDI and Coding Village online community! It’s impossible to know it all, it takes a village! What you’ll get:

• Learn from me and other experienced professionals (perfect for new coders / CDI’s and those pivoting their career)

• Participate in monthly interactive webinars hosted by me, weekly 1 minute CDI quick tip videos, and anatomy and physiology reviews not found anywhere else

• Share ideas & solutions to new problems with peers in real-time discussions

2. Check out The Practical Guide to Attending Documentation video course.

• Stop underbilling your hard work & MASTER the new billing rules

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. The course that helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe:

Get 25% off my attending and resident courses using coupon “25FOR2025”! This is my biggest discount ever. Coupon expires in 72 hours!

The Practical Guide to Attending Documentation — RobertOubreMD.com

www.robertoubremd.com/the-practical-guide-to-attending-documentation

Resident Guide Video Course — RobertOubreMD.com

www.robertoubremd.com/resident-guide-video-course

Staring down at the new MDM chart, I asked the profee (professional fee) auditor, “If I think it’s severe, do I need to explicitly document that or will it be obvious by the problem?”

“You don’t need to state it as severe,” she told me.

She audited my charts. She agreed with 19 of 20 of my codes. The one she disagreed with?

I had dropped my code based on a severe problem but didn’t specifically document it. She did not think the problem was severe.

So, she confirmed my question. I do need to state it as severe.

There’s one consistent theme that comes up every time I teach or discuss the MDM chart and professional coding with physicians: disagreements between the thought process of physicians versus coders.

Coders are not physicians.
Physicians are not coders.
Yet, the system we’ve found ourselves in requires a marriage that at times seems to have irreconcilable differences. But there’s hope for this relationship yet and it boils down to… clear communication.

Professional Coding Arrangements

For physicians reading this who might be confused, there are typically 3 different types of arrangements between physicians and their professional coding.

1. Physician self-coding where the physician selects the CPT (and ICD-10-CM) codes themselves. While, yes, you’ll “get the RVUs you expect,” this puts you at risk of undercoding or overcoding as well as missing complex coding nuances such as modifiers. Physicians in this set-up should pay for frequent audits.

2. Coder review and reconciliation where the physician drops the code themselves, but a coder comes behind them and reviews documentation to ensure that code is supported. If not, the code is changed. Depending on the arrangement, the physician may be made aware of that code change for debate, or deference may be given to the coders. While this process may result in higher accuracy and allows quality assurance by the physicians, it could also be time-consuming and significantly delay claim submission and payment.

3. Full outsourced coding where the physician does not drop a code and coders code based on the documentation. This allows physicians to focus on care and documentation, not coding. This decreases your risk of fraud as, ideally, only codes that are supported by the documentation are submitted without any outside influence. But, quality varies between vendors and you may even have to a get a third party vendor to audit the coding company from time to time.

Bridging the gap

So, what’s the problem?

First, when discussing disagreements with physicians, physicians have told me they “felt” that a visit deserved a level 5 (99205, 99215, etc.). After hearing what they did, they are often correct in that “feeling” but… the documentation didn’t support it. Remember, this isn’t just about getting paid, but for protecting yourself if The Feds ever show up to your clinic. They won’t care what you said you did, they’ll only care about what the documentation shows.

Second, and the most frequent, is assumptions made by physicians that either a coder is not allowed to make or is subtly suggested in documentation.

Example: A hand surgeon sees a patient for follow up of a fracture of a pointer finger. The word “stable” is used frequently throughout documentation. The physician codes a 99214 based on a moderate level problem. However, the coder disagrees and suggests documentation only supports a low-level problem as a stable acute illness / uncomplicated injury and should be coded as a 99213. This was sent to an auditor, who agreed with the coder. This was then sent to me.

I noticed that documentation included evaluation of capillary-refill and sensation distal to the fracture. As the problem primarily involved the bone but assessment required evaluation of “body systems not directly part of the injured organ” (i.e. vasculature and peripheral nervous system), I concluded the problem met the definition of an “Acute, complicated injury”(see page 15). However, I didn’t fault the coder for not coming to that same conclusion.

Examples like that come up over and over and over again. I remind physicians that not only are coders not allowed to infer to determine the level of severity (nor have the background to determine if a limb is at risk of permanent injury), but they are also held to their own productivity standards. I spent way more time reviewing that chart than that coder was probably allowed. While that’s not ideal, it is simply reality. And, unless you’re going to change your arrangement to one of the other two types mentioned above, the problem will occur again (and frequently).

So, my marriage counseling:

Clear Communication

Learn the MDM chart (my version above) and know the buzzwords.
Create dot phrases / smart phrases (or whatever your EMR supports) that includes those buzzwords.
Put that somewhere near your A&P ( I’ve written about this before as the “billing timeout.”). Then, bam, you get the RVUs you expect and you make it black/white for the coder and they efficiently agree with your coding.

Example: If you’re a hand surgeon frequently taking care of finger fractures and the initial visits require assessment of the neurovasculature, then make a dot phrase that inserts, “Closed displaced transverse fracture of a phalanx is an acute complicated injury as it required evaluation of body systems not directly part of the injured organ.”

BUT, remember, you’re not documenting that so that you can bill higher, you’re documenting that so your documentation clearly reflects what you actually did and that can be converted to the appropriate code.

Clear communication. Happy Marriage.
Send me postcards from the honeymoon.

That’s all for now. Don’t hesitate to ask questions as they help inspire future issues!

Cheers,

Robert

Thanks to Kristi Knight, CPC CPPM for editing this newsletter!

When you’re ready, there are 3 ways I can help you:

1. Join our CDI and Coding Village online community! It’s impossible to know it all, it takes a village! What you’ll get:

• Learn from me and other experienced professionals (perfect for new coders / CDI’s and those pivoting their career)

• Participate in monthly interactive webinars hosted by me, weekly 1 minute CDI quick tip videos, and anatomy and physiology reviews not found anywhere else

• Share ideas & solutions to new problems with peers in real-time discussions

2. Check out The Practical Guide to Attending Documentation video course. Use Coupon 25FOR2025 in the next 72 hours to get 25% off!

• Stop underbilling your hard work & MASTER the new billing rules

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. Use Coupon 25FOR2025 in the next 72 hours to get 25% off! This course helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe:

Ever wonder how cirrhosis coding affects your hospital’s HCC risk-adjustment? In this quick excerpt, I explain how accurate liver failure coding gets missed, how it impacts HCC capture, and how to get physicians and coders speaking the same language

If you like topics like that, there are >10 hours of webinars just like it. Check out our CDI and Coding Village online community with a 7 day free trial.

Has everyone forgotten about our anesthesiologists?

I rarely hear them discussed in CDI circles. A search of my favorite documentation podcasts turned up...nothing. Not a single episode.

(If I missed one, share it in the comments on the web version of this newsletter!)

Anesthesiologists are rarely queried, which means they’ve largely flown under the CDI radar.

But it’s time we gave them some attention.

The Elephant in the Room

I have to admit: no provider wants us reviewing their documentation. It may mean more work or a change in workflow…and sometimes they’re not wrong.

But gone are the days when physicians could ignore the financial and legal implications of their notes. The eye of Sauron - I mean CDI - now gazes upon anesthesiology. But with a good team, the shift doesn’t have to be painful.

What I'm Not Talking About: Professional Billing

I won’t go deep into professional billing requirements here. Anesthesiologists tend to have a solid handle on that. Their work is repetitive, time-based, and often well-templated.

Still, for the uninitiated, anesthesia documentation typically includes:

• Pre-operative (or pre-anesthesia) evaluation

• Intraoperative record

• Post-anesthesia care documentation

My focus today is on that pre-anesthesia evaluation and why it matters for CDI and risk adjustment.

