Overhauling diagnosis coding

By Christopher Guadagnino, Ph.D.

The government’s recent proposal to replace the ICD-9 code sets – now used to report health care diagnoses and procedures – with massively expanded ICD-10 code sets by October 2011 has provoked an outcry by the physician and health insurance communities, who warn that the deadline will wreak havoc by not giving ample time to adjust to a tremendous increase in coding and billing complexity.

In a proposed rule published in late August, the U.S. Department of Health and Human Services’ Centers for Medicare & Medicaid Services (CMS) said the 27-year-old ICD-9 (International Classification of Diseases, Ninth Revision), with its 17,000 codes, cannot adequately accommodate new procedures and diagnoses, and will start running out of available codes next year. ICD-10, by contrast, contains more than 155,000 codes (approximately 68,000 of which are physician diagnosis codes) and CMS touts its ability to accommodate a host of new diagnoses and procedures; as well as bring far greater “granularity” or detail to diagnosis and procedure coding; provide more detail in electronic transactions; facilitate a nationwide electronic health information environment; and fully support quality reporting, pay-for-performance, bio-surveillance, and other critical activities.

In a separate proposed regulation, CMS wants the health care industry to adopt an updated standard for Health Insurance Portability and Accountability Act (HIPAA) electronic transactions – the ANSI X12 standard, Version 5010. The new standard is an essential prerequisite for claims, remittance advice, eligibility inquiries, referral authorization, and other widely used transactions using the ICD-10 codes, which are not supported by the current Version 4010 electronic transaction standards. CMS proposed that implementation of the new 5010 standards be completed by April 2010 – a deadline which the physician and health insurance communities believe will cause massive confusion and potentially catastrophic disruption of the health care coding, billing and reimbursement system because it overlaps the system-testing period for ICD-10 implementation by 10 months.

ICD-10 will have a huge impact on the costs to physicians of clinical documentation and administrative transactions. According to its regulatory impact analysis, CMS estimated that the cost associated with physician training for adopting ICD-10 could be as high as $165 million, while physician practice productivity losses could be a high as high as $27 million, and total lost productivity costs to providers and health plans for improper and returned claims could be as high as $1.1 billion.

The magnitude of the impending changes has catalyzed physician and insurer group opposition to the deadlines, details of which they will communicate to CMS by the October 21 public comment deadline. CMS will then review the comments and publish the final implementation rules sometime thereafter.

“This is a massive administrative undertaking for physicians and must be implemented in a timeframe that allows for physician education, software vendor updates, coder training and testing with payers – steps that cannot be rushed and are needed for a smooth transition,” according to Joseph Heyman, M.D., board chair of the American Medical Association (AMA). “CMS’ efforts to go full-steam ahead on the transition to the ICD-10 coding system without first pilot-testing the newest HIPAA electronic transaction form (5010) that will be needed to process claims boggles the mind,” says Heyman, noting that “the timetable of just three years for simultaneous implementation of these two new major systems is woefully inadequate, and CMS is setting the stage for major implementation problems.”

Because ICD-10 contains more than nine times the number of codes as ICD-9, moving to ICD-10 is the “largest, most massive change to the health care industry in 30 years, and has the potential for tremendous disruption,” according to Robert Tennant, senior policy advisor on government affairs for the Medical Group Management Association (MGMA). The move will require massive system and workflow changes – including coordinated actions among medical groups and their vendors, clearinghouses and health plans – and it is a recipe for disaster to force such a change without sufficient pilot testing before implementation, Tennant adds.

Recent MGMA surveys indicate that 95 percent of respondents in medical practices would have to purchase software upgrades for their practice management systems or buy all new software; 64 percent concluded that they would have to purchase code-selection software, and 84 percent stated that they did not think public and commercial health plans would be ready to accept claims with ICD-10 codes by October 2011.

The Blue Cross and Blue Shield Association (BCBSA) agrees, warning that CMS’s proposed ICD-10 and 5010 implementation timeframes are unworkable and will cause a meltdown in the health care industry, including inaccurate and delayed payments to providers and consumers, an inability to detect fraud and abuse, and unnecessarily higher total costs of implementation due to the accelerated timeline, according to Joel Slackman, BCBSA’s managing director for policy in the office of policy and representation.

