Confessions of a Vegetarian Coder—or, why I hate M.E.A.T.!!

First-- I hope I’m not offending anyone.   The spirit of this post is lighthearted, and I’m having a little bit of fun. BUT, I’m also serious about proper coding, and that’s my only intent.  Better documentation and coding helps us all in many ways, and should lead to better and more efficient medical care for the members of our medical groups and health plans. It also affects our reimbursement. Done correctly, though—a wealth of important information is obtained, to help our members to obtain access to excellent treatment programs, that enhance their lives and living.

So, there is NOTHING official about “MEAT”, it’s just a handy acronym that someone thought up one day.   It’s cute.   It sort of addresses what’s needed for ANY diagnosis, not just risk adjustment or so-called “HCC Coding” (which also doesn’t exist).  By law, under HIPAA, what is required for any ICD-9 code is defined by what is in the ICD-9-CM Official Guidelines for Coding and Reporting,  found here ,  and Coding Clinic (see the 2nd paragraph of the guidelines).  You can’t replace 107 pages with 4 letters, two of which mean the same thing!

If you look at MEAT for just a minute, you can see it’s not official, and pretty meaningless:

Monitoring (or Medication)

Evaluation

Assessment;  or

Treatment

Evaluation and Assessment mean exactly the same thing!  And no “Official Guideline”  for anything, much less something that leads to Federal Government payment, would be complete at 4 letters, 6 if you include the “or”.   

EVERY condition that is submitted as an ICD-9 code has to meet the Official Guidelines.  There’s no exception because something doesn’t risk adjust.  Just writing the word “Stable” does not magically mean you can code something.  Or just because the diagnosis listed under the “Assessment” heading in a chart note doesn’t mean someone assessed it.  What is under that word are conclusions, 90 times out of 100, not an evaluation of the patient or an assessment.   If you look at the word “evaluation” in dictionary.com, it says this:

evaluation

e•val•u•a•tion

[ih-val-yoo-ey-shuhn]

noun

1 an act or instance of evaluating or appraising.

2. (especially in medicine) a diagnosis or diagnostic study of a physical or mental condition.  (emphasis added)

Here’s where the word “meat” or “substance” comes into play. There really has to be some substance to an evaluation.   Just writing:

Diabetes with neurological manifestations

in a chart note does not mean the doctor evaluated ANYTHING or can code 250.60

First of all ________manifestations isn’t a diagnosis at all. By itself, it means nothing.  No matter what someone from somewhere told you, you cannot just write “stable” or “continue on meds” next to a couple of words and you now have something you can code.  I have seen it fail in CMS RADVs more than once.  And I think that failing something like that is proper.  It never should have been coded in the first place! 

On the other hand, if you can see in that chart note that the  physician did a foot exam, and documented the patients symptoms of burning or tingling in their feet, did a monofilament exam—NOW YOU HAVE SOMETHING!!!!!!!!!

Now he or she can say in the “Assessment” section:

Diabetes with diabetic polyneuropathy.  Patient started on Neurontin, 300 mg, TID

TA DA!! We have something that we can code!!!  YAY!!!!!!!!!!!!!  There is the "meat" that's needed.  An evaluation and/or some indication that the condition was either assessed, being treated (actively) or it has an impact on the treatment of other diseases as the ICD-9 requires.

When people talk about “HCC Coding” being somehow different, there isn’t really a citation they can point to, or an example.   On the other hand, CMS DOES have specific signature requirements for a medical record, but that isn’t a coding issue per se. It is a Medical Record issue.  So, in my diet, I’m a happy omnivore, savoring a juicy steak when I can.  But when it comes to “M.E.A.T.” and coding, I’m a vegetarian coder all the way.

Stacey Hernandez, CCS-P

Coding versus Interpreting

As coders, we're required to code based upon the physician's documentation, and not to read into what the physician has actually written. 

Recently, we received the following question and follow up questions in the Coding@scanhealthplan.com inbox. 

Original Question: 

I would like to know some opinions regarding to coding the following documentation by Physician:  “ Vascular Dementia”.   

Our Answer: 

Vascular dementia NOS is indexed to 290.40.  I don’t see any alternative code.

Then, we received a follow up:

Follow up Question: 

In my opinion I’ll use 290.40 plus 437.0.   Please see instructional notes “ Use additional code to identify cerebral atherosclerosis”.  I need your opinion.

Our Answer:

You only use an additional code if it’s documented. It’s not something you automatically do.  Based on your email, the physician only documented vascular dementia.  If he had said vascular dementia due to cerebral atherosclerosis, then you would code it.

 The instructional notes for 250.40 say:

Use additional code, to identify manifestation, as: 
        chronic kidney disease  (585.1-585.9) 
             diabetic: 
             nephropathy NOS  (583.81) 
             nephrosis  (581.81) 
             intercapillary glomerulosclerosis  (581.81) 
             Kimmelstiel-Wilson syndrome  (581.81) 
 But you don’t code all those things unless they are documented.
I hope that clarifies.

In response, the questioner indicated that they planned on coding cerebral atherosclerosis  (437.0) because:

"Vascular dementia, Arteriosclerotic Dementia, Multi-infarct Dementia and Atherosclerotic disease are synonyms.   In my opinion the documentation of Vascular Dementia is the same as  Atherosclerotic Dementia or multi-infarct Dementia.    

