A. No. The Inpatient Coding Guidelines are for use by the facility only. Physicians use the Outpatient guidelines, no matter what the place of service is.

A. The ICD-9 does not categorize hypertension by level of control. Hypertension is categorized as benign or malignant. If a note indicates only “uncontrolled hypertension”, the correct coding is 401.9, “hypertension, unspecified”. If there is hypertensive heart or kidney disease present, these should be noted and coded in the 402.X or 403.X range. You should not code 401.1 unless your documentation says that the hypertension is benign.

A. No, the correct code for patients on long term opioid treatment is V58.69 – long term (current) use of other medication. Code 304.0, based on the DSM-IV (Diagnostic and Statistical Manual- 4th revision) is to be used for dependent drug abuse, not the prescribed use of medication, even if the patient has a physical dependence on the drug. Per CMS, clinicians should continue to use the DSM descriptors under HIPAA. See the DSM describes dependence as follows:

DSM-IV defines dependence as:

• A maladaptive pattern of substance use, leading to clinically significant impairment or distress, as manifested by three (or more) of the following, occurring at any time in the same 12-month period:
  1. tolerance, as defined by either of the following:
     • a need for markedly increased amounts of the substance to achieve intoxication or desired effect
     • markedly diminished effect with continued use of the same amount of substance
  2. withdrawal, as manifested by either of the following:
     • the characteristic withdrawal syndrome for the substance
     • the same (or a closely related) substance is taken to relieve or avoid withdrawal symptoms
  3. the substance is often taken in larger amounts or over a longer period than was intended
  4. there is a persistent desire or unsuccessful efforts to cut down or control substance use
  5. a great deal of time is spent in activities to obtain the substance, use the substance, or recover from its effects
  6. important social, occupational or recreational activities are given up or reduced because of substance use
  7. the substance use is continued despite knowledge of having a persistent or recurrent physical or psychological problem that is likely to have been caused or exacerbated by the substance (e.g., continued drinking despite recognition that an ulcer was made worse by alcohol consumption)

DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, ed. 4. Washington DC: American Psychiatric Association (APA). 1994.] (For documentation regarding the continued use of DSM Descriptors, please see the Frequently Asked Questions Database on the CMS website at: http://questions.cms.hhs.gov/ Search term “DSM-IV”)

A. If the only documentation is “CVA” the default code is 434.91 (CVA –stroke, ischemic). However, if the physician specifies the type of stroke (embolic, hemorrhagic, etc) there are separate codes in the 434.XX section of the ICD-9.

However, these codes are for use during the acute event—i.e., while the patient is still hospitalized. Once the patient has been discharged from the hospital, then coding and documentation should indicate a history of stroke (V12.89). More importantly, sequelae, what ICD-9 calls “late effects” should be documented and coded. Late effects such as hemiplegia/hemiparesis secondary to CVA (438.2X), aphasia (438.11) should be documented and coded. (Coding Clinic, Q4 2004)

A. No, you cannot code a diagnosis if you cannot determine who the rendering the provider is. CMS will only accept diagnoses from complete documentation, which must have the diagnosis documented, as well as have the patient's name, date of service, a unique patient identifier (e.g. date of birth or medical record number) and be performed and legibly signed by an acceptable rendering provider (e.g., physician, physician extender, inpatient hospital, or outpatient hospital).

A. No, it is not upcoding or overcoding for two reasons:

1. The Coding Clinic calls the use of V58.67 optional ("if desired") (Coding Clinic 2004 Q4)
2. It doesn't lead to a higher or extra risk score. Both V58.67 and 250.00 or 250.01 group to HCC 19. You don't get paid for HCC 19 twice no matter how many times that you submit the code during the collection period.

There is no reason not to code all applicable diagnoses, since they give CMS a more complete picture of our members. Many diagnoses not in the Risk Adjustment model are part of the RxHCC (Part D) model, so it is important to code all diagnoses, not just diagnoses that lead to an HCC.

A. Yes, submit the hospitalist’s data to the health plan. Follow the physician/outpatient coding guidelines (Coding Clinic 2000 Q3). All physician services should be submitted to the health plan and the plan submits them to CMS.

