The Effect of ICD-10-CM Coding on Nonmalignant Hematology

Clinical Advances in Hematology & Oncology
November 2015, Volume 13, Issue 11

Andrew D. Boyd, MD
Assistant Professor
Biomedical and Health Information Sciences
College of Applied Health Sciences
University of Illinois at Chicago

H&O Why was the switch to the International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) coding made in October 2015 after more than 35 years with ICD-9-CM?

AB Medicine has changed since the World Health Organization (WHO) developed version 9 of the ICD. The WHO classification is the starting point for the clinical modification that we use as the basis for reimbursement in the United States. Most countries transitioned to ICD-10-CM in the 1990s, but our government has postponed the transition several times—although it decided against jumping straight from ICD-9-CM to ICD-11-CM when it comes out in a couple of years.

The goal of the new version, with its increased number of codes, is to provide finer-grain detail about clinical disease to health insurers, hospitals, researchers, and physicians. The concept is that increased fidelity in describing disease will permit better management of our patients and deeper understanding of our health care system. For example, detailed codes will let us know how many physicians and nurses are needed, what types of medications to stock, or what type of infection is most likely after a particular procedure. In addition, payers will better be able to detect fraud.

H&O What are the most obvious differences between ICD-9-CM and ICD-10-CM?

AB The number of codes has increased from approximately 14,000 to 80,000. The codes are now as long as 7 digits rather than 5 digits, and all of them rather than just some of them are now preceded by a letter. The hierarchy is also different, and this new structure will be carried into the next version of codes.

H&O What are some of the most important differences between ICD-9-CM and ICD-10-CM in the field of nonmalignant hematology?

AB The good news is that compared with other specialties, nonmalignant hematology will have one of the least complex transitions. In a 2013 study in the Journal of the American Medical Informatics Association, my colleagues and I found that the proportion of convoluted diagnosis mappings—in which codes do not map backward and forward between ICD-9-CM and ICD-10-CM—was 5% in hematologic diseases of the blood or blood organisms. This is far from perfect, but much better than the 60% rates of convoluted diagnosis mappings in obstetrics and in injuries.

There are, however, some tricky areas. One example is D57.00 in ICD-10-CM, “Hb-SS disease with crisis, unspecified.” That maps backward to 282.62 in ICD-9-CM, which refers to “Hb-SS disease with crisis.” What hematologists need to be aware of when they collect data on sickle cell disease is that 2 additional ICD-10-CM codes map backward to 282.62: “Hb-SS disease with splenic sequestration” (D57.02) and “Hb-SS disease with acute chest syndrome” (D57.01). The level of detail is much higher, which can make it difficult to do comparisons across years.

Another change is the code for “sickle cell disease without crisis” in ICD-10-CM (D57.1), which maps backward to “sickle cell disease, unspecified” in ICD-9-CM (282.60).

As we found in a 2014 article in Journal of Oncology Practice, with Venepalli as the first author, “secondary hypercoagulable state” (289.82) has mapped forward to “other thrombophilia” (D68.69). In other words, we have lost the concept of secondary thrombophilia in ICD-10-CM.

We also found that 287.30, which represents “primary thrombocytopenia, unspecified” originally was mapped forward to D47.3, which represents essential thrombocythemia. This has been corrected so that now its maps forward to D69.49, which represents “other primary thrombocytopenia.” I mention this because some electronic health record vendors or insurance companies may not have made the correction.

H&O What other issues have been coming up?

AB When patients with hematologic disorders have comorbidities, that gets into a highly complex area. For example, we found that 40% of the codes in infectious and parasitic disease—the 001 through 139 codes—are complex. The same goes for codes related to mental disorders. Disorders of the circulatory system are 20% complex. So even though the primary reason for the visit will generally be simple to code, the comorbidities are much more complex.

H&O Do you think that ICD-10-CM will enhance patient care?

AB In the long term, ICD-10-CM will enhance patient care. However, the transition is challenging. In addition to making sure the computer systems work properly, we are still learning which diagnosis codes in ICD-10-CM get reimbursed and which do not. Physicians obviously cannot use incorrect codes in order to obtain reimbursement, but in the case of overlapping codes, of course they will want to use the one that produces reimbursement.

H&O Is there a different effect of ICD-10-CM on community hematologists vs academic hematologists?

AB In general, the large academic centers and hospital systems are most prepared for ICD-10-CM. Community-based hematologists with just a few staff members have had more difficulty designating certain people to learn about the ICD-10-CM coding systems. In September, some reports were saying that up to 75% of the small- to medium-sized practices were not prepared, so hopefully they have been studying up in their evenings or afternoons and making use of resources from the American Society of Hematology and other groups.

H&O What are some of the disadvantages of transitioning to ICD-10-CM?