A coalition of organizations including BCBSA, MGMA, AMA, some 50 other national physician organizations, and nearly every state medical society, supports the move to ICD-10 but calls upon CMS to extend the implementation date by two years. The coalition notes that the government’s own advisory body, the National Committee on Vital and Health Statistics (NCVHS), said the health care industry should adopt ICD-10 only after the 5010 transaction standards are fully implemented and tested. The coalition endorses NCVHS’s recommendation that CMS allow consecutive implementation: two years for conversion to 5010, then another three years before conversion to ICD-10 – by October 2013.

Said CMS Acting Administrator Kerry Weems, in a statement announcing the proposed rules, “We recognize that the transition to ICD-10 will require some upfront costs, but each year of delay would create additional costs, both because of the limitations of ICD-9 and because of the need to employ the greater precision that ICD-10 codes provide to support value-based purchasing of health care and other initiatives. We will continue to work collaboratively across the health care system to ensure a smooth transition to use of the updated transaction standards and ICD-10.”

Since the transition to ICD-10 won’t be for another three years, at the earliest, physicians do not yet need to be trained in the specific codes. They should, however, prepare for the overhaul now by researching its likely impacts on their practice and care delivery systems; by developing a long-term budget for new practice management software, systems and staff; and by contacting their billing system vendor and their health plans, according to Tennant.

Finer Granularity

CMS notes that ICD-10 provides much more information and detail within the codes than ICD-9, including:

· Significant improvements in coding primary care encounters, external causes of injury, mental disorders, neoplasms, and preventive health.

· Advances in medicine and medical technology that have occurred since the last revision.

· Codes with more detail on socioeconomic, family relationships, ambulatory care conditions, problems related to lifestyle, and the results of screening tests.

· More space to accommodate future expansions.

· New categories for post-procedural disorders.

· The addition of laterality – specifying which organ or part of the body is involved when the location could be on the right, the left, or could be bilateral.

· Expanded distinctions for ambulatory and managed care encounters.

In its proposed rule, CMS illustrated how ICD-10 would overcome limitations of ICD-9. For example, ICD-9 contains a single procedure code that describes the endovascular repair or occlusion of head and neck vessels, and does not describe the artery or vein on which the repair is performed, the precise nature of the repair, or whether the approach is a percutaneous procedure or is transluminal with a catheter.

CMS summarized the shortcomings of ICD-9:

· ICD-9 is outdated, with only a limited ability to accommodate new procedures and diagnoses.

· ICD-9 lacks the precision needed for a number of emerging uses (for example, pay-for-performance and biosurveillance).

· ICD-9 limits the precision of diagnosis-related groups (DRGs) with very different procedures being grouped together in one code.

· ICD-9 lacks specificity and detail, uses terminology inconsistently, cannot capture new technology, and lacks codes for preventive services.

· ICD-9 will eventually run out of space, particularly for procedure codes.

Adoption of the ICD-10 code sets, according to CMS, will:

· Support value-based purchasing by accurately defining services and providing specific diagnosis and treatment information, such as identifying cases of MRSA and other specific conditions, and would further Medicare’s ability to detect and prevent program abuse.

· Support comprehensive reporting of quality data.

· Ensure more accurate payments for new procedures, fewer rejected claims, improved disease management, and harmonization of disease monitoring and reporting worldwide.

· Allow the United States to compare its data with international data to track the incidence and spread of disease and treatment outcomes because the United States is one of the few developed countries not using ICD-10.

CMS maintained that ICD-10 will lead to fewer rejected claims by reducing the number of cases where copies of the medical record need to be submitted for clarification for claims adjudication. For example, ICD-10 injury codes identify in detail the fracture site of a malunion or non-union, while the ICD–9 codes for malunion and non-union do not identify fracture site. If the payor required this information to adjudicate the claim, the provider would need to send a claims attachment. As another example, because ICD-10 injury codes provide finer detail in identifying bilateral fractures, if a patient fractured both wrists, two codes could be assigned – one code identifying the left wrist fracture and a separate code identifying the right wrist fracture. ICD-9 does not provide this detail and if a provider wanted to report fractures of both wrists and reported the diagnosis code twice, the claim would be rejected, CMS noted.

The increased granularity of ICD–10 would allow case management organizations to better identify candidates for disease management programs, and to better adapt the disease management program to the individual once enrolled, according to CMS.

ICD–10 can also improve quality measurements, patient safety and the evaluation of medical processes and outcomes because it allows new procedures, diagnoses and technologies to be easily incorporated as new codes for both existing and future clinical protocols, CMS said. In an age of electronic health records, it does not make sense to use a coding system (ICD-9) that lacks specificity and does not lend itself well to updates, CMS noted.