Why the Pre-Anesthesia Note Matters

Anesthesiologists assess the patient holistically, not just for surgical suitability, but for how the patient’s comorbidities may impact the anesthesia plan.

And if you’ve read anything I’ve written before, you know that “comorbidities” is a magic word.

The documentation of comorbidities are not only important for reimbursement reasons, but also the growing importance of risk-adjustment.

A Quick Refresher: Risk Adjustment

Risk adjustment ensures providers aren’t penalized for caring for sicker patients. Many hospital and provider quality metrics account for illness burden, using submitted ICD-10 codes as their basis.

But those codes can only come from documentation and only if it’s codable.

Here’s what that means:

• Diagnoses must be spelled out (e.g., “hypokalemia” not “low potassium”)

• Diagnoses can only be coded from that encounter (So “rest of medical history per EMR” aint cuttin the mustard)

• Diagnoses cannot be coded from auto-generated problem lists

• Diagnoses typically require documentation of some form of M.E.A.T (Monitor, Evaluate, Assess, or Treat)

• Information in the EMR can be used to clarify, but not code, some diagnoses. Example: A coder cannot code a BMI without an associated “nutritional diagnosis” such as obesity. If obesity is documented, then the coder can go to the EMR or non-provider documentation to capture the BMI.

Why Anesthesia Notes Matter (More Than You Think)

In short-stay surgical inpatients, there may only be two physicians documenting in the medical record:

1. The surgeon – whose notes typically only focus on the surgical issue (unless they’ve been involved with CDI-initiatives)

2. The anesthesiologist – a bit more aware of, and more in the habit of documenting, additional comorbidities.

That makes anesthesia documentation critical to capturing comorbidities for accurate coding, especially when the surgeon’s note is limited.

Plus, the patient is likely in and out of the hospital before a CDI professional can review the chart.

But there’s a problem…

What I Found in an Audit

In a recent review of our anesthesia documentation, I found:

• Nonspecific terminology: Examples such as “BMI > 40,” “CHF,” “cardiac arrhythmia.” Those are effectively worthless in terms of coding and risk-adjustment.

• Key diagnoses missing or too vague: “Chronic renal dysfunction” instead of “ESRD” (an MCC!)

• Lack of connection between comorbidities and the anesthesia plan as the comorbidities were listed under a “review of systems” section.

And then I uncovered two bigger systemic issues:

Two Systemic Problems You Might Also Have

Problem #1: Coders Couldn’t Find the Notes
In EPIC, our coders were reviewing notes using a particular workflow tab and the pre-anesthesia evaluations were buried and difficult to find.

This wasn’t just a one-off. Other hospitals in our online community confirmed this was an EPIC-based issue.

So if you’re a coding manager, make sure your coders know where to find these notes. Show them exactly how to access them and make it a priority.

Problem #2: The EMR was Working Against Us
Many vague terms (like “cardiac arrhythmia”) were pre-built in the EMR for quick-click selection. Manually typing/adding specific diagnoses (such as “persistent atrial fibrillation”) required a lot more clicks. Anesthesiologists simply were not going to do that and I don’t blame them.

We’re working on fixing that, but that represents bad EMR design and highlights the importance of including your CDI and coding department in any documentation-related EMR changes / builds.

So…Why Should Anesthesiologists Care?

Fair question.

I haven’t found any anesthesiology-specific risk-adjusted claims-based quality metrics. So I tried a few alternative motivators:

• Helping their surgical colleagues improve scores on metrics like Failure to Rescue (see my last newsletter).

• Reducing badgering from the UM department about the usage of that new, expensive reversal agent with improved hospital reimbursement.

Frequently Missed Comorbidities

Below is a list of frequently missed comorbidities in my audit

• Obesity

• Electrolyte derangements (hyponatremia, hypokalemia, metabolic acidosis)

• Chronic systolic and/or diastolic heart failure

• CKD (with stage) and ESRD

• Presence of pacemakers and the underlying arrhythmia

• Acute blood loss anemia and chronic anemias

• Previous stroke with residual effects

• Pulmonary Hypertension

• Underweight and/or cachexia

• Chronic respiratory failure with hypoxia

Final Thoughts

Anesthesiologists don’t need a complete overhaul of their documentation, just a bit of refinement (and visibility).

If you're a CDI/coding leader, take a look at:

• Where anesthesia notes live in your EMR and whether your coders are trained to capture their content

• How anesthesia templates and workflows are built (and how they can be optimized)

• And whether your anesthesia group understands their quiet but vital role in the hospital’s financial and quality metrics

And like David Glaser who ties in themed-songs on almost every episode of Monitor Mondays, Sarah McLachlan’s infamous song is playing in my mind’s ear. Anesthesia, I will remember you. Will you remember me?

Cheers,

Robert

Thanks to Laura Samson, RN BSN CCDS for editing this newsletter!

When you’re ready, there are 3 ways I can help you:

1. Join our CDI and Coding Village online community! It’s impossible to know it all, it takes a village! What you’ll get:

• Learn from me and other experienced professionals (perfect for new coders / CDI’s and those pivoting their career)

• Participate in monthly interactive webinars hosted by me, weekly 1 minute CDI quick tip videos, and anatomy and physiology reviews not found anywhere else

• Share ideas & solutions to new problems with peers in real-time discussions

2. Check out The Practical Guide to Attending Documentation video course.

• Stop underbilling your hard work & MASTER the new billing rules

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. The course that helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe:

If you like quality and risk adjustment topics like this, check out our CDI and Coding Village online community with a 7 day free trial. I discuss risk adjustment in every webinar and we have two expert webinars dedicated to risk adjustment!

In its 2025 Inpatient Prospective Payment System (IPPS) final rule, CMS announced that it was replacing PSI-04 (Death among surgical inpatients with serious treatable complications) with a new one: 30-day Risk-Standardized Death Rate among surgical inpatients with Complications (also known as “Failure-to-Rescue”).

Over a year later, this metric remains an enigma.

First, if you like visuals (and I do) check out this algorithm that summarizes the metric (click the picture to go to direct source):

It is a claims-based metric, meaning it’s monitored (both what is included and excluded) entirely by submitted ICD-10-CM and ICD-10-PCS codes on claims. This contrasts with abstracted metrics where humans (abstractors) search the chart and submit a report to track the metric. Examples of abstracted metrics include NHSN’s Surgical Site Infection Rate and CMS’s Sepsis Bundle (Sep-1).

One thing that surgeons and hospitals should understand is that this measure assesses the percentage of surgical inpatients who experienced a complication and then died within 30-days from the date of their first operating room procedure. It does not track the percentage of complications themselves. Failure-to-rescue (FTR) is defined as the probability of death given a postoperative complication.

Oh, and if you didn’t already know, we’re already in the reporting period (which started July 1, 2023 and goes through June 30, 2025) for the fiscal year 2027 payment determinations.

Which Patients are Included?

It’s not all inpatient surgical patients. It only includes discharges assigned to three groups:

1. General surgery

2. Orthopedic surgeries

3. Cardiovascular procedures

Congrats, y’all!

The procedure that triggers the metric can occur 3 days prior to the admission, all the way through discharge of that admission.

Before we move on, if you want to know specifics, go to this link (automatically downloads a .zip file) and download the excel file. When I refer to “Tables” below, I am referring to the Tables in that excel file. (If the direct link above doesn’t work, go here + go to "measure specs" then "measure calculation then click on "Specification_Failure-to-Rescue_Final.zip.")

The metric also includes discharges associated with ECMO (extracorporeal membrane oxygenation) or tracheostomy MS-DRGs, which must also meet ALL the below criteria (They’re confusing. I’ll explain after.):

• Had an eligible MDC.