A. No, hospice enrollees are not paid using the Risk Adjustment method.

Please see pgs 23-24 in the CMS Managed Care Manual, which can be downloaded at http://www.cms.hhs.gov/manuals/downloads/mc86c08.pdf regarding payment for hospice enrollees.

Also, payment lags 1 year behind the service in the risk adjustment model, and you don't receive payment for a patient who expires. The nature of hospice (a physician certifies that the patient has less than 6 months to live) means that most patients won't be alive in the payment year. A patient must be a member in the payment year in order for the group to receive payment.

A. Under the Official ICD-9 Coding Guidelines, a diagnosis can only be coded when it is explicitly spelled out in the medical record. It cannot be inferred (even when a provider does the coding) that depression documented multiple times in a record is "major recurrent depression." Also, the Diagnostic Coding and Reporting Guidelines for Outpatient Services, pgs 88-91, explain that a diagnosis is often not established during the first visit and it may take subsequent visits to confirm that diagnosis. All diagnoses should be supported by physician documentation.

The physician should clearly document the type of depression in order to assign a more specific diagnosis code such as major depressive disorder. If only depression is documented, code 311 "depression not otherwise specified."

Official ICD-9 Coding Guidelines can be downloaded at: http://www.cdc.gov/nchs/datawh/ftpserv/ftpicd9/icdguide07.pdf.

A. CMS allows physicians to create an addendum to medical records to reflect confirmation of a diagnosis under certain circumstances. However, the example given by CMS is several days, not weeks, after a visit. Most laboratory and radiology services have results within days, not weeks, and it is expected that addenda will be completed as soon as possible after receipt of the additional information. CMS published this guidance in a training guide that can be downloaded at: http://www.csscoperations.com/new/usergroup/2007raps/ra-particpantguide_120607.pdf.

It reads as follows:

6.4.2 Unconfirmed Diagnoses
Physicians and hospital outpatient departments shall not code diagnoses documented as “probable,” “suspected,” “questionable,” “rule out,” or “working.” Rather, the condition(s) shall be coded to the highest degree of certainty known for that encounter/visit, such as symptoms, signs, abnormal test results, or other reason for the visit. CMS recognizes that this is an area where the physician-reported diagnosis and hospital inpatient diagnosis for the same encounter may disagree since hospital inpatient rules allow for coding of suspected conditions as if they were confirmed.

It also is understood that the physician record is not a static document. Positive test results and notation regarding contact with the patient for a revised plan of treatment often are added to the record several days after the patient encounter. When these addenda are made, corrections or additions to the diagnoses submitted to the MA organization may be recommended especially if the HCC assignment is impacted.

Example: 4
A physician removes a mole during an office visit and sends the specimen for pathology. The diagnoses documented are “suspicious skin lesion” (709.9, not in model) and “rule out melanoma.” At this point, the diagnosis 709.9 may be submitted, but the diagnosis of melanoma may not. The pathology report is returned several days later and confirms malignant melanoma. The physician reviews the findings, initials the report, and documents in the record the results and notification to the patient. Since the removal of the mole was done during the office visit, the new code (172.9, melanoma) should be submitted with that date of service.

A. The ICD-9 indexes the term "neuropathy" to 355.9. The physician’s specialty or type of provider does not influence coding, since one can't infer anything when coding. Even though a podiatrist most likely means polyneuropathy, we can't use that knowledge to choose a code. 355.9 is neuropathy NOS and is the correct code if any provider documents "neuropathy."

A. No, it is not acceptable to report the 2 additional diagnoses from the “chart review.” Coding from “chart reviews,” “failed visits,” or any other progress notes that are not from a face-to-face visit with a patient are not valid sources for risk adjustment coding. This guideline is in the CMS Risk Adjustment Participant’s Guide (section 7.2.4.5), which can be downloaded at: http://www.csscoperations.com/new/usergroup/2007raps/ra-particpantguide_120607.pdf.

A. The coding guidelines for late effects and acute fractures vs. aftercare of fractures is below. The Official Coding Guidelines can be downloaded at: http://www.cdc.gov/nchs/datawh/ftpserv/ftpicd9/ftpicd9.htm#guidelines.

