Home

Focus on documentation for new ICD-9 codes

Association of Clinical Documentation Improvement Specialists, August 6, 2008

Education among coders and physicians will help prevent a flood of physician queries that might otherwise result when the 2009 IPPS code changes take effect. “It’s great to have the new codes, but if we don’t have the documentation, we can’t use them,” says Kathryn DeVault, RHIA, CCS, manager of professional practice resources at the American Health Information Management Association (AHIMA) in Chicago. “The landscape has changed so much. We’re coding for data, coding for reimbursement, coding for profiling. You really need to show that specificity for all of these tasks.”

Document cause to code secondary diabetes

Some of the proposed code sets could present particularly difficult documentation requirements. For example, secondary diabetes (249.xx) might be a huge challenge to code, DeVault says.

“We have a hard time getting physicians to document type 1 and type 2 [diabetes],” she says. “I would suspect that a lot of diabetes we see as an ICD-9-CM code 250.xx might actually be secondary diabetes, but we don’t know that because there’s no documentation.”

Provide coders with clinical education about secondary diabetes, and communicate to physicians the new secondary diabetes codes so they know what additional documentation they need to supply, DeVault says.

For example, patients with cystic fibrosis or malignant neoplasms can have secondary diabetes, so facilities need to strive for thoroughly documented medical histories and urge physicians to make connections in their notes between other conditions and secondary diabetes when appropriate, DeVault says.

“We need to look more carefully at the patient’s medication history,” she says. “Coders can’t assume secondary diabetes just because something else such as Cushing’s disease or cystic fibrosis is present.”

Physicians must specifically note the tie between another condition or medication and diabetes in their documentation so coders can assign the correct codes.

Let’s say the physician documents “diabetes with chronic kidney disease.” Even if the patient has a history of steroid use that may have caused secondary diabetes, without documentation that states this specific connection, coders should not assign secondary diabetes codes, DeVault says.

Guidelines that the ICD-9-CM Cooperating Parties (American Hospital Association [AHA], AHIMA, CMS, and the National Center for Health Statistics) will issue in November will clarify how to appropriately assign the secondary diabetes codes. With all of the changes on the way, it would be wise to get a head start on educating your medical staff and coders.

Consider a case in which a patient has ovarian cancer, and she has had chemotherapy and chronic medication. The physician also documents stage 2 chronic kidney disease due to diabetes. According to this documentation, you should code the malignant neoplasm, chronic kidney disease, and diabetes as an ICD-9 code 250, which is primary diabetes.

If there is no documentation that states diabetes resulted from the treatment of the neoplasm, you should not assume a relationship between the diabetes and either the carcinoma or medications and, therefore, you should not code secondary diabetes.

Pay close attention to list of HACs

The addition of a specific code for ventilator-associated pneumonia (997.31) will also be of significance because it falls under the hospital-acquired condition (HAC) provision.

Facilities need to focus on HACs because Medicare will no longer recognize a complication and comorbidity (CC) or secondary diagnosis for higher reimbursement if that condition was not present on admission (POA).

Coders shouldn’t assume that because a patient is on a ventilator and has pneumonia, they can code ventilator-associated pneumonia. “The coders should be relying on physicians to make the linkage between the condition and the procedure. They need to document that connection,” says Nelly Leon-Chisen, RHIA, director of coding and classification at the AHA in Chicago. “You don’t know the sequence of events, and it’s not for the coder to make those decisions.”

Also, look at new codes for methicillin-resistant Staphylococcus aureus (MRSA) infections. The new codes, which will differentiate MRSA from other infections, include:

038.12—Methicillin-resistant Staphylococcus aureus septicemia

041.12—Methicillin-resistant Staphylococcus aureus

Coders need to look in the physician documentation to differentiate between MRSA determined to be present in a patient’s screening versus cases in which a patient has a full-blown MRSA infection, Leon-Chisen says.