  • These include:

    • Circulatory system (MDC 5)

    • Digestive system (MDC 6)

    • Hepatobiliary system and pancreas (MDC 7)

    • Musculoskeletal system and connective tissue (MDC 8)

    • Skin, subcutaneous tissue and breast (MDC 9)

    • Endocrine, nutritional, and metabolic diseases (MDC 10)

• O.R. procedure code (Table 1), that in the absence of ECMO or tracheostomy, would assign discharge to an eligible MS-DRG (Table 2)

• Without a procedure code for ECMO

  • ICD-10-PCS codes (Table 5): 5A1522F, 5A1522G, 5A1522H

• Without a procedure for tracheostomy occurring before or on the same day as the first non tracheostomy procedure

  • ICD-10-PCS codes (Table 6): 0B110F4, 0B110Z4, 0B113F4, 0B113Z4, 0B114F4, 0B114Z4

👆️ That bit about the ECMO and tracheostomy exclusions confused me at first. Let me explain. There are two DRGs that they’re referring to:

• DRG 003 ECMO or Tracheostomy with Mechanical Ventilation >96 hours or Principal Diagnosis except face, mouth, and neck with Major O.R. Procedure

• DRG 004 Tracheostomy with Mechanical Ventilation >96 hours or Principal Diagnosis except face, mouth, and neck with Major O.R. Procedure

As you can see, you can fall into that “ECMO DRG” without ECMO being done. So, if ECMO is performed, then you’re excluded from the metric.

Otherwise, if the tracheostomy is done before or during the O.R. procedure that triggered the metric (and that would have fallen into the original 3 surgical groups), then you’re excluded. If the tracheostomy is done after the O.R. procedure, then it’s still included.

There are a few other factors that can exclude a patient from the metric. These include any of the following:

• Patients aged >90 years old.

• DNR status (ICD-10-CM Code Z66) that is present on admission.

• Patients admitted from a hospice facility.

• Discharged against medical advice.

• Being included in an ungroupable MS-DRG

• No qualifying O.R. procedure with reported date

• Contradictory death information such as:

  • Reported date of death before admitting date.

  • Death date before discharge date when patient was reportedly discharged alive.

  • Discharge disposition reported as died but enrollee has subsequent claims.

• With missing discharge disposition, gender, age, quarter, year, or principal diagnosis

Note that otherwise this metric includes patients aged 18 - 89 enrolled in Medicare.

Which Complications Trigger the Metric?

The actual list of ICD-10-CM diagnosis codes and ICD-10-PCS procedure codes can be found in Table 3 and Table 4 in the referenced excel file above. But, in general, complications include:

• Cardiac events

• Congestive heart failure

• Hypotension or shock or hypovolemia

• Pulmonary embolus or deep vein thrombosis or phlebitis

• Cerebrovascular accident (CVA) or transient ischemic attack (TIA)

• Coma

• Seizure

• Psychosis

• Nervous system complications

• Pneumonia or pneumonitis

• Pneumothorax/effusion

• Respiratory compromise or bronchospasm

• Internal organ damage or perforation

• Peritonitis

• Gastrointestinal bleed and blood loss

• Sepsis

• Deep wound infection or wound complication

• Renal dysfunction

• Gangrene/amputation

• Intestinal obstruction or ischemia

• Retained foreign body

• Pressure injury

• Orthopedic complication

• Hepatitis or jaundice

• Pancreatitis

• Necrosis of bone (thermal or aseptic)

• Osteomyelitis

• Disseminated intravascular coagulation (DIC)

• Pyelonephritis

• Or other postsurgical complications

Again, these must not be documented as present on admission.

How Is It Risk-Adjusted?

This is the question that initially led me to explore this new metric further. It’s the one thing that I, as a CDI physician advisor, can impact. However, my research led me to increased confusion, as I’m not a big enough statistical numbers nerd to understand it all.

You can explore their risk adjustment methodology by visiting here, selecting the “Scientific Applicability” Tab, and then navigating to “Risk Adjustment” on the left. The excel file of their methodology can be found here.

Their risk-adjustment methodology for comorbidities uses Elixhauser variables, which (admittedly) make my head spin with a relatively considerable number of variables which seem like they should increase your risk of complications/death but actually decreases it. These negative Elixhauser variables often contradict some of the Vizient variables that our institution tracks. Being unable to reconcile these conflicting methodologies, I came to an important conclusion and reminder: instead of trying to pinpoint this (and other) metrics with particular comorbidities, we should encourage complete and accurate documentation to the highest degree of specificity possible. No more and no less. It’s the central dogma of CDI.

If you want to know more about Elixhauser, I encourage you to watch Penny Jefferson’s wonderful webinar that she recently presented in our online community (which you can access with a 7-day free trial). Dr. James Kennedy also has a lot of great posts and webinars on the topic on LinkedIn.

A Reminder

Lastly, remember that for the sake of our patients, hospitals and health care providers can benefit from knowing not only their institution’s mortality rate, but also their institution’s ability to rescue patients after an adverse occurrence.

We shouldn’t game the system to make quality metrics perfect, as this can lead to missing important opportunities to identify clinical care issues and ultimately improve patient care at the system level.

That’s all for now. Don’t hesitate to ask questions as they help inspire future issues!

Cheers,

Robert

Thanks to Penny Jefferson for co-writing this newsletter with me. Also thank you to Laura Samson, RN BSN CCDS for editing this newsletter!

When you’re ready, there are 3 ways I can help you:

1. Join our CDI and Coding Village online community! It’s impossible to know it all, it takes a village! What you’ll get:

• Learn from me and other experienced professionals (perfect for new coders / CDI’s and those pivoting their career)

• Participate in monthly interactive webinars hosted by me, weekly 1 minute CDI quick tip videos, and anatomy and physiology reviews not found anywhere else

• Share ideas & solutions to new problems with peers in real-time discussions

2. Check out The Practical Guide to Attending Documentation video course.

• Stop underbilling your hard work & MASTER the new billing rules

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. The course that helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe:

We recently surprised our members in our online community (The CDI and Coding Village) by announcing an affiliate program where they get 40% commission of people they invite (way above the industry standard) which could cover their costs and then some!

Browse the forums (which have been on fire lately) and the 10+ hours of recorded webinars with a 7-day free trial!

“My hospital is telling me I need to document every comorbidity. Do I really need to do that?”

I get some form of this question frequently. So, it’s time to dedicate a newsletter to it. No, it’s not all about reimbursement.

I frame these discussions around the four purposes of documentation:

1. Synthesizing (To help you think and stay organized)

2. Communicating

3. Legal Protection

4. Billing

Let’s dive in.

Why Clinically?

You should be systematic with every aspect of clinical care, and documentation should be a part of that.

Say you pre-round on a patient and notice they’re mildly hyperkalemic. You schedule a few doses of Lokelma and move on.

The next day, it’s no better. You give a few more doses.

The next day, it’s even higher. You up your aggressiveness of treatment and make sure the patient is on a low K diet. That afternoon, a nurse mentions something about potassium supplements. You take a look at the patient’s MAR and realize they’re not only on potassium supplements but also on spironolactone and Lisinopril (both medications that can impact the potassium level). “Oh. Oops,” you say to yourself as you hold those medications.

I like to avoid “oops” in medicine. I’m sure you do too. How could documentation have prevented this?

If you not only had listed their Chronic HFrEF (a comorbidity) as well as their medications, these would have jumped out to you during the process of documenting:

Lisinopril, Aldactone and potassium supplements are staring you in the face, even if you’re just skimming over your own notes while writing. That mistake would have been caught days earlier. In my example, everything turned out okay. But if not, that would look quite bad in a court of law (which is my 3rd purpose of documentation: legal protection. Use documentation to prevent medical errors and the lawyers never come around in the first place)

Why for Communication?

