This month's Focus on Technology section goes beyond the circuits and wires that typically frame any discussion about information technology. This two-part section looks at the language that makes IT work -- or not work -- in improving patient safety and the quality of clinical care and the ability to code, bill and pay for that care. -- David Burda, editor
AHIMA's push for updated coding
As the Bush administration pushes its electronic medical-record agenda forward, the American Health Information Management Association is seeking to add another piece to the puzzle, but its efforts are not getting an enthusiastic welcome from the payer community.
While testifying before the House Ways and Means health subcommittee recently, AHIMA Chief Executive Officer Linda Kloss warned that the benefits of adopting 21st-century health information technology will be diminished if the U.S. continues using a 1970s-era health-data coding system.
Kloss says HHS needs to immediately initiate the regulatory process for adopting the 10th revision of the World Health Organization's International Classification of Diseases for mortality and morbidity, or ICD-10, as 99 other countries have already done. By sticking with the ICD-9's clinical modification for morbidity coding, she says the U.S. is stuck with a system that fails to capture current medical knowledge, wastes money through vague or incorrect coding and requires excessive reliance on supporting paper documents.
Kloss cites how the ICD-9-CM lacks the specificity to classify blood vessel surgeries. The current system has only a single code for "other revision of vascular procedure," while the new ICD-10 procedural coding system, or ICD-10-PCS, captures the specific type of surgery, the specific artery or vein and the use of a graft or prosthesis.
Joel Slackman, managing director of the Blue Cross and Blue Shield Association's Office of Policy and Representation, says opposition to ICD-10 conversion is based on resource allocation and not on any philosophical differences.
Slackman points to a 2003 report from the Blues association that calculates the cost of converting to ICD-10 at $14 billion. Coupled with the recent Annals of Internal Medicine report estimating that building a national health information network will require a $156 billion capital investment, he says the ICD-9 will have to do for the time being. But Dan Rode, AHIMA's vice president of policy and government relations, points to a 2004 Rand Corp. study that estimates the cost of converting to ICD-10 from $425 million to $1.115 billion in one-time costs, with additional annual costs ranging from $5 million to $40 million in lost productivity for coders and physicians during the transition. -- Andis Robeznieks
In search of a standard
Patient-safety reports most often are written in narrative, with hospitals having their own particular forms and styles. But computers like numbers and uniform data fields, without the variability of free text.
That's a problem: Without standardization of how patient-safety events are recorded and turned into data, the new Patient Safety and Quality Improvement Act can't yield the local, regional and national reports on trends and patterns that are among its key goals. Simply said, standardization is not the status quo.
(To learn more about the new federal patient safety law, read the cover story of the Aug. 8, 2005, issue of Modern Healthcare.)
"This stuff has to be computerized, and computers are demanding. You have to define things fairly clearly," says William Munier, acting director of the Center for Quality Improvement and Patient Safety at HHS' Agency for Healthcare Research and Quality. The AHRQ will implement the law and provide technical assistance to the patient-safety organizations that will collect and analyze the data.
What healthcare needs, experts say, is a common framework -- a taxonomy -- for preparing patient-safety reports and a coding system linked to the taxonomy to make the reports computer- and researcher-friendly. Otherwise, a broad national database can't exist. The law gives HHS the authority to select reporting formats, but there is no deadline. A likely scenario is the government will seek industry consensus on an initial taxonomy that will be gradually expanded. First, the AHRQ is taking inventory of available resources.
There are enough taxonomies in existence that arriving at a common one will be both technically and politically challenging, says Dennis O'Leary, president and chief executive officer of the Joint Commission on Accreditation of Healthcare Organizations. For starters, there's the taxonomy for reporting patient-safety events developed by the JCAHO and endorsed recently as a national voluntary consensus standard by the National Quality Forum.
The JCAHO began work on the taxonomy five years ago and is doing some pilot-testing now, says Andrew Chang, director of the Joint Commission's Center for Patient Safety Research. The JCAHO taxonomy is "going to be helpful," Munier says. "It doesn't go down to the level of defining every patient-safety event we might want to look at, but it's a good first step."
A number of other sources exist. Those include the NQF's list of "never events," which have been coded using a popular nomenclature developed by the College of American Pathologists. Also, Munier said there are a score of state-based reporting groups -- all using different methods. And there are reporting systems developed by the Veterans Affairs Department, the Defense Department, the Centers for Disease Control and Prevention, and the national health services in Australia and the United Kingdom.
The AHRQ has hired Rand Corp. to assess state reporting programs and their definitions. The JCAHO, which has an AHRQ grant, will work with Rand to sample hospital patient-safety measures, Chang says. The JCAHO also has begun meeting with developers of major international classifications as part of its contract with the World Health Organization to develop an international patient-safety taxonomy. -- Joseph Conn