For example, a nasal swab in a routine screening could show MRSA, but the patient might not have an infection. On the other hand, you could have a patient who has pneumonia due to MRSA, and this is the bacteria causing the infection.

To use the new MRSA code, the physician needs to make the connection in his or her documentation between the infection and MRSA.

And just because a patient tests positive for an infection doesn’t mean it is a MRSA infection. Coders should query about the type of infection if the physician did not provide the information initially. Coders need to clarify whether the MRSA culture is from the specific infection site or simply part of an initial patient screening on admission, says Leon-Chisen.

“Perhaps the patient developed something else, and this infection could be something entirely different,” she explains.

Clarify where patient is in disease process

The most difficult part of assigning the new leukemia (203.xx–208.xx), multiple myeloma (203.02), or sarcoma in relapse (205.32) codes is obtaining adequate physician documentation on where the patient is in the disease process. The proposed changes mean physicians must now state whether the patient is still in the active status of the disease or in remission. For example, physicians may document that a patient has lymphoma but neglect to state where the patient is in the disease process. Convey to physicians that although it is easy to overlook this detail, they need to begin including it, DeVault says.

The new code sets for these diseases now have a fifth digit, identifying those conditions that are in relapse. The ICD-9-CM Cooperating Parties’ intention is that these more-specific codes will help distinguish between cases in which the condition is under control versus cases in which the condition is still active and the patient is not able to achieve remission.

Urge coders and physicians to communicate

This wave of new codes presents an opportunity for coders and physicians to work together to head off the potential barrage of physician queries that might otherwise occur.

“Educating physicians is a big challenge, but you have to do it because it’s the only way you can use these specific codes,” Leon-Chisen says. “They need to know what to document so they don’t have to be queried every time for the same thing.”

Consider the following ways to help open lines of communication between the two parties:

Have coders sit in on physician meetings to give the physicians an overview of the new codes and the dates they take effect. This would also present an opportunity for physicians to provide coders with education on the clinical aspects of the codes.

Provide information on new codes in physician or medical staff newsletters. “We don’t need [physicians] to learn the codes, but we need them to understand that we are looking for greater specificity in their documentation,” DeVault says. For example, now that CMS proposed ICD-9 codes for different stages of pressure ulcers, physicians must define the severity of illness by noting the stage of the pressure ulcer in their documentation.

Distribute educational handouts in physicians’ mailboxes that list new codes and explain why detailed documentation is increasingly important.

Invite a physician (e.g., surgeon, cardiologist, or cardiothoracic surgeon) to a lunch meeting with your coders, and provide the physician in advance with a list of clinical or anatomy questions, for example.

This gives coders a chance to ask about procedures they might find difficult to code or to address aspects of that particular physician’s documentation practices that they find confusing.

“Then, once they’ve met you, they see you in the hospital, and you’ve bridged that gap a little bit and they might be less likely to put off your queries,” DeVault says.

Concerning procedure coding, you may need to negotiate between two different “languages.” Many physicians code according to CPT language, whereas inpatient coders code in ICD-9-CM language.

And it can be a challenge to persuade physicians to try to speak ICD-9 for the coders. “It can’t just be a one-step approach of just going to meetings,” DeVault says. “You have to provide outreach in creative ways and present different avenues of education.”

Show physicians the value of the more-specific documentation you are asking them to provide. Explain how specificity in documentation affects not only hospital coding, but also practice coding, DeVault says.

It might be a good idea to remind the physicians that good documentation and accurate coding not only affect quality reporting for the institution but can affect individual physician profiles as well, Leon-Chisen says.

Encourage your coders and physicians to communicate to reduce the potential for careless coding errors.

“You don’t want coders assigning complications because they misread something or didn’t get the right information in the documentation,” Leon-Chisen says. “You don’t want coders putting down ‘no’ for POAs because something wasn’t documented properly.”

Editor’s note: You can contact DeVault by e-mailing her at Kathryn.devault@ahima.org.

Leon-Chisen can be reached at nleon@aha.org.

Most Popular