A patient has partial aphasia and right hemiparesis from a previous stroke. However, no one documents that fact in any of the admission documentation. For various reasons, the patient is no longer on antiplatelets nor statin medications. A nurse notices the patient’s aphasia and right hemiparesis and sees no documentation of it. She therefore calls a code-stroke. That patient gets a stat CT head and CTA head/neck before someone notices older documentation suggesting that’s the patients baseline.

That was a medical error. Clear documentation of this aphasia and right hemiparesis from a prior stroke could have prevented that situation.

Why for Legal Protection?

Both of the examples above could result in lawyers getting involved. Use documentation to prevent those medical errors from occuring in the first place, and you don’t have to get to the other legal reasons documentation helps you.

I could tell examples all day long to drive home these 3 points. Unfortunately, most clinicians do not understand the far-reaching impacts documentation has on ourselves, our patients, and our colleagues, and do not take full advantage of this built-in tool to help us care for patients.

Quality Metrics

One way or another, and whether you’re aware of it or not, you’re being tracked by quality metrics. I’ve discussed these at length in previous newsletters (One for surgeons but applicable to non-surgeons too, and another about mortality ratios).

But, briefly, many metrics are “risk adjusted” these days. Meaning, they put everyone on an equal playing field because patient populations vary. So, these risk-adjusted metrics calculate the expected outcomes versus the observed outcomes (You may hear this as O:E ratio).

What impacts the expected outcomes? Those are based on how sick your patient population is.

What is that level of sickness based on? Often claims data.

What is claims data based on? Submitted codes.

What are the codes based on? Your documentation.

A neurosurgeon recently told me, “I wouldn’t go to me if I was a patient and saw these metrics. How do I fix these?” I left him with four major points:

1. Coders cannot code from previous documentation. It’s only from documentation during THAT encounter.

2. They cannot code from problem lists. It must be clear the diagnosis is impacting your care in some way (often summarized by M.E.A.T.: Monitored, Evaluated, Assessed, or Treated).

3. Coders are not allowed to assume. You must be explicit in your specificity of diagnoses and comprehensiveness. Examples:

   • Class 3 Obesity with BMI 44 > “Elevated BMI”

   • Hyponatremia > “Low sodium”

   • Chronic Systolic Heart Failure > “CHF”

   • Chronic respiratory failure with hypoxia > “On home O2”

4. Use your CDI and coding departments as a resource.

Lastly, Reimbursement

There are two main types of billing: Professional Billing and Hospital Billing.

Professional billing is how you get paid. That’s your RVUs based on CPT codes - 99214, 99223, etc. If you’re billing based on MDM, the list of comorbidities can help your complexity of problems.

And then there’s hospital billing, that’s how the hospital gets paid for… everything else. I’ve discussed DRGs in a previous newsletter, but, briefly, many inpatient stays are based on the DRG system whereby hospitals get paid a lump sum payment based on the main reason the patient is in the hospital. Very little can move that DRG one way or the other, but some comorbidities are considered Comorbid Conditions or Major Comorbid Conditions, and they can impact the DRGs and result in higher reimbursement.

On the outpatient side, if you belong to an ACO and/or take care of Medicare Advantage patients, then you’ve probably heard of HCCs. The comprehensiveness and specificity of diagnoses apply there as well.

These reasons, as well as quality metrics for the hospital, are the reasons you’re being told by your hospital to document as many comorbidities as possible. But, there’s a reason I listed those last. There are many other clinical, and self-preservation reasons, to optimize this practice.

That’s all for now. Don’t hesitate to ask questions as they help inspire future issues!

Cheers,

Robert

Thanks to Laura Samson, RN BSN CCDS, for editing this newsletter!

When you’re ready, there are 3 ways I can help you:

1. Try 7 days FREE in our CDI and Coding Village online community! It’s impossible to know it all, but you’re not alone. It takes a village! What you’ll gain:

• Become the local expert by getting real-time crowd-sourced answers from peers.

• Learn from monthly webinars (by me and other subject experts)

• Advance your career by networking with me and other CDI and coding professionals.

• Grow your skills (and your income) with paid opportunities (via our affiliate program and by giving your own webinar).

2. Check out The Practical Guide to Attending Documentation video course.

• Learn how to satisfy billing requirements with ONE sentence (meaning you can document less, write notes faster, and get home sooner)

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. The course that helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe:

The 7-day free trial remains open in our CDI and Coding Village online community. Sign up and get immediate access to over 7 hours of webinars by me and other experts from the community (and 247 forum discussions).
We have a huge announcement coming soon for members only… 😉 

Part of my job is fighting insurances against removing diagnoses thereby keeping the money we’re owed. The CDI lingo for this is clinical validation denial appeals and “DRG downgrades.”

These types of denials are becoming more common compared to traditional coding denials where insurances deny codes because coding guidelines weren’t properly followed, or they disagree with sequencing . These coding denials are relatively black and white.

But now they’re denying that the clinical conditions themselves even exist, which, of course, is not a coding decision but a clinical one. Even when international consensus definitions exist, such as the 4th universal definition of MI and 3rd international consensus definition of Sepsis, health plans still make up their own definitions of diagnoses, impose their own criteria, and then use those self-created parameters to form the basis of an adverse clinical validation determination. If the patient doesn’t have all the criteria that a plan requires, it deems the bedside provider’s diagnosis “clinically invalid.” The health plan will downgrade the payable DRG with the invalidated diagnosis excluded, and ask that the hospital pay money back for the difference between the hospital’s billed DRG and that which the payer re-calculates.

Many hospitals attempt to appeal those decisions and make an argument to keep that money. These are most often argued upon on clinical grounds. We may show how the clinical indicators were present and the patient did meet their definition of that condition. Or, we may disagree with the health plan’s clinical definition altogether and use high-quality evidenced-based publications to support that argument.

But I noticed a concerning trend.

Our appeals weren’t being given the good-faith consideration that they deserved.

Nuance seemed to be welcomed when it favored the payer (insurance), but not when it favored us (the hospital).

Instead of high-quality peer-reviewed publications, health plans were quoting blog posts and online articles.

They were taking international consensus definitions and adding additional layers of restrictions unsupported by any published literature.

It began to feel that clinical arguments were an exercise in futility.

Worse yet, Medicare Advantage payers seemed to be openly refusing to follow CMS regulations despite being tasked with “providing access to care that is equal to or better than Traditional Medicare” (or at the very least, exploiting loopholes). I knew there was probably a problem, but I wasn’t sure, nor did I know where to find the proof.

But I had a friend and a secret weapon.

That secret weapon was healthcare law attorney Richelle Marting (who co-wrote this newsletter with me).

She helped me start tackling these issues, and give me the legal sources to stand behind those arguments. There are many, but below are some of the concerns I had and the legal arguments against Medicare Advantage plans’ adverse clinical validation decisions.  

#1. Their criteria don’t exist under Traditional Medicare rules for DRG assignment, and clinical validation is even explicitly prohibited for several Traditional Medicare contractors.

Medicare Advantage plans have to cover and pay for Traditional Medicare benefits in a manner that is no more restrictive than Traditional Medicare, including the type and level of service. 42 U.S.C. 1395w–22. The benefit in this case is hospital inpatient care. The manner in which the type or level of service is covered and paid for hospital inpatient care is the MS-DRG system. 

If Traditional Medicare does not restrict the level or complexity of the DRG based on specific definitions of individual diagnoses, in theory, nor can a Medicare Advantage plan.

#2. They were not giving us an explanation for their denials, just the final determination, especially for subsequent levels of appeal.

In CMS’s 42 CFR 422.101 (b): eCFR :: 42 CFR 422.101 -- Requirements relating to basic benefits.

