AHIMA Comments to PITAC's Recommendation Regarding ICD-10 and SNOMED-CT

April 21, 2004

Jonathan C. Javitt, MD, MPH
Co-chair PITAC Health Care Delivery and IT Subcommittee
President's Information Technology Advisory Committee
c/o National Coordination Office for Information Technology
Research and Development
4201 Wilson Boulevard, Suite II-405
Arlington, Virginia 22230

Dear Dr. Javitt:

On behalf of the Board of Directors and members of AHIMA, I am writing to express the Association's strong opposition to the PITAC subcommittee's April 13, 2004 draft recommendation to study the “cost-benefit” of upgrading the country's diagnosis and procedure code system from ICD-9-CM i to ICD-10-CM ii and ICD-10-PCS. iii We believe this action would be counterproductive and would like to suggest an alternative recommendation.

First, your recommendation wrongly implies that SNOMED-CT can be an alternative for ICD-10-CM and ICD-10-PCS (referred to collectively as ICD-10). This is not a case of choosing one or the other. Even when SNOMED-CT, a reference terminology, is universally implemented, we will still need ICD. It is the classification system for grouping concepts in the terminology for a number of critical uses such as health and vital statistics trending, health policy and planning, reimbursement, and many administrative uses. So, we need both SNOMED and ICD-10.

We agree that the goal is the use of SNOMED-CT by all health care organizations-provider and payer-but this will require a fully computerized health care industry, as it is not practical to use SNOMED any other way. We cannot delay ICD-10 implementation until electronic health records with SNOMED have been universally adopted. The ICD version in use today, ICD-9, is obsolete and must be replaced by ICD-10 immediately. This is not just AHIMA's opinion. Rather, it is the industry-wide consensus of nearly every stakeholder organization.

Your recommendation seems to ignore the extraordinary depth and rigor with which this issue has already been studied. The National Committee on Vital and Health Statistics (NCVHS) recognized the need to replace ICD-9 over a decade ago. Since then, the National Center for Health Statistics has developed the clinical modification to ICD-10 for use in the US, and the Department of Health and Human Services (DHHS) has developed ICD-10-PCS for coding procedures. And last fall, the NCVHS concluded its comprehensive evaluation of this issue by recommending that DHHS publish the Notice of Proposed Rule Making for implementation of ICD-10. NCVHS's decision was informed by a comprehensive and highly credible cost benefit study conducted by the Rand Corporation, and hundreds of hours of testimony by experts who examined all sides of the issue.

ICD-9 needs to be replaced by ICD-10 because it is a thirty-year old classification system that can no longer accurately describe today's practice of medicine. Testimony before the NCVHS from physician groups, CMS, public health and many others cited example after example of how ICD-9 compromises data quality and usefulness. Here's an example: The ICD-9 diagnosis code for “unspecified essential hypertension” is a commonly used code, though it imparts little specific information. However, the only other more specific choices in the ICD-9 hypertension category are “malignant” and “benign” essential hypertension, terms that are no longer even used in contemporary clinical practice. There are countless other similar examples that demonstrate how ICD-9 reflects old medical knowledge and practice. Yet, this is the data that our health care system uses for all of its critical decision-making!

The benefits of improved data from ICD-10 were quantified by Rand and shown to far outweigh the cost of implementing the updated classification system. Also, moving to ICD-10 will allow the US to compare its health care data to that of other countries, most of whom already use ICD-10. This is particularly important in an era of SARS and bio-terrorism surveillance.

Instead of recommending a cost benefit study on the alternative of using SNOMED-CT for reporting diagnoses and procedures, which we assert is not even a viable alternative, we believe that the Subcommittee should recommend:

  1. Expedited implementation of ICD-10
  2. Research into how to speed implementation of SNOMED-CT and demonstrate its use cases including the value of mapping to ICD-10

Since HIPAA requires that implementation cannot occur until two years after the issuance of a final rule, it will already be 2007 or 2008 before we begin to benefit from ICD-10. Your draft recommendation threatens to cause further delay and, in the meantime, the quality of our health care data continues to rapidly deteriorate. It is in our nation's best interest that your recommendations speed the adoption of ICD-10, not delay it.

AHIMA would appreciate the opportunity to share with you its expertise and resources regarding ICD-10, including the very positive results of an ICD-10 field-testing project conducted last year in conjunction with the American Hospital Association.

I look forward to a continued dialogue on this matter. Please feel free to contact me for further discussion or clarification. I can be reached by phone at (312) 233-1166 or by e-mail at linda.kloss@ahima.org. You can also contact Dan Rode our Washington, DC-based vice president of policy and government relations at (202) 659-9440 or by e-mail at dan.rode@ahima.org.


Linda L. Kloss, RHIA, CAE
Executive Vice President and Chief Executive Officer

cc: Members of the PITAC
Melanie Brodnik, PhD, RHIA, 2004 AHIMA President
Dan Rode, MBA, FHFMA, AHIMA Vice President

i    International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) is the current disease and inpatient facility procedure classification system used in the US. ICD-9 is a World Health Organization classification system copy written by WHO and no longer supported by the WHO. Clinical Modification represents “additions” to the ICD-9 modified in the case of the US, by the National Center for Health Statistics (NCHS) – Volumes 1&2 (diagnoses), and the Centers for Medicare and Medicaid Services (CMS) – Volume 3 (Inpatient procedures).

ii    International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) is the US modified version of the current international WHO system ICD-10. ICD-10-CM would upgrade or replace ICD-9-CM, Volumes 1&2 diagnosis coding system, in the US.

iii    International Classification of Diseases, Tenth Revision, Procedural Coding System (ICD-10-PCS) is the US replacement for ICD-9-CM, Volume 3.