Once initial confusion stemming from ICD-10’s complexity subsides, fraud and abuse could be reduced because finer granularity leaves fewer “gray areas” and less ambiguity in coding, CMS added.

Impact on Physicians and Insurers

Critics of CMS’s ICD-10 and 5010 implementation proposals say it is unrealistic to expect the health care industry to handle the massive complexities and to conduct sufficient pilot testing in such a short timeframe, while some question whether the benefits of transitioning to ICD-10 outweigh the costs at all.

The finer granularity of the new codes, touted by CMS, will impose considerable administrative burdens on physicians – who will have to diagnose patients with far greater specificity to satisfy the new data fields, and burdens on health plans – which have to adjudicate claims that contain far more complicated data. As an example, Tennant notes, there are four ICD-9 diagnosis codes for sprained and strained ankles, while there are 72 ICD-10 codes – adding details such as cause, location and angularity of injury. “ICD-10 has a code for ‘struck by a parrot,’ and a different code for ‘struck by a macaw.’ Should physicians track down what kind of bird caused a patient’s injury?” asks Tennant.

“If health plans are requiring the most specific, most granular codes or else reject the claim, who will pay the physician for arriving at the most specific code?” asks Tennant. As an example, to determine the type of Down Syndrome – required under the ICD-10 taxonomy – a genetic test is needed.

Whether physicians will utilize the finer granularity is another question. A study of an Arkansas Blues plan found that physicians who diagnosed patients with acute sinusitis – which has six available codes under ICD-9: maxillary, frontal, ethmoidal, sphenoidal, pansinusitis, and unspecified – chose the “unspecified” code on their claims 82 percent of the time. There are 14 ICD-10 codes for sinusitus. “If providers are not using the distinctions available in ICD-9, what will it take to get them to use the finer gradations of those previously unused distinctions in ICD-10?” asks Slackman.

Reimbursement implications of that question are not yet known, Slackman says. “Until we start seeing a couple of years of ICD-10 coding, we won’t know enough about how physicians are changing their coding behavior – whether they’re upcoding or downcoding – to know how to respond. The first year of data won’t be robust enough to know,” he notes.

Slackman points out that CMS said nothing about how ICD-10 will affect hospitals’ prospective payments. “To us, that is a major omission that makes it very difficult to project the impact on health plans’ cash flow, and the impact on physician coding behavior,” he says. “Until we know how DRG payments will change, we won’t know how physician payment will change,” Slackman adds.

It also remains to be seen what level of coding specificity local Medicare carriers and commercial health plans will require for physicians to be reimbursed. Tennant says CMS requires diagnoses to the most specific code available, but he doesn’t know whether carriers will require the full spectrum of granularity as a precondition to reimbursement.

According to Slackman, many commercial health plans will “pend” a claim that uses an “unspecified” code, and request further information, which he says can delay payments and lead to a cash flow problem for physicians. Over time, given the movement toward quality performance measurement and pay-for-performance, public and private payors are going to start to ask for greater specificity in claims data, or else pend the claim, Slackman predicts.

Physicians will experience more burden to qualify for pay-for-performance bonuses, which could defeat the purpose of such programs, says Tennant. MGMA analysis has found that many physicians fail to recoup the extra expense their practice goes through to achieve their P4P bonus payments. Under ICD-9, such programs may have used metrics requiring 10 to 15 codes. Under ICD-10, those same P4P programs could potentially require 60 codes, he adds.

Slackman believes that physicians may eventually see higher P4P reward payments under ICD-10 as health plans move toward, and devote additional funding to, outcome-based payments. Incentives will eventually align with those codes requiring greater levels of specificity, he says.

Coding aids which physicians have come to rely upon may no longer be available under ICD-10. The primary way many physicians submit codes is the one-page “superbill,” which contains their most commonly used ICD-9 codes for them to circle. Under ICD-10, they won’t be able to do that, predicts Tennant, as a vastly greater number of codes will make the superbill too unwieldy. “Physicians will need to have some type of code selector software in their exam room, either on a desktop computer or hand-held device,” Tennant says.

Because of the extreme specificity of diagnosis notes required under ICD-10, physicians may also lose one of the major efficiency strategies they had to deal with HIPAA requirements: relying on coding support entities known as clearinghouses to convert non-compliant claims data into a format that is acceptable to a health plan, says Slackman.

