Bart’s Corner: Coding MedDRA and ICD-10
On October 1, 2015 it became official that health care providers must start using ICD-10 codes for billing to insurance companies and the government. Work on ICD-10 was started in 1983 and completed in 1994. So this is not a sudden or new development.
You can try searching for codes in ICD-10 using the browser from the Centers for Medicare & Medicaid Services. There are many private companies that offer coding software and services.
Although providers have been using diagnosis and procedure codes (ICD-9) for many years primarily for billing, the arrival of ICD-10 represents a major change. The number of available diagnosis codes has increased from 13,000 in ICD-9 to 68,000; the length of the codes has increased; the codes are much more specific and they now have “laterality” (left side vs right side).
In addition, and adding complexity, there are now many combination codes. That is, before one would use two codes for two medical problems, there are now combo codes (e.g. 415.0 acute cor pulmonale and 415.12 septic pulmonary embolism are now combined in ICD-10 to I26.01 septic pulmonary embolism with acute cor pulmonale). There are also procedure codes which increase from 3,000 in ICD-9 to 87,000 in ICD-10. There are similar changes in code length, detail, laterality etc.
There are many guides issued on this by various agencies and companies. A brief summary is available from the American Medical Association.
As you can easily imagine, the arrival of ICD-10 has met with multiple reactions ranging from bemusement (see below) to anger where they are “fixing something that is not broken.” Under “bemusement”:
- Burn due to water-skis on fire, initial encounter (V91.07XA)
- Spacecraft collision injuring occupant (V95.43XS)
- Struck By Turtle (W59.22XA)
- Hurt at the Library (Y92.241)
- Knitting and Crocheting (Y93.D1) This is “an external cause of morbidity.
For more funny codes, see this link.
On a more serious note, this change has significant costs and has been addressed in multiple articles. See the article in Bloomberg Businessweek.
Hopefully the transition will be smooth and the government agencies (e.g. Medicare and Medicaid) as well as private insurers are ready. The fears are that there are so many choices that coding will be harder and that reimbursement might be denied or changed (read: lowered) based on the codes. Providers will have to learn new “tricks” to maximize reimbursements. Hospitals and providers fear that there may be cash flow problems because of coding issues holding up payments.
Question: Why am I discussing this?
Answer: Because it is interesting and instructive to compare our drug safety/PV coding with MedDRA to those who have to code with ICD-10.
My major thesis here is that we are lucky to have MedDRA which, after nearly two decades of use, is stable and useable. Although the MSSO continues to make changes and updates, most of them are small and are refinements of codes or SMQs (Standardized MedDRA Queries). No major changes are anticipated.
Full Disclosure: I am biased. I think MedDRA has been a tremendous success. I was one of the original MedDRA testers in the 1990s and still do medical reviews of new and changed terms for the MSSO.
Comment: As most readers of this blog know, MedDRA is relatively easy to learn – presuming one has some knowledge of medicine and medical terms. The number of terms is not overwhelming and there are many fine training programs and browsers available. There are guides (Points to Consider: See the 2015 release 4.9 based on MedDRA 18.0). It has proven itself over the years both for data input and retrieval. The hierarchy works well and the choices usually logical. To date, combination terms (as in ICD-10) have been avoided.
So why is MedDRA a success and ICD-10 a potential problem?
MedDRA was developed relatively quickly with a clear and limited goal (use for pharmaceutical and, to a lesser degree, device safety coding and retrieval). ICD-10 development started over 30 years ago and only now is it being implemented. Over the years more and more was added on in an attempt to please all stakeholders (hospitals, health care professionals, government, coders, insurance companies, etc.) thus making it an unwieldy tool that gives all the stakeholders some functionality but is not optimized for anyone. This is the nature of committee designed projects.
So let us be happy with MedDRA. If it ain’t broken, don’t fix it.
Tags: drug safety, ICD-10, MedDRA, Pharmacovigilance