• Plans must comply with “General coverage and benefit conditions included in Traditional Medicare laws, unless superseded by laws applicable to MA plans”

• Per the Medicare manual 100-04 Chapter 29: Medicare Claims Processing Manual under “Explanation of the Decision”

  • “<Instructions: This is the most important element of the redetermination. Explain the logic/reasons that led to your final determination. Explain the coverage policy (LCD, NCD), regulations, policy guidance (IOM provisions), and/or laws used to make this determination. Make sure the rationale for the decision is clear and that it includes an explanation of why the claim can or cannot be paid for the particular set of facts at issue in the appeal. For example, the explanation should demonstrate how the beneficiary’s condition or circumstances do not meet specific coverage policy requirements. Statements such as "not medically reasonable and necessary under Medicare guidelines" or "Medicare does not pay for X" provide conclusions instead of explanation, and are not sufficient to meet the requirement of this paragraph.>

• For example, a Medicare Advantage plan decision that summarily concludes a diagnosis of sepsis is invalid because it is unsupported by clinical information in the record is a mere conclusion. It offers no citation to a coverage policy, regulation and/or law used to make the decision.

#3. They were not providing the credentials of the reviewer issuing the denial.

Per the statement of work of the recovery audit contractors, 2011:

• Page 23: "Clinical validation is a separate process, which involves a clinical review of the case to see whether or not the patient truly possesses the conditions that were documented. Clinical validation is beyond the scope of DRG (coding) validation, and the skills of a certified coder. This type of review can only be performed by a clinician or may be performed by a clinician with approved coding credentials"

• This has been referenced by AHIMA and AHA coding clinic as well (Coding Clinic, Fourth Quarter ICD-10 2016 Pages: 147-149):

  • "Although ultimately related to the accuracy of the coding, clinical validation is a separate function from the coding process and clinical skill. The distinction is described in the Centers for Medicare & Medicaid (CMS) definition of clinical validation from the Recovery Audit Contractors Scope of Work document and cited in the AHIMA Practice Brief ("Clinical Validation: The Next Level of CDI") published in the August issue of JAHIMA: "Clinical validation is an additional process that may be performed along with DRG validation. Clinical validation involves a clinical review of the case to see whether or not the patient truly possesses the conditions that were documented in the medical record. Clinical validation is performed by a clinician (RN, CMD, or therapist). Clinical validation is beyond the scope of DRG (coding) validation, and the skills of a certified coder. This type of review can only be performed by a clinician or may be performed by a clinician with approved coding credentials."

• There may be state law rules that define the practice of medicine which could influence whether a clinical validation review must be performed by a licensed physician in your state. For example, some states define the practice of medicine as including any written or verbal opinion about a patient’s medical diagnosis.

• Medicare Advantage plans’ adverse clinical validation review findings often provide no indication of the qualifications of the reviewer at all. Some state the review was performed by a nurse with coding experience. Some indicate the review was performed “under the supervision” of a physician, suggesting the review was not personally performed by a physician.

#4. Their clinical criteria were not publicly available and
#5. They were not using high-quality clinical literature.

• In CMS's final rule 4201-F, page 22202 CMS states:

  • "Finally, in response to whether prior authorization policies or procedures that dictate specific definitions of medical diagnoses is considered more restrictive than Traditional Medicare, we consider coverage policies that dictate specific definitions of medical diagnoses to be additional coverage criteria that are only authorized in accordance with § 422.101(b)(6) as finalized in this rule."’

• If you then go to 422.101(b)(6) it states:

  • "MA organizations may create publicly accessible internal coverage criteria that are based on current evidence in widely used treatment guidelines or clinical literature when coverage criteria are not fully established in applicable Medicare statutes, regulations, NCDs or LCDs. Current, widely-used treatment guidelines are those developed by organizations representing clinical medical specialties, and refers to guidelines for the treatment of specific diseases or conditions. Acceptable clinical literature includes large, randomized controlled trials or prospective cohort studies with clear results, published in a peer-reviewed journal, and specifically designed to answer the relevant clinical question, or large systematic reviews or meta-analyses summarizing the literature of the specific clinical question."

    • Publicly accessible. For internal coverage policies, the MA organization must provide in a publicly accessible way the following:

      • (A) The internal coverage criteria in use and a summary of evidence that was considered during the development of the internal coverage criteria used to make medical necessity determinations;

      • (B) A list of the sources of such evidence; and

      • (C) An explanation of the rationale that supports the adoption of the coverage criteria used to make a medical necessity determination. When coverage criteria are not fully established as described in paragraph (6)(i)(A), the MA organization must identify the general provisions that are being supplemented or interpreted and explain how the additional criteria provide clinical benefits that are highly likely to outweigh any clinical harms, including from delayed or decreased access to items or services.

#6. We were never provided the criteria in advance.

Even if we were able to get a written copy of the definitions and criteria the plan was using - and many times, we aren’t given this documentation at all - many health plan contracts require the plan to make their policies available before a hospital is responsible for following them. If they were not made available until after we began receiving unfavorable decisions - or we were denied copies entirely - maybe there is contract language to support our efforts.

Making it count

The key to this is not necessarily including these arguments in the appeal itself, but keeping track of each one of these violations and submitting a complaint to CMS and even your managed care contracting department.

No, don’t group them together. For each individual case, file a complaint. Create templates and standardize the process so your hospital can efficiently file these complaints. Wait until you get a complaint number, then file the next. Then repeat. (Dr. Ronald Hirsch has discussed this process before)

Why does this matter? CMS complaints go into the MA plan’s star rating, which is worth hundreds of millions of dollars. (It’s such a big deal that multiple MA plans sued CMS over changes to how the star rating was calculated).

These are only a few examples of the arguments Richelle has discovered to assist in the fight against Medicare Advantage plans. She’s a unicorn in this space as a certified coder turned lawyer…and she’s creating an army of others just like her. I can’t recommend her enough.

That’s all for now. Don’t hesitate to ask questions as they help inspire future issues!

Cheers,

Robert

When you’re ready, there are 3 ways I can help you:

1. Join our CDI and Coding Village online community! It’s impossible to know it all, it takes a village! What you’ll get:

• Learn from me and other experienced professionals (perfect for new coders / CDI’s and those pivoting their career)

• Participate in monthly interactive webinars hosted by me, and anatomy and physiology reviews not found anywhere else

• Share ideas & solutions to new problems with peers in real-time discussions

2. Check out The Practical Guide to Attending Documentation video course.

• Stop underbilling your hard work & MASTER the new billing rules

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. The course that helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe:

The above represents my opinion only, and not that of my employer’s.

We’ve decided to keep the 7 day free trial option…for now. Don’t miss the opportunity to jump into our CDI and Coding community with a free trial!
This is what others have said this past week:

“Just tell me what to pick.”

I can’t do that.

“I know I know but… just tell me this time."

When I get questions about queries from physicians, the interaction often goes something like that. Physicians understand healthcare enough to know that I can’t tell them the answer because some rule somewhere says that I can’t. That “some rule” is the Guidelines for Achieving a Compliant Query Practice and we stick to it, as most CDI and coding professionals hold themselves to a high ethical and legal standard.

What Are Queries and Why Do We Need Them?

For inpatient physicians, queries need little introduction. For the uninitiated: a query is a communication tool designed to clarify documentation in the health record, typically sent by a CDI professional or coder. Think of it as our way of ensuring the integrity and accuracy of a diagnosis, procedure, or service code (In fact, the official guidelines essentially command us to do this). In essence, queries help us:

• Clarify Ambiguous Documentation: Because, sadly, coders aren’t allowed to “read between the lines.”

• Specify Implicit Diagnoses: When a condition is clinically evident, even if it isn’t explicitly named.

• Resolve Conflicts: For example, when a consultant documents that a patient has diastolic heart failure while the attending documented systolic. Is one or the other? Or both?

• Question Unsubstantiated Claims: Like documenting hypoxic respiratory failure when the patient was neither hypoxic nor placed on supplemental oxygen.

