2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
When CMS switched from ICD-9 to ICD-10 last year, the new diagnosis classification list allowed coders to translate additional elements of doctors’ language to billers’ language. However, some coders would like to see that vocabulary expand further, particularly in the area of comorbidities and complications.
Cindy Parman
When CMS switched from ICD-9 to ICD-10 last year, the new diagnosis classification list allowed coders to translate additional elements of doctors’ language to billers’ language. In a sense, ICD-10 added “adjectives” to coders’ vocabularies: oncology coders now have the ability to describe additional aspects of a tumor, such as whether it resides in the upper left or lower right quadrant of a patient’s lung.
Some coders would like to see that vocabulary expand further, particularly in the area of comorbidities and complications. “Oncology needs to get better at coding things that are not neoplasms—complications of care, or patients who show up with comorbidities,” said Cindy Parman, a principal at Coding Strategies Inc., a Georgia-based coding education and consulting company. “If a patient already has congestive heart failure, it may change the drug we're going to give the patient. I think oncology coders could use more codes like that—more codes for adverse events of care.”
In interviews with Oncology Business ManagementTM, oncology coders and practice administrators also said they see room for improvement in the coding of genetic mutations, staging information, and tumor behavior.
When Hoang Nguyen codes an inpatient oncology visit at the Boston-area hospital where he works, he pays special attention to patients’ comorbidities.
“The comorbidities will complicate treatment. If they've had kidney failure or congestive heart failure, that will make it harder,” said Nguyen, a coding instructor who does work for South Shore Hospital in South Weymouth, Massachusetts. Medicare bases its payment system for inpatient hospital stays around a concept called the Medicare Severity-Diagnosis Related Group (MS-DRG).
Hospitals bundle procedure codes under a DRG, and payers use it to assign a flat fee that they stratify based on complications and comorbidities. “So, diagnosis coding drives the payment for MS-DRGs,” Parman said.
Independent oncology practices, which specialize in outpatient care, don't use the MS-DRG system. Physicians are paid based on relative value units, which take into account the amount of work expended by a physician to treat a patient. However, a payment scheme more akin to the MS-DRG-based model may be in the works for outpatient care, Parman said.
Under the Patient Access and Medicare Protection Act of 2015, Congress directed CMS to submit a report on the development of an alternative payment model for radiation services in a nonfacility setting. Such a model would allow for greater specificity in billing, which would require greater specificity in coding.
Room for Improvement
“They're doing this already on an inpatient basis, and translating it to an outpatient basis wouldn't be that hard. But it would be hard for independent oncology practices that aren't fully classifying each one of their patients,” Parman said. “If everybody's sending in the same procedure code, how are we going to differentiate payment based on quality and value? Well, what we have are the ICD-10 diagnosis codes.”The inadequacies of ICD-10’s diagnosis codes once sparked a conversation among Tracie Whitley’s staff at the Regional Medical Oncology Center in Wilson, North Carolina. The biggest bugbear was the code set's lack of accommodation for “the genetic mutation pieces” of a cancer diagnosis, Whitley said. “It would be great if we could have those as an add-on digit to the cancer code,” she said. “A lot of the drugs now are specific to those genetic mutations, and that way, we can use it to explain why we coded for that drug.”
She also wishes a portion of the codes accounted for the progression of prior therapy. “We have a lot of drugs now where they want to see a patient had a previous line of therapy on a platinum drug, and they failed or progressed,” Whitley wsaid. “That's something we have to put in as a comment, whereas if we had a code, I think that would increase the efficiency of that process.” Amy Lawhead, a business analyst at the Oklahoma Cancer Specialists and Research Institute, says tacking additional elements onto codes and making them more precise would only benefit the oncology field. Coding expert Bobbi Buell said billers probably wouldn't like it—“If you're a biller, you're stressed out enough,” she said.
Several coders added that the people on their end would be more than happy to accommodate sharper attention to detail. “I’m one of those coders—and most coders are, I have to say—where I like codes to be as detailed as possible,” said Heather O’Leary, of the Santa Barbara Cancer Center at Sansum Clinic. “So, when these new codes come out and it's so specific, we like that.”
Payers and practice managers tend to welcome new codes as well, because the more specific the codes are, the more likely they will justify a prescription. This lessens “the prior authorization problem,” Buell said. New codes are released about twice a year. CMS occasionally augments regular updates in January and July with releases occasioned by a new drug or procedure, O’Leary said.
With new codes comes training. A new book with updated codes comes out every year, and practices with more than one coder often offer training when new codes come out. For the release of ICD-10, O’Leary took a refresher course in anatomy and physiology. She and Nguyen both mentioned how essential an understanding of anatomy is to coding well. "You really have to have a deep knowledge of anatomy and physiology. You have to know the body and how it works,” O’Leary said. “To be able to do codes to the highest specificity, you have to know exactly where that lymph system goes, so you can see, ‘Does it end here, or here?’”
Once coders acquire that understanding of anatomy and physiology, they come to realize how difficult their job of “translating” is, because “ICD- 10 does not reflect the true way physicians practice,” Lawhead said. “For instance, in lung cancer, there are numerous subcategories such as non— small cell, small cell, and squamous, which are treated differently. Yet, we have one set of ICD-10 codes—C34—to fit everyone into,” Lawhead said. “It would make sense for the codes to evolve to this type of detail.”
An enhanced level of detail would benefit oncology coding specifically, she said, because it is “one of those areas that is not always black-andwhite. In cardiology, it is easy to state, ‘The patient has a blockage in this artery, from this branch, on this side of the body,’ but in oncology, the disease may spread over multiple areas of the body, and it may not be easy to pinpoint the origin of the disease to give the patient one diagnosis,” Lawhead said.
In cases like this, it would be helpful to have an ICD-10 component for tumor behavior, she said. “It does not allow us to code for how the tumor is acting. For instance, the patient may have a gynecological tumor, but we cannot state it truly is ovarian, cervical, or uterine, but we can clearly state why and how this tumor is to be treated,” Lawhead said. “But there is no ICD-10 code that allows us to state ‘treating like ovarian cancer.’ This is an area lacking for oncology.”
Another area that is lacking is staging information, she said. “The codes have helped with areas such as laterality and some characteristics of diseases; however, more and more requirements are being put on the oncologists to submit staging information for reimbursement or to even get authorizations for services,” Lawhead said. “Therefore, if staging was added into the neoplasm category, this would tremendously help staff as well as payers to streamline the patient’s care and allow for faster, more efficient service.” Lawhead said she thinks a coding system with a broader number of codes and higher level of specificity would decrease the wiggle room in diagnosis interpretation, which would better streamline authorization and reimbursement.
Enhancing the codes’ focus on comorbidities and complications would benefit payers and the public health researchers for whom the International Classification of Diseases was originally designed as well, Parman said. Codes containing detailed complication and comorbidity information would give payers and public health researchers, who aggregate code data to determine health care trends, the ability to better assess how well certain treatments are working. “If somebody is going to pay X-thousand dollars for a patient, they want to know how many patients have adverse effects and how many just sail through. Is it 10%? Is it 50%?” Parman said. “That makes a difference.”