AHIMA Testimony on Health Information Technology, Quality Data and the Need for ICD-10-CM and ICD-10-PCS before the Subcommittee on Health of the House Ways and Means Committee

Testimony of Linda Kloss, MA, RHIA, CAE, Chief Executive Officer, AHIMA
Before the U.S. House of Representatives Committee on Ways and Means, Subcommittee on Health 

July 27, 2005

Chairman Johnson, Mr. Stark, and members of the committee, thank you for this opportunity to address the quality of health data and actions that are needed to improve it as part of the overall US health IT initiative.

The American Health Information Management Association and its 50,000 health information management professional members are deeply committed to and actively participating in the adoption of standards-based and interoperable health IT. We are on the front lines in implementing electronic health records and other technologies as well as the implementation of local and national health information exchange and continue to be on the forefront of professional activities including privacy, confidentiality, security, data integrity, consumer and professional education.

My comments this morning relate to the urgent need for the Department of Health and Human Services to immediately initiate the regulatory process for adoption and implementation of ICD-10-CM and ICD-10-PCS code sets (referred to as ICD-10), rules, and guidelines as a replacement for the 30-year old ICD-9-CM. ICD-9-CM is not meeting current healthcare data needs and cannot support the transition to an interoperable health data exchange in the USⁱ. HHS must issue a final rule for adoption of ICD-10 as soon as possible to reverse the trend of deteriorating health data and to allow the healthcare industry to prepare for a smooth transition to modern classification systems by 2008.

Specifically we are calling for the following action by HHS and the healthcare industry, and urge your support for these actions:

• HHS must immediately initiate the regulatory processes to permit final implementation and use of upgrades to the deficient ICD-9-CM classification system by October 1, 2008. This upgrade will affect all diagnoses coding currently Volumes 1 and 2 of ICD-9-CM, as well as inpatient procedural coding, currently Volume 3 of ICD-9-CM.
•   Adoption of final rules as early a possible in 2006 will give the healthcare industry ample notice to commence systems conversion and other steps necessary to ensure a smooth and efficient implementation.
•   A coordinated, collaborative implementation strategy should be developed by industry stakeholder representatives to ensure broad input and a consensus-driven transition process.
• System conversions and upgrades to implement ICD-10-CM and ICD-10-PCS should be accomplished by healthcare entities in conjunction with the UB-04 and CMS 1500 (diagnosis codes only) system changes.
• Robust, rules-based, maps among SNOMED-CT®, ICD-10-CM and ICD-10-PCS, and ICD-9-CM should be developed promptly and distributed via the Unified Medical Language System (UMLS). ⁱⁱ

[I have placed a simple “understanding ICD-10” at the end of this testimony and this includes how ICD-10 would impact electronic health records.]

ICD-9-CM should have been replaced nearly 10 years ago. Each year that passes results in further deterioration of the classification system and the data that it produces:

• The ICD-9-CM coding structure and capabilities are in crisis. There are very few unassigned codes remaining to accommodate new diagnoses and procedures.
• In addition to no further capacity for expansion, many of the codes now in use do not accurately describe the diagnosis or procedure concepts they are assigned to represent.
• While the US has used ICD-10 coding to report mortality data since 1999, we are now virtually the only industrial nation that has not upgraded its morbidity classification system. This failure threatens our ability to track and respond to international threats to public health and bioterrorism. Rather than being a world leader in the collection of high-quality health data, the US lags far behind.

At a time when Congress and the Administration are making significant progress toward improving our health information infrastructure, the critically needed upgrade of ICD-9-CM has been delayed with little acknowledgement of the serious consequences and no clear plan for fixing the problem. Further delays in adoption of ICD-10-CM and ICD-10-PCS increase the cost of an eventual implementation once ICD-9-CM completely breaks down. While the US is working hard to adopt health information technology, it must also accommodate a robust 21 st -century classification system.

According to the 2003 Rand study commissioned by the CDC, the benefits of implementing ICD-10-CM and PCS outweigh the costs within a few years of implementation. Rand further noted that the cost of doing nothing may be greater than actual implementation and further delay in adoption is likely to increase future implementation costs. This research did not examine the upgrade to ICD-10 as a component of the overall health IT improvements and thus it did not factor in the potential to change the paradigm of coding through accelerating the development of computer assisted coding tools. Thus, the potential benefits may be accelerated.

