Testimony To The National Committee On Vital And Health Statistics
~ Subcommittee on Standards and Security ~
Panel On The ICD-10 Next Steps – Testimony and Public Comment

Michael Lundberg
Vice-Chair, Board of Directors, National Association of Health Data Organizations
Executive Director, Virginia Health Information
Silver Spring, Maryland - October 29, 2003

My name is Michael Lundberg and I am the Vice-Chair of the Board of Directors of the National Association of Health Data Organizations (NAHDO) and the Executive Director of the Virginia Health Information (VHI).  I want to thank you, on behalf of both NAHDO and VHI, for the opportunity to offer testimony about the adoption and implementation of the ICD-10 as a coding system for the health care industry in the United States.

The National Association of Health Data Organizations is a non-profit membership and educational organization, established in 1987, to promote the public availability of health care data and improve statewide health care surveillance systems.   VHI is an independent non-profit organization collecting health care data on cost, quality and access for use by Virginia citizens. 

We believe the health care industry in the United States is at a very important crossroads in its adoption and use of national coding standards.  NAHDO feels that the ICD-9-CM has outlived its usefulness and support replacing it with ICD-10-CM for the identification and coding of diagnoses and conditions. NAHDO also recommends, initially a dual approach to procedure coding with ICD-10-PCS for hospital inpatient services and CPT for outpatient services, while an index and cross-reference system is developed for all outpatient services.  NAHDO recommends NCVHS immediately announce a two-year voluntary transition process from ICD-9-CM to ICD-10-CM and establish a defined date for the adoption of ICD-10-CM as a new code standard under HIPAA.

My testimony will focus on the following three major areas:

• Overall Perspectives on ICD-9 and ICD-10
• Examples of Benefits that can be achieved with the transition to ICD-10
• NAHDO Recommendations to NCVHS

1.      Overall Perspectives on the ICD-9 and the New ICD-10 Coding and Classification System

We believe the ICD9 is no longer sufficient for today's purposes and is not consistent with the direction that the rest of the world is going.  The current HIPAA regulations require the use of ICD-9-CM Volumes 1 and 2 for the reporting of inpatient conditions and the ICD-9-CM Volume 3 for the reporting of hospital inpatient procedures.   For all outpatient services, including hospital outpatient procedures, the regulations require the use of HCPCS codes. 

In the 90's, our members have noticed the following flaws with the ICD-9-CM classification system:

• The ICD-9-CM classification system is outdated and inconsistent with current health care practice. The ICD-9-CM process is reaching a point where it cannot be adequately updated and cannot keep pace with the changes in healthcare.
• The lack of space will not accommodate newly identified diseases or the healthcare encounters for reasons other than treatment of disease or injury, such as preventive medicine.
• The codes lack sufficient clinical detail to describe the severity or complexity of diagnoses and as a result, lead to inconsistent code assignment.
• The most striking limitation is the inability in most cases to distinguish between events that happened in the hospital and those that occurred prior to admission[1]. 
• Frequently, manual review of health records are required to meet the information needs of researchers and to fulfill other data mining functions.

The ICD-9-CM classification system is currently being used for the following purposes:

• Measuring the quality, safety, and efficiency of health care services.
• Conducting research, epidemiological studies, and clinical trials.
• Improving clinical, financial, and administrative outcomes performances.

In order to perform the listed purposes, the 24-year-old ICD-9-CM system needs to:

• Expand distinctions for ambulatory and managed care encounters
• Expand to include emerging diseases and more recent medical knowledge
• Develop consistent and current use of standard terminology
• Capture procedures being done with new technology.

The two new classification systems (ICD-10-CM for diagnoses and ICD-10-PCS for procedures) will capture greater specificity and clinical detail and new technological procedures. The significant improvements will result in:

• Improving the ability of providers, payers, government agencies, and others to measures the quality, safety, and efficiency of health care services
• Reduce the need for manual review of health records to perform research and data mining
• Enhance public health decision making
• Improve the ability to forecast the healthcare needs

While data will be more robust under I-10-CM, we must be mindful of the costs:

Data collection costs

• According to AHIMA, the average learning curve for other countries was 4-6 months and they reported that they did not find ICD-10 any more or less difficult to learn than ICD-9. [Since the coding rules are basically the same in applying the codes for either system, it should not require extensive retraining for coding professionals]. The changes to the data systems, groupers, and software will incur a one-time cost similar to the HIPAA changes for other standards.

IT/vendor costs

• The move to the ICD-10 system will increase the workload for the information system personnel, similar to the HIPAA and Y2K standardization. The software needs to accommodate the 7 digit diagnosis codes and 7 digit procedure codes, alphanumeric data type, edit logic changes, and expansion of files containing ICD-10 codes. The changes will impact the data retrieval and analysis between the old ICD-9 and the new ICD-10, which will necessitate the use of the crosswalks developed by CMS and NCHS.
• Vendors and IT staff will also need to modify existing audit and edit systems both prospectively and retrospectively (for the analysis of past records).