Please look Dementia
                                       Multi-infarct(cerebrovascular) ( see also Dementia, arteriosclerotic)"

 
I don't believe I did a good enough job explaining why you wouldn't code something not documented.  But the reality is that cerebral arteriosclerosis is not the only cause of vascular dementia, although it may be the most common cause.  Multi-infarct dementia (MID), due to multiple strokes or TIAs,  or mixed type due to MID and Alzheimers, or many other

 conditions that reduce blood flow to the brain, including certain autoimmune diseases (e.g., lupus eythematosus, temporal arteritis), certain inherited (genetic) diseases, infections of the heart (endocarditis), brain hemorrhage, profoundly low blood pressure can also cause vascular .   Since arteriosclerosis is not the sole cause, coding 437.0 without documentation of cerebral arteriosclerosis is not appropriate.  As coders, we must code what's documented, without inserting our knowledge of disease states or opinions into the process.

Full Text of the 2012 ICD-9 Now Available!

The full text of the 2012 ICD-9 Code book, effective October 1, 2012 is now available on our HCCUniversity.com website.  You can download the book, including the official guidelines here: 2012 ICD-9 .

When CMS posts the update to the CMS-HCC model, we'll post them on HCC University.

Have questions about risk adjustment or ICD-9-CM coding?  Contact us at coding@scanhealthplan.com.

2012 ICD-9-CM Guidelines Posted

The Official ICD-9-CM coding guidelines, effective 10-1-2011, were posted on the National Center For Health Statistics website.  A copy of the new guidelines will be posted on the Tools page at hccuniversity.com shortly.

Few changes are included, but there are some of note.  There are new sections added to describe:

• Appropriate coding of post-procedural infection and post-procedural septic shock (I.C.1.b.10.c.)
• Appropriate coding of types and stages of glaucoma (I.C.6.b.) and
• Guidelines for complications of care (I.C.17.f.1.)

The Myth of Coding the "Rule Out" Diagnosis

We get a lot of questions in our "Ask A Coder" inbox, coding@scanhealthplan.com.   There is no doubt that far and away, the most frequent question we get is some variation of "can I code all "rule-out" diagnoses for an inpatient"?

I've been coding a lot of years (about 25).  As a disclaimer, I have never been employed by a hospital as a coder --BUT--in my role at SCAN and other health plans, I've coded hundreds (maybe thousands) of inpatient charts. I can honestly tell you, the number of times I've coded a rule-out diagnosis can be counted on the fingers of both of my hands.  I'm sure that hospital coders have done so many more times than I have, but I'd bet they would tell you that this is not a common occurrence. 

Why isn't it?  Let's take a look at the Official Coding Guidelines themselves.  I've highlighted particularly important passages of the guideline.

H. Uncertain Diagnosis If the diagnosis documented at the time of discharge   is qualified as "probable", "suspected", "likely", "questionable", "possible", or "still to be ruled out", or other similar terms indicating uncertainty, code the condition as if it existed or was established. The bases for these guidelines are the diagnostic workup, arrangements for further workup or observation, and initial therapeutic approach that correspond most closely with the established diagnosis.    This guideline is applicable only to inpatient admissions to short-term, acute, long-term care and psychiatric hospitals.  

First, note that the condition is still probable, uncertain--i.e., not yet ruled out, then if the documentation indicates  at the time of discharge it is still a possible condition, then it is possible it can be coded.Second, you must have the entire hospital chart--why? Because you can't know if the treatment during the hospitalization was directed at that condition without it.

As to the related question of whether or not this can be used for the physician's record...the answer is NO.   Coding Clinic and the Official Guidelines indicate that physicians use the Outpatient coding guidelines, no matter what the place of service.

I hope this clarifies when coding a "rule-out" condition is acceptable.

What coding questions do you have?  Send them to us at coding@scanhealthplan.com.  We'll do our best to answer within 72 hours. We de-identify questions of general interest and post them on HCCUniversity.com under Ask A Coder.

What Diagnoses are not appropriate in a Physician Office Setting?

There are a number of diagnoses not likely to be treated in a doctor’s office.  Often, these are documented in the medical record and submitted to us.  These conditions are things like CVA (stroke), Acute Respiratory Failure, Sepsis, and Acute coronary syndrome and Acute MI that are normally treated in a hospital setting.

In almost all cases, these conditions are submitted based on a misunderstanding of ICD-9 rules.   Usually, the physician documents something like  “CVA” in the medical record, and then chooses the diagnosis code for CVA for submission to the health plan. 

The problem is that most patients who are having a stroke are not seen in a doctor’s office. It’s true that on rare occasions, the patient does appear in a doctor’s office with a life threatening condition, but usually the physician is trying to convey that this is a problem that occurred in the past.

Physicians generally document as a way of capturing what’s wrong with the patient.  But there is another reason—it can serve as a reminder for the physician what they thought about at the last visit.  So, if they see “CVA” or “acute respiratory failure” listed on the last note, they know that this is a patient who had a CVA or acute respiratory failure in the past.  

Acute MI is a little different, and coding for Acute MI is appropriate for 8 weeks after the event.  Because there’s a lot of confusion about acute MI coding, we’ll be doing a separate blog posting on this in the future.

Physicians need to be educated that there are codes for the history of many illnesses, most of them are in the “V” section of the ICD-9.  History of MI is an exception—it’s ICD-9 412.  History of something like pneumonia or septicemia can be coded to V12.09 (personal history of other infectious disease) if the physician documents the history of and considers it a significant diagnosis.

Documentation of these historical conditions should make it clear that it is a disease that occurred in the past. Any condition that is no longer being treated should be noted to be a history of the condition.  Where a code for the history of the disease exists, it should be submitted using that code, not a code for the acute condition.

Coders need to be aware that even when a physician documents one of these conditions, they should not code them. It’s important to educate the physician and explain that these acute conditions should be documented and submitted as a “history of” code.

Do you have a have a confusing coding issue that you're dealing with? Drop us a line at coding@scanhealthplan.com, or leave us a comment and we'll do what we can to help!