Adoption of national electronic health records (EHRs) and interoperable information networks require improved classification systems for summarizing and reporting data. Government and industry leaders cite healthcare initiatives that rely on data but are in fact compromised by the continued use of ICD-9-CM. These include quality measurement, pay-for-performance, medical error reduction, public health reporting, actuarial premium setting, cost analysis, and service reimbursement. ^{iii iv} Classifications systems are key elements of the health information improvement strategy. Failure to upgrade ICD-9-CM diminishes the value of the US investment in SNOMED-CT®. The anticipated benefits of an EHR cannot be achieved if SNOMED-CT must be aggregated into an antiquated classification system like ICD-9-CM. Conversion to ICD-10-CM and ICD-10-PCS will not only produce better information and support development of computer-assisted coding, they will serve as the necessary foundation for continued improvements and expansion of 21 st -century classification systems, nationally and internationally

Healthcare providers, payers, and vendors are waiting for a notice from HHS signifying the intent to implement ICD-10 in order to begin planning and preparing for an anticipated use date. Vendors also need this notice to ensure new products will be available to accommodate these more advanced classification systems. US healthcare entities will soon be converting databases and applications systems to accommodate the upgrades to UB-92 and the CMS 1500 claims forms and data sets. It would be effective and efficient to make ICD-9-CM upgrades at the same time. Without some indication that implementation is on the horizon, healthcare providers, payers, and vendors will be reluctant to make these necessary changes concurrently.

As I have noted, the ICD-9-CM coding standard is in serious crisis. Terminologies and classifications from the 1970's no longer fit with the 21 st century healthcare system as numerous conditions and procedures are outdated and inconsistent with current medical knowledge and application. New advances in medicine and medical technology and the growing need for quality data cannot be accommodated. Data incomparability continues to increase globally and within the US due to the use of these antiquated code sets. As of the spring 2005 ICD-9-CM coordination and maintenance cycle, the US now has less than 70 remaining codes to represent health technology in the future. Two simple examples of the gross inadequacy of this classification system:

• ICD-9-CM offers two codes for asthma, extrinsic and intrinsic. Current medical knowledge no longer considers this a clinically relevant distinction. In ICD-10-CM asthma codes are differentiated by mild persistent, moderate persistent; and severe persistent, which are the terms used in evidence-based practice guidelines.
• In the area of procedures, ICD-9-CM simply lacks important specificity. There is a single non-specific code for “other revision of vascular procedure” encompassing a wide variety of surgeries on blood vessels. ICD-10-PCS in contrast will allow capture of the type of surgery, the specific artery or vein involved, and use of a device such as a graft or prosthesis. This kind of detail is essential for evaluating outcomes and efficacy and may decrease the supplemental information that is required to adjudicate a claim, in the form of a paper attachment or actual review of the medical record.

Data coded under the ICD-9-CM system are the foundation for billing, claims processing, payment and pricing. It is used for public health and quality reporting, biosurveillance, research, pay for performance, provider credentialing and fraud detection. In other words, it underlies all the major programs that this Committee oversees and is looking to advance. However, ICD-9-CM does not meet any of the following criteria:

• Code set standards outlined by the Health Insurance Portability and Accountability Act of 1996 (HIPAA);
• New services and technology that must be acknowledged in CMS payment systems according to the Benefits Improvement and Protection Act of 2000 (BIPA); or
• Characteristics of a procedural coding system outlined by the NCVHS in 1993.

Significant costs are incurred by continued use of severely outdated and limited coding systems. For example, failure of our coding systems to keep pace with medical advances results in the use of vague or incorrect codes often taken from the claims form and thereby requiring excessive reliance on supporting paper documentation (attachments or copies of the health record).

When the need to replace ICD-9-CM was identified in 1993, steps were taken by the National Committee on Vital and Health Statistics (NCVHS), the National Center for Health Statistics–CDC (NCHS) and the Centers for Medicare and Medicaid Services (CMS – then HCFA) to develop a migration plan to ICD-10 for morbidity and mortality coding. ICD-10 use for mortality coding in the US was initiated in 1999, however, while the rest of the industrial countries are now using their variations of ICD-10 for all reporting, the US continues with the unsupported ICD-9-CM (the World Health Organization (WHO) now exclusively supports ICD-10) leading to data incomparability with the rest of its global community.

Extensive work and dedication has gone into developing and evaluating these systems as replacements for ICD-9-CM. While there is significant support for this ICD-9-CM upgrade, there is also a segment of the healthcare industry, clinging to antiquated legacy systems, who continues to argue for further delay choosing to forgo the benefits of improved data and information available through 21 st -century terminology and impeding progress toward achieving critical US healthcare goals.

In November 2003, the NCVHS recommended that HHS initiate the regulatory process for the adoption of ICD-10-CM and ICD-10-PCS as replacements for the 30-year old ICD-9-CM. At that same time, Congress – in language included in the Medicare Prescription Drug, Improvement and Modernization Act (MMA) – urged the HHS Secretary to move forward with promulgation of rules for adopting and implementing ICD-10-CM and ICD-10-PCS. It is now 2005 – a year and a half after these distinguished recommendations – and HHS has taken no action.