There isn’t quite enough information to precisely answer the question, but I’ll give you enough information so you can determine what the correct 5th digit is. I assume by the way your question is phrased that the physician did not state the area of the heart muscle affected by the MI, which led you to the fourth digit of “9”.

Generally, for a patient being seen in the office it won’t ever be a fifth digit of 1.

First, the diagnosis code will always be determined by the documentation in the physician record.

The fifth-digit of “0” (unspecified episode of care) means the physician did not provide enough information to determine what episode of care the patient is in.

The fifth-digit of "1" (initial episode of care) covers all care provided to a newly diagnosed myocardial infarction patient until the patient is discharged from medical care (i.e., discharged from the hospital). This includes any transfers to and from other facilities prior to the patient's discharge and occurring within the eight-week time frame.

The fifth-digit of "2" (subsequent episode of care) covers care (further observation, evaluation or treatment) rendered after the initial treatment (discharge), but the myocardial infarction is still less than 8 weeks old.

Once the MI is more than 8 weeks old, the physician should document old MI (ICD-9-CM 412)

A. The correct code for coronary artery disease depends on what information is provided in the documentation. CAD is commonly miscoded as 414.00 (coronary atherosclerosis of unspecified type of vessel, native or graft). However, this is incorrect unless the physician specifically notes that both native and graft vessels are present. Hence, the documentation must include information regarding a previous coronary artery bypass graft (CABG) in order to use this code. When there is no documentation of a prior CABG, then the correct code for CAD is 414.01, coronary atherosclerosis of native coronary artery even though the physician does not state “native artery.” This rule was published in Coding Clinic For ICD-9 CM 2nd quarter, 1995 and reiterated in 1st quarter, 2004.

A. If the documentation indicates only diabetes and peripheral vascular disease, then the documentation is 250.00 (uncomplicated type II diabetes) and 443.9 Peripheral vascular disease, unspecified. Unless the physician specifically states that the peripheral vascular disease is a complication of the diabetes then the two diseases are considered unrelated.

A. Pathologic fractures of the vertebrae are coded 733.13.Although the Official Coding Guidelines indicate that the condition should not be coded on an ongoing basis, a recent Coding Clinic (Q3 2008) indicates that if the patient is being treated for a non-healing pathologic fracture, it can be coded as often as treated.

A. ICD-9 does not generally assume a cause and effect relationship between diabetes and other illnesses. Because the physician did not state a causal relationship, through terms like "diabetic retinopathy", or “retinopathy secondary to diabetes", the correct coding is 362.10, background retinopathy NOS and, if the note indicates that the patient is diabetic, 250.00, DM II, not stated to be uncontrolled.

A. If the documentation indicates only diabetes and peripheral neuropathy, then the documentation is 250.00 (uncomplicated type II diabetes) and 357.2 diabetic polyneuropathy. Unless the physician specifically states that the polyneuropathy is a complication of the diabetes, or documents “diabetes with polyneuropathy”, then the two diseases are considered unrelated. This is a change from the longstanding Coding Clinic advice that the physician must state the causal relationship.

A. Signature logs are not the same as a signature stamp. A signature stamp replaces a physical signature. As of 1-1-09, CMS does not accept signature stamps under any circumstances.

CMS has said (on a conference call) that they will accept signature logs if they are a normal part of the provider’s practice. The biggest issue we see with signature logs is that the signatures on them rarely match the signature in the record. In that case, they are useless.

A. Providers do not have risk scores. Certain of our reports show the risk average risk score of the members assigned to a PCP, or members seen by a given specialty, but these are always a reflection of the members’ risk scores.

Most physician extenders are not individually credentialed by health plans. IF they are identified as the rendering provider on incoming encounters (something that doesn’t happen consistently) it might be possible to determine the average risk score of members seen by the physician extender. But, it would not be a unique score, since there would always be overlap with the PCP that the members are assigned to.

A. No, CMS specifically says that superbills are not a part of the medical record. See the RAPS training participant guide, page 134 at:

http://www.csscoperations.com/new/usergroup/2007raps/ra-particpantguide_120607.pdf

A. What a provider can code is always driven by two things:

• What the provider documented
• What the ICD-9 coding rules allow

The provider cannot code for any condition he did not document himself, irrespective of what the patient was hospitalized for.