• Confirm Diagnoses: For example, if a radiology report mentions something the treating physician hasn’t documented.

These are just a few reasons. There are many are others but you get the picture.

Compliant Queries: Dos and Don’ts

What We Must Do:

• Resolve Discrepancies: We query the attending provider unless a specialist is the better subject matter expert. Otherwise, the attending is the tie breaker and has “the final say.”

• Support with Clinical Indicators: We include relevant clinical details from the medical record to back up the query.

• Offer Alternatives: We provide an “other” option so providers can craft an alternate, accurate response. If you don’t like an answer, by all means, give an “other” answer but please write out your reasoning.

  • This is a good opportunity to remind you that these are part of the legal medical record so… don’t respond to a query with curse words directed towards your neighborhood CDI nurse.

What We Can’t Do:

• Send Multiple Queries for the Same Issue: We can’t send repeated queries until we get the answer we want. We also can’t send the same query to multiple physicians and keep the answer we like.

• Lead: We cannot bold, underline, or otherwise nudge you towards a particular answer (so I definitely can’t tell you the answer). This is one CDI professionals take quite seriously and why an answer may not be clear, because they want to give you the freedom to choose based on your experience with the patient.

• Rely Solely on Past Encounters: Queries must be based on the current encounter, although we can use previous encounters to help inform current queries

  • I’m looking at you, surgeons, who tend to document “rest of medical history in the EMR.” Except for demographic information, the coding of a patient’s medical comorbidities reset to 0 within each inpatient encounter. Failing to capture these will sabotage your risk-adjusted quality metrics.

• Query Non-Treating Providers: We cannot code from pathology and radiology reports and we cannot query non-treating providers such as pathologists and radiologists. An interventional radiologist, however, would be fair game.

• Include Reimbursement or Quality Measure Impact: Talk about leading… Yeah can’t do that either.

• Mine the Record: We cannot “mine” the prior medical record to query for the current encounter.

What We Can Do:

• Use open-ended, multiple choice, or yes/no questions.

• Send queries after discharge.

• We do not have to include “unable to determine” as an option (except if we’re asking if something was present on admission or a yes/no query).

  • By the way, as I’ve written on many times before, queries exist to help you either directly or indirectly. So, it’s not in your best interest to always answer “unable to determine” just to get the query out of your way.

Beyond the Query

While educating our hospitalists about some conditions that we were under documenting, someone asked, “Why are our capture rates so low if y'all should just be querying us for these diagnoses?”

It was a fair question.
My response?
We do consider the fact that queries contribute to burnout, so we limit queries to those that directly impact reimbursement and quality metrics.

By the way, despite answering a query appropriately, you might be encouraged to document that diagnosis in your notes anyhow. This is neither a coding guideline nor a suggestion from the compliant query practice guideline. That recommendation is a result of insurances fighting to deny diagnoses that only exist in queries.

Lastly, a physician advisor may contact you to answer a query because hospitals must stick to certain timelines to bill. Certain details are needed for charts to be billed, especially for procedures/surgeries. So if the deadline is approaching, those unanswered queries often get “escalated” to your physician advisor.

Compliant Query Practice: Beginnings

I had the opportunity to interview Brian Murphy, the original director of ACDIS, who gave us rare insight into the early days of ACDIS and how he “extended an olive branch” to AHIMA to co-write and update the compliant query practice brief in the early 2010s.

Check out that episode here. The discussion starts at about 32:45.

That’s all for now. Don’t miss out on our 7 day free trial option of our CDI and Coding Village, and don’t hesitate to ask questions as they help inspire future issues!

Cheers,

Robert

Thanks to Laura Samson, RN BSN CCDS for reviewing this newsletter! And thanks to Alicia Whiteford and Robin Dunlop for their contributions!

When you’re ready, there are 3 ways I can help you:

1. Join our CDI and Coding Village online community! It’s impossible to know it all, it takes a village! What you’ll get:

• Learn from me and other experienced professionals (perfect for new coders / CDI’s and those pivoting their career)

• Participate in monthly interactive webinars hosted by me, weekly 1 minute CDI quick tip videos, and anatomy and physiology reviews not found anywhere else

• Share ideas & solutions to new problems with peers in real-time discussions

2. Check out The Practical Guide to Attending Documentation video course.

• Stop underbilling your hard work & MASTER the new billing rules

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. The course that helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe:

Starting TODAY, get a 7 day free trial of our CDI and Coding Village online community! This offer ends February 25!
Next webinar:

Story time:

1. A group of gastroenterologists felt their GI bleed average length of stay was higher than they expected. When they reviewed the charts, they suggested many were coded wrong.

2. A group of cardiologists felt their heart failure readmission rates were higher than they expected. When they reviewed the charts, they suggested many were coded wrong.

3. A 3rd party vendor emailed some members of hospital administration suggesting that when a patient’s hospitalization is extended due to something that developed after admission, we should recode it to that diagnosis.

Spoiler: They weren’t coded wrong and that 3rd point is absolutely wrong.

Why? Because those people didn’t understand the selection of the principal diagnoses… which is okay. Physicians and administrators aren’t supposed to be coders (but please consult your coding team before jumping to conclusions).

What is the principal diagnosis?

I suggest any administrator or physician who tracks clinical metrics understand it. It’s addressed on page 106 of the Official ICD-10-CM guidelines.

The principal diagnosis is “that condition established after study to be chiefly responsible for occasioning the admission of the patient to the hospital for care.”

“The admission” are the keywords here.

Why?

Because a condition that is not present at the time of admission cannot be coded as the principal diagnosis. They say it’s what “bought the bed.” Now that you know that simple coding 101 fact, you know why that 3rd point above was wrong.

Also, when two conditions are both present at the time of admission, that condition which requires more resources may be chosen as the principal diagnosis.

Why does principal diagnosis matter?

The principal diagnosis determines the DRG (except when a procedure is performed, then the procedure typically drives the DRG). I explained DRGs in a previous newsletter.

Design and development of the Diagnosis Related Group (DRG)

That’s important because many physicians track their metrics based on DRG, and MANY different codes “map to” that DRG.

Misconception Examples

Let’s go back to the gastroenterologists. They decided to have their metrics tracked by the typical DRG for GI bleeds with EGDs (DRGs 326 - 328).

The problem? Though they were doing their jobs well and scoping those patients within 24 hours, they were often staying in the hospital longer due to other issues such as heart failure or waiting for skilled nursing facility placement. But, those circumstances would not change the principal diagnosis and ultimate DRG assignment.

Why? Again:

1. A procedure would be considered “more resources” than some IV lasix.

2. Often the heart failure exacerbation was not present at the time of admission.

So, the GI bleed DRG was assigned every time (appropriately).

Speaking of heart failure, let’s talk about those cardiologists and their heart failure readmission rates. Ignoring some CMS oddities that I won’t bore you with, they had a different issue: dreaded combo codes and sequencing rules!

Another basic coding 101 fact? Coding assumes a relationship between some conditions such as:

• Hypertension and heart involvement

• Hypertension and kidney involvement

Even if you don’t say they’re linked, coding assumes they are (they will not be linked if you specifically document that they are not). Moreso, sequencing rules will often require that these combo codes get coded first (often resulting in the principal diagnosis).

How did that impact our cardiologists?

Patients were coming in with things like complications from their hypertension and chronic kidney disease. Their heart failure was chronic, stable, not in exacerbation, and not a main focus of care.

But because of the combo codes and sequencing rules?
BAM.
It was included in the principal diagnosis and mapped to the DRG by which they were being tracked.

Does it make clinical sense?
Not really.
Was it coded wrong?
Nope.

My suggestions to physicians

Understand how your metrics are tracked! Do not agree to metrics that determine your pay without fully understanding them. Demand that your lead coder is in the room when negotiating with your administration, or ask them lots of questions after the meetings. Use your CDI and coding teams as resources.