Analytic costs

• The specificity of ICD-10 codes will require a much more robust review of small cell sizes to further prevent patient identification.
• It will also be important to consider the need to develop methods for bridging trend data pre- and post- adoption of ICD-10

People costs

• This is one of the most significant costs --- the need to prepare, train and educate coding personnel across the health care industry.
• The increased specificity in the ICD-10 classification systems will make documentation more important.  Health information managers and coding professionals will need to help radiologists, pathologists, and other physicians and clinicians become more aware of documentation requirements for accuracy and completeness about the patient’s healthcare.

With these challenges in mind, the possible uses of ICD-10 provide greater return in improving healthcare needs for all people. The aging ICD-9-CM system will hinder the health care progress as it does not provide data specificity when associated with other current data information such as rising healthcare costs, healthcare performances, and other outcomes factors.

2.      Examples of Benefits That Can be Achieved With the Transition to ICD-10

• Greater specificity of asthma diagnosis coding would help researchers and public health determine if a hospitalization was avoidable or not[2].    ICD-10-CM diagnoses codes would provide more clinical information about ACSCs.  As an example, of ACSC, the ICD-9-CM describes asthma (493) as extrinsic, intrinsic, chronic obstructive, and unspecified.  ICD-10-CM describes asthma (J45) as mild intermittent, mild persistent, moderate persistent asthma, severe persistent, and other & unspecified----terminology reflecting the recent medical standard terminology.
• For outcomes studies on acute myocardial infarctions, there are codes that can lend more specific information about the cardiac arrest.  ICD-9-CM describes cardiac arrest (427.5) with no further information on the cause.  ICD-10-CM describes cardiac arrest (I46) as due to underlying cardiac condition, due to other underlying condition, or cause unspecified.
• For trending conditions in cerebrovascular disease, there are codes that can provide more information about the specific cerebral artery causing the subarachnoid hemorrhage.  ICD-9-CM describes subarachnoid hemorrhage (or cerebral aneurysm) (430) with no further information as to which cerebral artery is affected.  ICD-10-CM renames the condition as nontraumatic subarachnoid hemorrhage (I69) and describes it further as to which of the nine specific cerebral arteries the hemorrhage is coming from.  This will provide more clinical details for identifying, tracking, and managing disease processes.
• For popular publications on pregnancy-related conditions, there are codes that can provide specific information about which trimester the treatment was being sought.  ICD-9-CM describes the pregnancy-related conditions such as hypertensive heart disease, infections of urinary system, placental separations, diabetes, drug and/or alcohol use, skin disorders, antepartum hemorrhage, and many more. ICD-10-CM further describes the same pregnancy-related conditions as to which trimester (unspecified, first, second and third) the treatment for such condition was being sought.
• For studies on procedures, we know that the ICD-9-CM (volume 3) lacks specificity and has an insufficient numeric structure to capture new technology.  In 1994, the DHHS Office of Inspector General issued a report describing vulnerabilities in maintenance, use and management of CPT. The report identified several flaws in CPT-4 codes, guidelines, and index that can lead to improper coding. The ICD-10-PCS will provide a marriage of dual procedure systems with a structure to capture new technology, greater specificity and detail, and new medical services and technologies.  A number of approaches and techniques used for procedures such as laser, scopes, and minimally invasive techniques can be readily captured by the ICD-10-PCS.  This will provide insight as to the improvement of the patient’s health outcomes with these new procedures.
• For studies on injuries and causes, there are codes that can lend more specific information regarding the patient’s encounter level.  ICD-9-CM describes fractures as closed or open or with complications such as nonunion. ICD-10-CM describes fractures further, such as initial encounter for closed fracture, initial encounter for open fracture, subsequent encounter for fracture with routine healing, subsequent encounter for fracture with delayed healing, and subsequent encounter for fracture with nonunion.  ICD-9-CM describes the cause, such as fall from playground equipment with no further information about the encounter.  ICD-10-CM describes the causes more specifically, such as fall from playground slide, fall from playground swing, fall from jungle gym, fall from other playground equipment, along with a 7^th character to denote initial encounter, subsequent encounter, or sequelae.
• ICD-10-CM added more conditions that are not available in the existing ICD-9-CM system.  For example, ICD-10-CM provides specific codes for blood types (Type O, Rh negative, etc) and for immunizations not carried out for specific reasons (patient’s decision, belief or group pressure, etc).
• ICD-10-CM provides greater specificity for services related to examinations.  ICD-9-CM describes medical examination for administrative purposes, with no further information about the purpose.  ICD-10-CM describes the encounter for examination with more specific reason codes, such as for admission to educational institution, or for pre-employment examination, or for admission to residential institution, or for recruitment to armed forces, or for driving license, or for participation in sports, or for insurance purposes.
• ICD-10-CM provides more possibilities for further study of the health care delivery system to better meet healthcare needs without having to manually retrieve the medical record documents.  For example, the decubitus ulcer (or bed sore) is more common for the elderly patients in long term health care facilities.  ICD-9-CM describes the condition as decubitus ulcer (707.0).  ICD-10-CM describes the decubitus ulcer (L89) with specific locations (right upper back, left upper back, right lower back, buttock, and so on) and the breakdown of the skin (fat layer exposed, necrosis of muscle, necrosis of bone, or unspecified severity).  This level of specificity can lead to enhanced health care measures preventing the severity of bedsores that can lead to loss of limb or death.
• The ICD-10 classification system will be more consistent and uniform with HIPAA standards.  There is already in place a federal committee (ICD-9-CM Coordination and Maintenance Committee) who will ensure continuity and efficient updating of this medical code set standard over time.  From what we’ve seen in the preview of the ICD-10 classification systems, it has greater flexibility to adapt easily to changes in the healthcare infrastructure.  Most notably, the development of these two classification systems was accomplished with the expert advice from physician groups in neurology, orthopedics, pediatrics, obstetrics and gynecology, psychiatry, dermatology, and surgeons of various specialties; professional organizations (ADA, AMA, AHIMA, ANA, NACHRI); and other ICD-9-CM users (federal agencies, workers compensation, epidemiologists, and researchers.  Since other countries (2000) and Vital Statistics in U.S. (1999) have been using the ICD-10 diagnoses, the use of ICD-10-CM by the health care industry within U.S. would provide comparability with international data and State/National mortality data.