We believe that adoption of ICD-10-CM and ICD-10-PCS is a component of the health IT strategy being advanced by Congress and the Administration, not a diversion from it. You are aware that the federal government has licensed SNOMED-CT® to make it available at no charge as a reference terminology in electronic health records. Mappings must be built from SNOMED-CT to ICD-10 so robust computer assisted coding applications will be available for adoption. ^v Today, the National Library of Medicine is preparing mappings to ICD-9-CM because there is not yet a final rule for ICD-10. I liken this to putting a model T engine in a Porsche.

Electronic health records based on a reference terminology, such as SNOMED-CT, offer a new paradigm for health data capture, aggregation and reporting. In the future, it will be possible to use a variety of classification systems to meet specific information needs without laborious manual coding. This is what AHIMA and others are working toward. But it will take years for these technologies to be built and fully deployed in all provider and payer organizations throughout the US . A date certain for implementation of ICD-10 will drive application development and accelerate adoption of reference terminologies, mappings and artificial intelligence-aided coding engines. It will permit IT vendors, providers and payers to prepare for the change and develop software “crosswalks” between ICD-10 and ICD-9-CM to accommodate organizations that cannot overcome legacy system limitations before the effective date.

In addition to adopting ICD-10, the US must redouble its efforts to ensure uniform coding practice and adherence to coding guidelines. The 1996 HIPAA legislation called for uniformity of transactions and code sets, but states and payers, including CMS, persist in adopting local rules and guidelines that further undermine data integrity and add to administrative costs. If we are to have reliable interoperable data and cost effective fraud-deterring systems, we must promote uniformity no matter where the data originates and no matter who the payer is. This is a first, but important step in improving the quality of health data through standards.

I commend your leadership in supporting health IT improvements and urge you to codify the important work of the Office of the National Coordinator in the Department of Health and Human Services. I also commend your action to safeguard the privacy and security of personal health information. Health information management is fundamentally a field that safeguards patient data—its integrity, its effective use, and its privacy and security.

In closing, we urge Congress to expedite adoption of ICD-10-CM and ICD-10-PCS and standards that will improve the accuracy and consistency of health care data. I stand ready to answer questions and to provide any additional information that is not covered in my written testimony. Thank you.

Understanding ICD-10

ICD-10-CM is a US clinical modification of the World Health Organization's (WHO's) International Classification of Diseases, 10th edition (ICD-10) is maintained by the National Center for Health Statistics (NCHS) . ICD-10 is now implemented or being implemented in all highly developed nations except the US . ICD-10-PCS was designed under contract by the Center for Medicare and Medicaid Services (CMS) , specifically to replace the ICD-9-CM procedural coding system.

The US is the only developed country that has not adopted ICD-10 for mortality and morbidity. A total of 99 countries are currently using ICD-10 for both mortality and morbidity. The US has used ICD-10 since 1999 for mortality reporting only. We need to implement ICD-10-CM in order to maintain comparability between mortality and morbidity data.

Improved Data

ICD-10 provides better data needed to meet the demands of an increasingly global and electronic healthcare environment. The ways in which coded data are being used today go well beyond the purposes for which ICD-9-CM was designed for back in the 1970s. Significant advances in the understanding of disease and treatment have been made over the last 30 years.

ICD-10 provides a significant opportunity to improve the capture of information about the increasingly complex delivery of healthcare. ICD-10 will provide:

• Better data to support quality and patient safety improvement activities
• Better data for improved public health and bio-terrorism monitoring
• Better data for more accurate reimbursement rates.

ICD-10 and the EHR

ICD-10-CM and ICD-10-PCS are better suited for use in electronic health record systems (EHR) than ICD-9-CM. The expanded availability of SNOMED-CT® made possible by recent government licensing agreement increases the urgency of replacing ICD-9-CM with ICD-10-CM/PCS so the development of mapping tools to the ICD-10-CM and ICD-10PCS can be initiated. Valid maps are urgently needed to link from a highly specific terminology to a classification system so that information captured in the reference terminology can utilize the power of summary required for healthcare reporting and indexing offered by the classification systems. ICD-10 medical coding system facilitates more robust mapping from SNOMED-CT clinical reference terminology in the EHR due to its greater size and granularity.