According to its regulatory impact analysis, CMS estimated that the cost associated with physician training for adopting ICD-10 will be about $82 million, while it could be as high as $165 million (the upper range, using CMS’s cost analysis assumptions). Physician practice productivity losses – the cost resulting from a slow-down in coding bills and claims because of the need to learn the new coding systems – were projected by CMS to be about $11 million, or as high as $27 million. CMS estimated the cost to providers (CMS lumped large and small provider groups, as well as institutional providers such as hospitals into one provider estimate) of system changes for software vendors to be about $96 million, or as high as $137 million.

Acknowledging that the new code sets will likely produce a temporary increase in coding errors, especially on the part of physicians, CMS estimated that additional returned claims processing would cost providers and health plans (no physician-specific figure was given) $329 million in the first year following implementation, $165 million in the second year, and $49 million in the third year. CMS’s global estimate of total lost productivity costs for improper and returned claims for transitioning to ICD-10 is $543 million, and could be as high as $1.1 billion.

According to Slackman, one-third of BCBSA’s plans have begun the ICD-10 conversion planning process through the work of interdisciplinary teams comprising chief medical officers and experts in medical policy, provider contracting, provider relations, benefit design, information technology, and fraud and abuse. “Early findings suggest that this is going to be the biggest change yet, far bigger than Y2K or the move to HIPAA transaction requirements,” says Slackman.

If CMS’s final rule keeps the Oct. 2011 implementation date for ICD-10, says Slackman, health plans will need to cut corners and will develop workarounds that will end up costing more money in the future, just to meet the deadline.

Experiences with other HIPAA mandates, including the original 4010 transaction and the National Provider Identifier (NPI), illustrate the time needed to implement even the simplest of transactions required under HIPAA, says Slackman. Implementation of the NPI transaction – the simplest of the mandated HIPAA transactions – took four years and four months. It makes no sense to provide less than three years for the massive overhaul that the 5010 and ICD-10 changes will require, Slackman adds.

The American College of Physicians (ACP) opposes adoption of ICD-10 diagnosis codes outright, as it is not convinced that the benefits of adoption outweigh the complexity and costly disruption to physician practices, particularly to small physician practices that are least able to absorb additional costs, according to Brett Baker, ACP’s director of regulatory and insurer affairs. In addition to the practical challenges of system conversion, physicians have no way of gauging the payment implications of the new codes, says Baker. Some payors may use the greater coding detail to expand medical necessity denials and limit payment for some codes, he adds.

Based on ACP’s work with committees of internists, Baker’s sense is that there is probably not much awareness in the physician community about ICD-10 and its implications.

Preparation Advice

Physicians should ask these questions of their practice management and billing system vendors, says Tennant:

· Will you be producing an upgrade to accommodate ICD-10 and 5010 transactions?

· Will the upgrade be available for my version of software (including older versions)?

· When will the upgrade be available?

· What will it cost my practice?

Physicians should ask these questions of their health plans, suggests Tennant:

· When do you anticipate being ready to test 5010 electronic transactions for: claims, remittance, claims status inquiries, patient insurance eligibility verification, and prior authorization?

· When will you “go live” with these transactions?

· Will you publish a “companion guide” (containing payor-specific coding format standards) to facilitate 5010 transactions?

· Will you be using CMS’s “crosswalk” from ICD-9 to ICD-10 (a short-term bootstrapping device used to map old codes to new codes, especially in the beginning of the transition), or will you be using your own crosswalk?

· What level of ICD-10 code specificity will you require?

· Will you pay for the additional tests to arrive at the more specific diagnoses?

Physicians should assign a point person in charge of fact-finding to keep up with the developments – particularly dates and costs – related to 5010 and ICD-10 phase-in, advises Tracey Glenn, director of practice management consulting at PMSCO Healthcare Consulting in Harrisburg, Pa. Free training programs for practice staff will be offered by the CMS, the American Health Information Management Association (AHIMA), and the American Academy of Professional Coders (AAPC), says Glenn.

The practice may ultimately have to purchase new coding software to translate diagnoses into ICD-10 codes, while some electronic medical record systems may be able to accommodate the new codes, according to Glenn. “Most of the vendors we’ve talked to are aware that this is coming, and the bigger vendors have an idea of what to do,” adds Glenn, noting that “we’ve seen practices with ancient legacy software that may not be able to make the change.”

The top five practice management system vendors, which represent some 80 percent of the market, look to be prepared to meet 5010 and ICD-10 requirements, while the other 20 percent of mostly smaller vendors will have a significant challenge to make a successful conversion, according to Lee Barrett, executive director of the Electronic Healthcare Network Accreditation Commission (EHNAC).