3.      Conclusions and Recommendations

• NAHDO feels that the ICD-9-CM has outlived its usefulness and support replacing it with ICD-10-CM for the identification and coding of diagnoses and conditions. Furthermore, NAHDO recommends NCVHS immediately announce a two-year voluntary transition process from ICD-9-CM to ICD-10-CM and establish a defined date for the adoption of ICD-10-CM as a new code standard under HIPAA. At the end of this two-year transition period, the new regulations would be adopted with ICD-10-CM as the new standard code set.  Because of the HIPAA rule-making process, this gives the industry an additional 180 days of compliance after the final rules are adopted.
• NAHDO believes that issues with the adoption of the new coding standard can be worked out during this two-year transition period.   Costs would be expected to be distributed throughout the transition period.
• With regards to a single procedure classification system for all services, we know that the ICD-9-CM (volume 3) lacks specificity and has an insufficient numeric structure to capture new technology and that the CPT-4 has a number of vulnerabilities in its maintenance, use and management [(as documented in 1994 by the DHHS Office of Inspector General].  The ICD-10-PCS provides a marriage of dual procedure systems with a structure to capture new technology, greater specificity and detail, and new medical services and technologies.  Nevertheless, the ICD-10-PCS would financially impact the outpatient health industry.  It would essentially impose a completely new medical terminology assigning new naming conventions to current procedures.  There would need to be a well-developed index and cross-reference system.    Therefore, we would support a dual approach to procedure coding with ICD-10-PCS for hospital inpatient services and CPT for all outpatient services.  Perhaps the ICD-10-PCS can be phased in over time for all services and thus satisfy the HIPAA requirement to eliminate duplicative efforts of dual procedural classification systems.
• It is very important that the changes in the ICD-10-CM be coordinated with HIPAA data standards.  If the new coding system increases the average number of codes per patient, then the requirements for number of codes included in administrative data sets (such as the health care claim) needs to be expanded.
• It is our understanding that the ANSI X12N standards are being updated to support coding ICD-10-CM and ICD-10-PCS. It seems that the way the maintenance was done will support allowing ICD-9-CM coding along with ICD-10-CM and ICD-10-PCS. That support should be specifically identified and included in the next iteration of HIPAA regulations.
• We also understand that the NUBC is in the process of approving the next version of the paper billing form, the UB-02 and believe that the new form should also support ICD-10-CM and ICD-10-PCS coding.

In summary, we believe the efforts undertaken by this Committee to look into the need to transition to a new coding system and to find the best approach for achieving this goal are commendable. The opportunity to continue to offer comments along the way, and in particular as final recommendations get drafted will be critical to the success and acceptability of any transition plan.   NAHDO will stand ready to continue to work with this Committee and other industry groups to ensure that...

I thank you again for the opportunity to provide this testimony.

[1] The current external cause of injury and poisoning code for hospital occurrences, E849.7, is not a hospital specific code, but rather is a code for "Residential institution". In addition to hospital this code includes seven other settings, including jail and prison. Further, even if E849.7 is assumed to mean a hospital event, there is no foolproof way to associate this place of occurrence code with the event that occurred in the hospital.

[2] Asthma is an ambulatory care sensitive condition, a condition in which timely and effective outpatient care can help to reduce the risks of hospitalizations by either preventing or controlling the onset of an illness or condition.