Wednesday, October 31, 2012

Inpatient Coding Guidelines

We often get questions about when to use the Inpatient Coding Guidleines--can they be used for a discharge summary alone, or a consultation that took place during an inpatient stay?  According to CMS' RADV Q&A from 2009, the answer is NO:

Q15: Are practitioner visits during a hospital admission acceptable as "PHYSICIAN" records? If yes, what coding rules will apply to these records?
A15: If a member has an inpatient hospital discharge that supports the HCC, it is usually best to select the inpatient discharge and submit the entire inpatient medical record for coding. However, if the entire inpatient medical record cannot be obtained, the organization can submit medical record documentation from an inpatient physician visit for review and it would be reviewed in accordance with the Diagnostic Coding and Reporting Guidelines for Outpatient Services. When submitting these forms of documentation please note the following:
In the outpatient setting, coders do not code diagnoses documented as "probable" "suspected," questionable," or "rule out" but rather coders code the condition to the highest degree of certainty for that encounter/visit (i.e., symptoms, signs, abnormal rest results.) This limited documentation may not support the HCC. Acceptable inpatient physician visit medical records are: inpatient history and physical examinations, progress notes, consultation reports, and discharge summaries. When submitting medical record documentation from an inpatient physician visit, the organization has two options:
1. Select a service date from the stored risk adjustment data listed in Section 3A of the coversheet (i.e., RAPS data) for a PHYSICIAN visit. The RAPS record most likely was for the physician claim for inpatient visit services. Be sure that the record you are submitting exactly matches the date of the selected service date. For example, the coversheet is
checked with a service date of 9/5/2003 through 9/5/2003 and a signed inpatient physician consultation report dated 9/5/2003 is attached for review.
2. Submit an "in lieu of" medical record by completing Section 3B of the coversheet for a PHYSICIAN visit. Be sure that the record you are submitting exactly matches the date of the selected service date. For example, Section 3B of the coversheet has a service date of 10/3/2003 through 10/3/2003 and a signed inpatient physician admission history and physical examination report dated 10/3/2003 is attached for review.
The reason for this is the nature of the Inpatient Coding Guidelines for coding uncertain diagnoses: 

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

As you can see, this guideline requires that the diagnostic workup and initial therapeutic approach correspond to that uncertain diagnosis.  Simply saying "rule out MI" in a consultation note, or even a discharge summary, does not show that diagnostic workup or therapeutic approach.  You need the entire chart (or the majority of it) to support this.  So, standalone documents from an inpatient stay are always coded as if they are outpatient documents.

Have a coding question related to risk adjustment, or risk adjustment data validation (RADV)?  Send it to us at

Wednesday, October 24, 2012

OIG Posts Video-- Outlook 2013

This morning, the Office of the Inspector General posted this video on their Outlook 2013 page. 
OIG senior executives discuss emerging trends in combating fraud, waste, and abuse in Federal health care programs, and the OIG's top priorities for 2013.   In addition, they discuss new and ongoing projects from the OIG's Work Plan.
We'll be posting about the OIG Work Plan in the next few days.

Monday, October 15, 2012

Where are the New ICD-9 Codes?

Since we're already halfway through October, some people are starting to wonder where the new  ICD-9 codes are.   By this time, they're usually up on the website for people to download.

Because of the transition to ICD-10 in October 2014, there is a moratorium on creation of new codes, except in very limited circumstances.  So, there are no new ICD-9-CM diagnosis codes for 2013.  There is also no change to instructions or the Official Coding guidelines, which were effective October 1, 2011.

There is one change-a single ICD-9 procedure code (used only by inpatient hospitals), 00.95, Injection or infusion of glucarpidase, was added.   This drug is used to treat a common side effect of methotrexate administration.

If there are any future changes to the ICD-9, we'll post it here and on HCC University.

Tuesday, October 9, 2012

HHS and DOJ Issue Joint Letter on Cloning of Medical Records

On September 24, 2012, Secretary Katherine Sebelius of the Dept. of Health and Human Services, and Eric Holder, Attorney General, of the Department of Justice, issued a joint letter to several health care associations, including the American Hospital Association and Federation of American Hospitals. You can view a copy of the letter on HCC University, on our Tools page.

Although the letter applauds the widespread adoption of EHRs, it notes that there are indications that some providers are using them to clone records and to game the system.  The letter points out that this type of false is not only dangerous to the patient, but also illegal. 

The letter notes that CMS is stepping up its medical record review activities to identify these issues, and that the DOJ, FBI and other law enforcement are monitoring these issues and will take action when warranted.

So--how do CMS, the DOJ and others identify cloned medical records?   Cloned records are often very obvious.  Although they may occur in handwritten records, it is most often found in EMRs with cut and paste and template functionality. 

Although templates are not inherently bad, auto-populating them indiscriminately with information can lead to obviously cloned records. While some things may not change significantly from visit to visit (for example, the bulk of the patient's history may remain the same), a patient's condition is rarely, if ever, static.  A patient's interim history, vital signs, symptoms and treatment are almost never exactly the same. Copying and pasting these components leads to inaccurate medical records.

Another area where cloning becomes obvious is in the assessment.  Often, previously treated or self-limiting diseases remain in the assessment for months or even years.  We've seen charts where a diagnosis of "acute sinusitis" continued on for over a year, with no treatment after the initial date of service where it appeared.

It's important to remember that once one part of the medical record is obviously wrong, and auditor is likely to discount the record in its entirety, since it's impossible to tell what is and isn't accurate.

Clinicians using EMR should use templates wisely, and cut and paste functionality very sparingly.  All information in the chart should accurately reflect the patient's complaints, conditions, and treatments on the current date of service.