There is no coding rule which says that these are mutually exclusive conditions. If your provider’s documentation supports both conditions, then both can be coded.

A. It’s hard to determine what the patient is being seen for. All of the conditions listed could only be coded as a “history of” IF they were assessed. What should be coded is what the patient was seen and assessed for today. None of these things listed seem to be current conditions.

Risk Adjustment never changes what it is appropriate to code. What we have to be mindful of is what physicians (especially) have often forgotten to code—current conditions that they assess, but don’t mark on a superbill ; and what physicians have often forgotten to DOCUMENT, so that they can be reminded to do so. For example physicians often have assessed long term chronic problems (like the labs on an A fib patient, or the cardiac status and evaluation for peripheral edema in a CHF patient) but have forgotten to write the diagnosis for these conditions. Education focused on documentation is critical so that physicians are accurately recording what they assess.

A. From a coding perspective, CAD can be “related to” or caused by or secondary to diabetes only if the physician says so in their documentation. This is why we strongly recommend that physicians be trained in documentation. It doesn’t matter if the physician thinks that the diabetes caused the CAD if they don’t say so. The coder cannot make that assumption.

If the physician documents CAD secondary to the diabetes, then the correct coding would be 414.01 (see earlier question in “Ask a Coder”) and 250.8X – diabetes with other specified manifestations.

A. Yes. You should also use V58.61 if long term use of anti-coagulants is documented.

A. No. Per the coding guidelines, you do not code conditions that no longer exist. From the Official ICD-9 Coding Guidelines:

Section IV. Diagnostic Coding and Reporting Guidelines for Outpatient Services

K. Code all documented conditions that coexist

A. In the first scenario, no, CMS doesn’t downcode it. It is just processed as submitted. Of course, it must be documented in the medical record appropriately.

In the second scenario, again, CMS does not change the code, but would not allow it in a Risk Adjustment Data Validation. If you identify that the physician didn’t document a complication, then you should submit an adjustment transaction to SCAN with 250.00, diabetes without complications.

A. In the case of “ETOH ½ pint Vodka everyday”, nothing can be coded. A coder can’t make any diagnosis, and the doctor has just stated how much the patient drinks. He/she hasn’t said that there’s abuse OR dependence (the codes you are asking about are dependence (i.e., addiction) codes). In order to code addiction/dependence, the physician must document that the patient is addicted or dependent on alcohol. In order to use the fifth digit of “1” the physician must indicate that continuous drunkenness is present.

In the case of the statement “ETOH abuse”, then you can only code 305.00, Nondependent alcohol abuse, unspecified drunkenness. The physician has stated that abuse (not dependence) is present. The physician has not specified continuous or episodic drunkenness, so only the unspecified fifth digit of “0” is appropriate.

A. The physician clearly indicated that there is a history of osteomyelitis. It’s unlikely that the patient would still have a bone infection 58 years later. The correct coding would be 736.81, acquired unequal leg length, and V13.59, personal history of other musculoskeletal disorders.

A. In the second instance, it appears as if you’re saying that the computer read-out from the EKG says that there’s septal infarct, then this cannot be coded. The physician MUST interpret the EKG in order for any coding to take place. I’m also concerned about the statement that the physician “scratches through” the terms. All medical record changes should be done by a single strike through.

If the physician has indeed indicated that he disagrees with an EKG reading, then the abnormality would never be coded. Note, as above, if it’s a computer reading, it cannot be coded from in any case.

If in the first instance, you’re also referencing a computerized EKG reading, remember that all test results must be interpreted by the physician before coding can happen.

A. Unless the physician states that the patient has hemiparesis due to/secondary to the CVA, it cannot be coded. CMS Risk Adjustment follows all coding rules and guidelines, including those in Coding Clinic. This issue was addressed in Q1 2005, and the question asked if weakness secondary to a CVA should be coded as hemiparesis secondary to CV Physicians should be educated about the need for clear documentation for ICD-9 coding purposes. This is one of the scenarios I use in training, since it’s so commonplace.

A. It's appropriate to report angina post stenting only if the patient still has angin