Consider having your metrics tracked by the principal diagnosis codes themselves, rather than DRG (therefore avoiding the “wide net” that DRGs cast). This is how the CMS AMI 30-day readmission cohort is tracked, which I discussed in a previous newsletter. 

And, of course, complete and accurate documentation is key.

• Ensure documentation is clear about what was present on admission, and what was not.

• Make sure documentation is clear about the main underlying etiology of symptoms.

• Limit copy / paste.

• Give a final diagnosis by the time of discharge (e.g. Chest pain vs GERD vs ACS. Acute right MCA infarct versus “neurological symptoms.”)

Know that coding is complex and confusing. It often doesn’t make sense clinically (and that’s not coders’ fault, they have to follow the guidelines). Coders go through lots of schooling and it takes YEARS of experience to get it right. Sure, it’s good to question things, but don’t jump to conclusions. And know that your health system routinely audits coders to ensure their accuracy.

Don’t miss out on a 7 day free trial of our CDI and Coding Village online community (offer expires Feb 25)!

That’s all for now. Don’t hesitate to ask questions as they help inspire future issues!

Cheers,

Robert

Thanks to fellow villagers Laura Samson, RN CCDS, Kelli Scardino, RHIT CCS, Penny Jefferson, RN CCDS, Alicia Whiteford, RN CCDS, and Revka Stearns, CCS CPC for their contributions to this newsletter!

When you’re ready, there are 3 ways I can help you:

1. Join our CDI and Coding Village online community! It’s impossible to know it all, it takes a village! What you’ll get:

• Learn from me and other experienced professionals (perfect for new coders / CDI’s and those pivoting their career)

• Participate in monthly interactive webinars hosted by me, weekly 1 minute CDI quick tip videos, and anatomy and physiology reviews not found anywhere else

• Share ideas & solutions to new problems with peers in real-time discussions

2. Check out The Practical Guide to Attending Documentation video course.

• Stop underbilling your hard work & MASTER the new billing rules

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. The course that helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe:

We’ve launched our “Village Advisor Program” where members of our CDI and Coding Village online community can give their own webinar and get paid.

The first will be by Laura Samson on Risk Adjustment. The 2nd will be by Penny Jenfferson on Elixhauser.

That’s not all.

7 days before the 1st presentation (TBD) we’ll have another exciting announcement. Stay tuned through my social media channels!

(Thanks to Dr. James Kennedy, who co-wrote this newsletter with me, and to hepatologist Dr. Steven Young for reviewing this newsletter)

One of the first questions I received as a new CDI physician advisor was, “Should I query for chronic hepatic failure?”

“Um…yes?…I think?” I thought to myself while being self-conscious about why that wasn’t an easy answer.

It wasn’t an easy answer because many physicians do not use the term “chronic hepatic failure.” Admittedly, we prefer to instead emphasize the consequences (hepatic encephalopathy, ascites, etc.) without mentioning the liver failure. This would be like documenting orthopnea, dyspnea on exertion, JVD, etc. without stating the patient has heart failure.

Therefore, physicians, pay attention. I’m going to get a bit deep into the ICD-10-CM coding here as many hospital systems (e.g. Value-Based Care) and physician groups (e.g., MIPS, ACOs) are increasingly focused on risk-adjusted “quality” and cost efficiency measures that are dependent on our documentation and its subsequent coding. So, you will start to hear more about this and see queries for acute, chronic, and decompensated (acute on chronic) chronic hepatic failure in the near future.

Physician Terminology Versus Coding

Some terms physicians generally use around liver conditions are:

• Acute

  • Liver injury (+ Acute) without specifying its underlying cause*

  • Acute Liver Failure

  • Fulminant Liver Failure (without specifying acuity)

  • Drug-Induced Liver Injury*

  • Obstructive Jaundice

  • Ischemic Liver Injury* / “Shock Liver”

  • Congestive Hepatopathy*

  • Alcoholic Hepatitis

  • Acute Viral Hepatitis

  • Hepatic Encephalopathy

  • “Transaminitis”*

• Chronic

  • Chronic Liver Disease (Ex: Hepatic Steatosis/Fatty Liver Disease, Chronic Viral Hepatitis)

  • Cirrhosis

    • Compensated (codes only as cirrhosis)

    • Decompensated (codes only as cirrhosis)

    • (not an exhaustive list)

Acute Liver Failure Versus Chronic

Physicians often reserve the term “liver failure” for only “acute liver failure” for patients without pre-existing liver disease who develop an acute liver injury with elevated LFTs (liver function tests), coagulopathy (INR >1.5), and acute hepatic encephalopathy. If the patient does not have those findings, especially encephalopathy, then the disease process is typically described by its acute condition (acute viral hepatitis, acute drug-induced liver injury, etc).

Otherwise, chronic conditions are listed by their chronic disease states but the term “failure” is typically not documented. The closest term to “chronic hepatic failure” physicians use is cirrhosis which is divided into:

• Compensated = No complications

• Decompensated = With complications (variceal bleeding, ascites, hepatic encephalopathy, hepatocellular carcinoma, etc.)

And coding?

For the chronic disease states such as viral hepatitis or fatty liver disease, there are individual codes for those.

For compensated cirrhosis, there’s cirrhosis codes with the underlying etiology.

BUT there is no code for “decompensated cirrhosis” and therefore this is generally captured by a cirrhosis code plus its associated complication (such as portal hypertension, ascites, etc.). Interestingly enough, “hepatic insufficiency” and “end-stage liver disease” are classified as chronic hepatic failure while “decompensated cirrhosis” is not. That’s just the way ICD-10-CM handles it.

The problem

The problem is that “chronic hepatic failure without coma” (e.g. K72.10) is a common diagnosis that highly impacts quality risk-adjusted metrics and HCCs (which i discussed in previous newsletters) which add to the severity of the underlying liver disease (e.g. alcoholic cirrhosis) or defined consequences (e.g. portal hypertension, esophageal varices).

Yet, while many of our patients have chronic hepatic failure, physicians are almost never going to document it unless they’ve been instructed to do so. So, our care will look worse “on paper” than in reality unless we address this.

Defining Chronic Hepatic Failure

If you asked most clinicians they’d likely equate “chronic hepatic failure” with “decompensated cirrhosis.” But, again, compensated and decompensated cirrhosis cannot be coded as hepatic failure.

Many hepatologists use the Child-Pugh classification which uses scores ranging from 5 to 15 to stage chronic liver disease as follows:

• Score 5 - 6 = Child-Pugh Class A

• Score 7 - 9 = Child-Pugh Class B

• Score 10 - 15 = Child-Pugh Class C

Per this UpToDate article, functional compromise starts at Child-Pugh Class B.

Per the 2023 Guidance on Hepatitis C, “Patients with a Child-Pugh score >7 have decompensated cirrhosis.” So, this gives us an evidence-based objective definition of chronic hepatic failure in patients with a Child-Pugh Class B score >7.

For those who want to argue for a more conservative definition, per the ICD-10-CM classification noted before, when a physician documents “End-Stage Liver Disease” it allows a coder to code “chronic hepatic failure.” As Child-Pugh Class B has a two-year survival rate of 60%, some may argue that’s not exactly “end-stage.” So, they may not feel comfortable equating Class B to chronic hepatic failure. But a 35% two-year survival rate with Class C (score >10) would almost certainly check that box.

Additionally, Dr. Kennedy has some thoughts on acute and acute on chronic hepatic failure:

Neither myself nor Dr. James Kennedy are official sources of guidance and you should defer to your institution’s clinical definitions and provider’s clinical judgment and documentation of these diseases. You may consider discussing this with a hepatologist.