Continued use of the outdated version of ICD (ICD-9-CM) diminishes the value of the US investment in SNOMED-CT®. The anticipated benefits of an EHR cannot be achieved if the reference terminology employed in the EHR, such as SNOMED-CT®, is aggregated into a 30-year old classification system such as ICD-9-CM for administrative use and indexing. Mapping from SNOMED-CT to ICD-10 will improve the value of clinical data as it will:

• Facilitate retrieval of coded data at the desired level of detail depending on the purposes for which the data are being used
• Allow for administrative reporting functions such as reimbursement and statistical analysis not possible with SNOMED-CT alone.

As part of their recommendations for patient medical record information terminology standards, NCVHS urged the federal government to promote the creation and maintenance of mappings between the recommended core set of terminologies, which includes SNOMED-CT®, and medical code set standards designated under the Health Insurance Portability and Accountability Act (HIPAA).

Replacing ICD-9-CM with ICD-10-CM is necessary in order to maintain clinical data comparability with the rest of the world concerning the conditions prompting healthcare services. The longer the healthcare industry continues to use ICD-9, the more difficult it becomes to share disease and mortality data at the time when such global data sharing is critical for public health. For example:

• ICD-10-CM would have better documented the West Nile Virus and SARS complexes for earlier detection and better tracking
• ICD-10-CM also provides the ability to track bio-terrorism events and other public health outbreaks.

i. ICD stands for the International Classification of Diseases. 9 stands for the 9 th revision and 10 for the 10 th revision. CM stands for Clinical Modification (a US version of ICD-9 or ICD-10) ICD-9 and ICD-10 were developed and copyrighted by the World Health Organization (WHO). The WHO no longer supports ICD-9. ICD-10-PCS is a procedural coding system designed by the Centers for Medicare and Medicaid to replace the current inpatient procedural coding system currently included as part of ICD-9-CM.

ii. See AHIMA's Position Statement on Implementation of SNOMED-CT® -- www.ahima.org/dc/positions . AHIMA has also authored a white paper Coordination of SNOMED-CT® and ICD-10: Getting the Most Out of Electronic Health Record Systems, which provides a complete description of the roles of terminologies and classifications in EHR systems and the importance of mapping to effectively use clinical information for multiple purposes.

iii. See AHIMA's Position Statement on Consistency of Healthcare Diagnostic and Procedure Coding – www.ahima.org/dc/positions .

iv. See Medicare Payment Advisory Commission March 2005 Report to Congress Chapter 4: Strategies to improve care: Pay for performance and information technology – www.medpac.gov .

v. Background on SNOMED-CT and Mapping

SNOMED-CT:

• Is a comprehensive, precise clinical reference terminology that contains concepts linked to clinical knowledge to enable accurate recording of data without ambiguity;
• Is specifically designed for use in an EHR:
• It is incompatible with a paper-based health record system.
• Integrated into software applications, it represents clinically relevant information in a reliable, reproducible manner;
• Supports clinical decision support systems, computerized physician order entry systems, and critical care monitoring;
• Facilitates communication among clinicians and improves the quality of data available for research and measurement of clinical outcomes;
• Ensures interoperability of patient information across software applications for disease management , treatments, etiologies, clinical findings, therapies, procedures, and outcomes;
• Provides a common language that enables a consistent way of capturing, indexing, storing, retrieving, and aggregating clinical data across clinical specialties and sites of care;
• Contains concepts linked to clinical knowledge to enable accurate recording of data without ambiguity;
• Works through implementation in software applications, representing clinically relevant information in a reliable, reproducible manner;
• Contains over 364,400 concepts with unique meanings and formal logic-based definitions; more than 984,000 English language descriptions or synonyms; and approximately 1.47 million semantic relationships.

Mapping

The purpose of mapping is to provide a link between one terminology and another in order to:

• Use data collected for one purpose for another purpose,
• Retain the value of data when migrating to newer database formats and schemas, and
• Avoid entering data multiple times and the associated risk of increased cost and errors.

See the AHIMA white paper, Coordination of SNOMED-CT and ICD-10: Getting the Most Out of Electronic Health Record Systems, for a complete description of the roles of terminologies and classifications in EHR systems and the importance of mapping to effectively use clinical information for multiple purposes.

[ About AHIMA | Schools/Jobs | Professional Development | HIM Resources | Foundation | Help | Site Map ]

Copyright © 2008 by the American Health Information Management Association. All rights reserved. No part of this site may be reproduced, reprinted, stored in a retrieval system, or transmitted, in any form or by any means, electronic, photocopying, recording, or otherwise, without the prior written permission of the association. All contents, including images and graphics, on this Web site are copyrighted by AHIMA unless otherwise noted. You must obtain permission to reproduce any information, graphics, or images from this site. Permission must also be obtained to link to any of AHIMA's Web sites. To request permission, fill out the AHIMA Permissions Request or contact permissions@ahima.org.