AB The biggest downside is just learning the new system, and having every payer and provider in the country go through the transition at the same time. This requires a huge cost investment and is disrupting work flow and reimbursement. The original cost analyses by the Rand Corporation, which were done in 2004, estimated the cost of conversion as $425 million to $1.1 billion and the loss of productivity as worth $5 million to $40 million. The cost of health care has gone up since then, so we will not know for several years the true cost of implementing ICD-10-CM.

All hematology practices have differences in the way they code, based on differences in disease prevalence, different ways of coding, and differences in the insurance companies they deal with. As a result, the coding advice for one practice does not always carry over to another practice.

One disadvantage that rarely gets discussed is that we lose data in 3% to 6% of the cases in which we move from ICD-9-CM to ICD-10-CM.

H&O How is ICD-10-CM affecting the levels of billing and collections?

AB The Centers for Medicare and Medicaid Services has announced that it will be very lenient for the first year about the specificity of the codes. As long as the code is in the right class, it will be reimbursed. This is helpful because it will prevent refused claims and interruptions in cash flow. On the other hand, it will cause physicians, health care groups, and hospitals to need to learn the system twice: first the general system, and then the precise system.

Although the government is temporarily relaxing the rules about specificity, billing and collections remain big unknowns. I anticipate some disruption to cash flow based on denied claims. My personal wish was that the insurance companies and the government would guarantee cash flow neutrality for at least the first month, but nobody was willing to commit to that.

Based on my conversations at conferences with people working at insurance companies, at least some of these companies have carefully mapped the codes in all of their contracts from ICD-9-CM to ICD-10-CM. Other companies have taken shortcuts, and that is where we are likely to see problems and denied claims. The adjudication process to sort this out will delay payment, although we have yet to see by how long.

H&O Will the benefits of ICD-10-CM be worth it in the end?

AB I think we will all be happy with ICD-10-CM in 5 or 10 years, but the transition is challenging.

One problem is that we will have difficulty determining, say, the average number of flu cases per year, which requires looking at data over the previous 3 to 5 years. Data from the 2015 to 2016 flu season are likely to be off, which will affect our ability to determine the size of the necessary medical response, such as how many extra nurses to hire, how much medication to stock, or how many face masks to buy.

H&O What steps can physicians take to make this transition easier on their office staff?

AB All physicians should sit down with their office staff if they have not yet done so to figure out their top 10 or 25 diagnosis codes in ICD-9-CM and determine their equivalent codes—or what additional detail they need—in ICD-10-CM. The physician should also meet regularly with the office staff, perhaps every morning or every evening, and ask, “Hey, did I give you enough detail in my note?” or “Did I choose the right code?”

H&O How should hematologists handle problems with the system?

AB As we go forward with ICD-10-CM, we may see denied claims for specific codes that were appropriate. Remember, this is not just about reimbursement. The codes represent medical knowledge, so it is important to note whether a patient has, for example, primary or secondary thrombocytopenia.

We are literally learning a new language, and there are unanticipated consequences that we really cannot describe now. So if a specific code continues to be denied although it accurately reflects the patient’s condition, call the insurer and ask to speak to some of the more knowledgeable people reviewing claims. Hematologists are the best people to advocate for their specialty. If you run into recurring problems, your best bet is to contact your professional society.

Suggested Readings

Boyd AD, Li J, Burton MD, et al. The discriminatory cost of ICD-10-CM transition between clinical specialties: metrics, case study and mitigating tools. J Am Med Inform Assoc. 2013;20(4):708-717.

Boyd AD, Li JJ, Kenost C, et al. Metrics and tools for consistent cohort discovery and financial analyses post-transition to ICD-10-CM. J Am Med Inform Assoc. 2015;22(3):730-737.

Boyd AD, Yang YM, Li J, et al. Transition to ICD-10-CM threatens patient safety by degrading adverse event monitoring [published online September 3, 2014]. J Am Med Inform Assoc. doi:

Caskey R, Zaman J, Na H, et al. The transition to ICD-10-CM: potential challenges for pediatric practice. Pediatrics. 2014;134:31-36.

Krive J, Patel M, Gehm L, et al. The complexity and challenges of the International Classification of Diseases, Ninth Revision, Clinical Modification to International Classification of Diseases, 10th Revision, Clinical Modification transition in EDs. Am J Emerg Med. 2015;33(5):713-718.

Libicki M, Brahmakulam I; Department of Health and Human Services. The costs and benefits of moving to the ICD-10 code sets. Published March 2004. Accessed September 23, 2015.

Venepalli NK, Qamruzzaman Y, Li JJ, Lussier YA, Boyd AD. Identifying clinically disruptive International Classification of Diseases 10th Revision Clinical Modification conversions to mitigate financial costs using an online tool. J Oncol Pract. 2014;10(2):97-103.