Have some thoughts on this? Comment on my post about today’s newsletter on LinkedIn or Twitter (X)!

That’s all for now. Don’t hesitate to ask questions as they help inspire future issues!

Cheers,

Robert

Thanks again to Dr. James Kennedy, Dr. Steven Young, and additionally to Laura Samson, RN BSN CCDS for editing this newsletter!

When you’re ready, there are 3 ways I can help you:

1. Join our CDI and Coding Village online community! It’s impossible to know it all, it takes a village! What you’ll get:

• Learn from me and other experienced professionals (perfect for new coders / CDI’s and those pivoting their career)

• Participate in monthly interactive webinars hosted by me, weekly 1 minute CDI quick tip videos, and anatomy and physiology reviews not found anywhere else

• Share ideas & solutions to new problems with peers in real-time discussions

2. Check out The Practical Guide to Attending Documentation video course.

• Stop underbilling your hard work & MASTER the new billing rules

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. The course that helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe:

Are you a CDI or coder who struggles with cardiology? I’ve hosted two 1-hour webinars on cardiology topics (with reviews on anatomy, pathophysiology, CDI, coding and quality implications) which are available to view in our online community.

Next up: Pneumonias and other lung conditions on January 3rd at 9 AM CST.

In the Fall of 2023, I received a message from a cardiology fellow about one of my twitter posts. He argued that I had used the term “Type 2 NSTEMI” but “STEMI” (ST-segment elevated myocardial infarction) and “NSTEMI” (non ST-segment elevated myocardial infarction) were terms that could only be used for Type 1 MIs. We respectfully exchanged ideas, neither really convincing the other.

But then I saw other cardiologists arguing the exact same point on social media. Then, in June 2024, I received a message from a fellow CDI physician advisor. His colleagues were also arguing that point. This made me pause.

Was I wrong?

Should I stop using the term “Type 2 NSTEMI”? I still saw it used all the time in practice, by cardiologists and non-cardiologists alike.

Were we ALL wrong?

I immediately asked a cardiac electrophysiologist friend of mine (and assistant professor at UAB) who confirmed my understanding…phew…

So where’s the misunderstanding? Let’s take a step back.

How Do We Define a Myocardial Infarction?

According to the 4th Universal Definition of Myocardial Infarction (4UDMI), two key criteria must be met:

1. Evidence of acute myocardial injury

2. Signs or symptoms of ischemia

Both criteria are required for a diagnosis of MI. 

Breaking down the definitions

• Acute Myocardial Injury: This occurs when cardiac troponin levels rise above the 99th percentile upper reference limit (URL). A rise and/or fall in troponin levels indicates an acute injury, while an unchanging pattern suggests chronic injury. 

• Evidence of Ischemia: The 4UDMI outlines four potential indicators, any one of which suffices: 

  • Symptoms of acute myocardial ischemia (e.g., angina or anginal equivalents) 

  • New ischemic ECG changes 

  • Development of pathological Q waves 

  • Imaging evidence of new loss of viable myocardium or regional wall motion abnormality consistent with ischemia 

If both criteria are met, then the diagnosis of MI is supported. The distinction between Type 1 and Type 2 MI depends on the underlying etiology: 

• Type 1 MI results from atherothrombotic coronary artery disease, often triggered by plaque disruption (rupture or erosion). 

Source: 4th Universal Definition of MI

• Type 2 MI stems from an oxygen supply-demand mismatch. 

Source: 4th Universal Definition of MI

Where does NSTEMI fit in?

The terms STEMI and NSTEMI describe ECG findings rather than MI types. Neither ECG finding is unique to a type. In fact, ST-elevation can occur in 3–24% of Type 2 MIs according to the 4UDMI.

So, let’s address that cardiology fellow’s question. Can you meet the criteria for a myocardial infarction with supply-demand mismatch aka “Type 2 physiology?”

Absolutely.

And if there’s no ST-elevation on ECG, then it is, by definition, a non-ST segment elevated myocardial infarction, aka NSTEMI. So, a Type 2 NSTEMI.

But here’s the catch (and where his argument originated): 

The 4UDMI does NOT use the term “Type 2 NSTEMI.” It only uses “Type 2 MI.” So, even though it’s an accurate descriptor, should we use “Type 2 NSTEMI” in documentation? The answer is actually no—and not just to avoid irking cardiologists but rather because there are important quality reporting implications at play. 

Quality Implications of NSTEMIs

CMS tracks 30-day unplanned readmissions for acute myocardial infarctions (AMIs) under the Hospital Readmissions Reduction Program (HRRP). This cohort isn’t based on DRG assignment but rather on the principal diagnosis code of the index admission:

Source: https://qualitynet.cms.gov/inpatient/measures/readmission/methodology

Notice that the ICD-10-CM code I21.4 Non-ST elevation (NSTEMI) myocardial infarction is included in CMS’s AMI cohort, but I21.A1 Myocardial infarction Type 2 is not. The program is designed to track Type 1 MIs. So if providers document “NSTEMI” without specifying it as Type 2, it may be incorrectly coded as I21.4. This misclassification may negatively impact your hospital’s AMI readmission metrics.

To best mitigate this, encourage providers to document “Type 2 MI” explicitly and reserve “NSTEMI” for Type 1 MIs. CDI teams should query for clarification when necessary.

Other Common Issues: 

1. I5A Non-Ischemic Myocardial Injury (Non-Traumatic): This diagnosis is a CC (complication or comorbidity) whether acute or chronic, impacting DRG assignment and risk adjustment more significantly than terms like “troponinemia” or “troponin elevation.”

2. Overuse of NSTEMI: Providers frequently document “NSTEMI” in cases where troponin levels rise and/or fall due to supply-demand mismatch but without ischemic signs or symptoms. These cases meet the criteria for acute myocardial injury, not myocardial infarction. Given that Type 2 MI (I21.A1) is an MCC, it’s a prime target for clinical validation denials. Providers should avoid documenting myocardial infarction unless both 4UDMI criteria are met. 

Source: 4th Universal Definition of MI

Outdated Language

Behind all of this is a larger point. Many cardiologists have already moved away from the term “NSTEMI” and instead use NSTE-ACS (an umbrella term for a Type 1 event which includes NSTEMI and unstable angina).

…Doesn’t quite roll off the tongue, though. But if the coding and quality implications change, you know I’ll be here to keep you all updated.

Don’t forget to check out our CDI and Coding Village online community and access hours of recorded video content and attend monthly live webinars.

That’s all for now. Don’t hesitate to ask questions as they help inspire future issues!

Cheers,

Robert

Thanks to Dr. Sean Dunn (EP Cardiologist, Assistant Professor at UAB) for his help over the last year on this topic. Thanks to Erica Remer, MD, and Laura Samson, RN BSN CCDS for editing this newsletter!

When you’re ready, there are 3 ways I can help you:

1. Join our CDI and Coding Village online community! It’s impossible to know it all, it takes a village! What you’ll get:

• Learn from me and other experienced professionals (perfect for new coders / CDI’s and those pivoting their career)

• Participate in monthly interactive webinars hosted by me, weekly 1 minute CDI quick tip videos, and anatomy and physiology reviews not found anywhere else

• Share ideas & solutions to new problems with peers in real-time discussions

2. Check out The Practical Guide to Attending Documentation video course.

• Stop underbilling your hard work & MASTER the new billing rules

• Get 3.5 hours of CME credit! (Use your CME funds!)

• Stop feeling guilty about writing shorter notes

• Use notes to PREVENT getting sued

3. Check out The Resident Guide to Clinical Documentation video course. The course that helps you:

• Impress your attendings and improve your evaluations.

• Prepare for real-world productivity pressures

• Gain the confidence to write shorter yet more effective notes.

• Use notes to make you a better, more prepared physician

If you were forwarded this newsletter and would like to subscribe: