Break to ICD-10’s Regulatory Issues


Published in Coding Edge – November 2008


It’s time to discuss the regulatory issues related to ICD-10. As you probably have heard, the proposed rule was published on Aug. 22, 2008. How many years have we been told ICD-10 is coming and it never happened? ICD-10 will happen sooner than we think. The comment period is over and the final rule is expected to be published by the year’s end or at the latest by the time President Bush leaves office.

Unfortunately, the proposed rule is riddled with problems. In the proposed rule, the implementation date is expected on Oct. 1, 2011 and the small health plans conversion on Oct. 1, 2012. What does that mean to us? We will use ICD-10-CM for the larger health plans and ICD-9-CM for the smaller health plans for a period of one year. This is problematic as some health plans will have issues just trying to figure out which system to use. We need more time to prepare. Oct. 1, 2011 may not afford enough time for larger group practices, universities, or our payer community. Let’s review the overall problems with the proposed rule.

Some data and studies conducted by various organizations do not appear accurate and the Department of Health and Human Services (DHHS) is downplaying the overall costs to physician practices. The AHIMA Field test of ICD-10 surveyed a mere 6 percent of physician services, which does not give us a clear picture of the overall costs for training.

Another problem with the proposed rule is that DHHS has labeled ambulatory and physician coders as part-time coders. This statement is not true. Physician coders are not typically part-time coders as most physician coders work at a minimum of 40 hours per week. There are probably some part-time outpatient or physician coders, but this does not equate to the majority of outpatient or physician coders. Given that the average certified physician coder earns at a minimum $21.00 per hour with benefits, lost work time is higher than the Department of Health and Human Services estimates. Benefits are typically estimated at 30 percent. The average coder earns approximately $1,100.00 a week including benefits. While the coder is training for implementation, this would equate to lost work time realizing at $197 million or even more.

Based on the Nolan study, which appears more accurate in the assessment for ICD-10-CM training requirements, training would take approximately 24-40 hours of time to complete. In reality, it will most likely take a minimum of 40 hours of training time. The average cost per day for training is approximately $320.00 at $40.00 per hour. For 40 hours of training, the cost is more likely to be $1,600.00 per physician coder. If the Nolan study is correct and there are approximately 142,000 coders in the United States, training a coder could cost at least $286 million.

Superbill Costs May be Unreasonable

In regard to physician training, the proposed rule indicates most physicians use superbills to select the ICD-9-CM code. That is true in many cases; however, a provider is responsible under the Federal False Claims Act for accurate and complete coding, physicians and other health care providers includes physician assistants and nurse practitioners who must be trained on ICD-10-CM coding. It is not optional for any physician to solely rely on superbills in all instances. Many providers list the most common diagnoses on their superbills, but there must also be a level of ICD-10-CM understanding for all providers. Many times the most appropriate diagnosis may not be on the superbill and the physician must look up the code or rely on a coder to find the correct diagnosis code. Some providers only include procedures on their superbills and code diagnoses using the diagnosis codebook.

With superbill use for reporting ICD-10-CM codes, the practitioner – whether it is a physician or non-physician practitioner – must understand the ICD-10-CM guidelines to properly code and sequence codes correctly and understand the level of specificity required on the claim. DHHS estimates the cost to re-create the superbill is $55.00 per practice. In reality, it can take up to 30-40 hours to redesign one superbill adding all the diagnosis codes and validating their use with the providers, costing as much as $1,000 per provider (not including print costs). If the provider uses a consultant to perform the service, the cost could be as high as $6,000 per practice to redesign the superbill.

The superbills will need revision and with the increase of approximately 17,000 ICD-9-CM codes  to possibly 155,000 ICD-10-CM codes, the use of superbills may become obsolete.

Free Internal Training Comes at a Cost

Another misconception is that physicians in university settings would seek training internally. Most providers who are trained internally contract with a consultant or sources outside of the system to provide training, and if physicians have internal trainers available, there will be a cost to the department to provide the training.

The proposed rule also indicates physician systems are less sophicated than hospital systems. Although an agreeable observation, most medical practices use medical billing software and many practices are moving to electronic medical records (EMRs) which are costly to maintain. The DHHS assumes vendors will upgrade the software at no cost. Based on the issues with HIPAA, there was not one vendor in the country who offered free updates. Many vendors went out of business and medical practices had to scramble to find other software vendors and  purchased new medical billing packages and/or EMRs.

At this time, it is unknown as to how many vendors will not be able to support the ICD-10 conversion; however, because the volume of codes are increased with ICD-10-CM, it might be necessary for some medical practices to purchase new hardware as well as pay for upgrading the medical billing system and/or EMR. This could equate to approximately $35,000 per practice to upgrade. If there are approximately 170,000 group practices in this country, costs could reach as high as $60 million or more.

Since physicians and non-physician practitioners (NPPs) provide the majority of health services, it is reasonable to expect all physicians and NPPs to need approximately 8-12 hours of training on the new coding system. A four hour training session is not an adequate amount of time for a provider of any specialty to learn guidelines plus the ICD-10-CM code sets. With the average cost of training at approximately $40-$50 per hour, 12 hours of training would roughly cost a provider $600.00 not including the 12 hours of lost productivity. If the Nolan study is correct and there are 754,636 practitioners in the United States, DHHS’ training estimate is too low. The approximate training cost would be more accurate at $453 million.

This estimate does not include nurses or medical assistants who sometimes assist physicians with coding nor does it include staff who enter data into the medical billing system and/or staff who perform billing functions. Most physicians in medium to large group practices use coders, billers, data entry staff, surgery schedulers, as well as nurses and medical assistants who all are exposed or work with diagnosis coding in some form. These clinicians and staff would need approximately 12-24 hours of training in ICD-10-CM, which could equate to approximately another $150-$200 million to train.

Productivity Losses Will Take Their Toll

The proposed rule attributes productivity losses to:

  • ¢  Being unfamiliar with the index
  • ¢  Using different main terms and subterms in ICD-10 versus ICD-9
  • ¢  Spending more time reviewing the medical record
  • ¢  Being unfamiliar with ICD-10 versus ICD-9 including guidelines, conventions, etc.

 It’s safe to assume these productivity losses could be as high as $30-40 million depending on the specialty and the complexity of the specialty coding.

For the long term, it has been proven by many sources that greater detail in ICD-10-CM will reduce denials, improper claim submissions, and reduce the number of requests by carriers for medical records; however, the initial transition might increase rejection and denial numbers causing more overtime for physician practices. Expenses could reach as high as $20 million for the first year after implementation.

There are 170,000 physician practices with less than five physicians in the United States. If each physician pays $75,000 for ICD-10 implementation, the cost would be $13 billion. Even at $40,000 per physician the cost would be $7 billion. Overall the cost for medical practices to implement ICD-10-CM could reach as high as $24 billion.

Payers Aren’t Accounted For

The proposed rule does not fully address the burden to the payer community.

Both hospitals and physicians will need their systems updated to accommodate running both ICD-9 and ICD-10 for a minimum of 1-2 years to collect old A/R. Current systems accommodate one form. Whether billing one form or having to accommodate both new and old, this would be a major programming issue. Oct. 1, 2011 does not allow enough time for programming and testing, due to field length and alpha characters. These costs will not be absorbed by software vendors. They will be passed on to their hospital and physician clients. Cost estimates were under estimated. ICD-10’s implementation is scheduled with 5010’s implementation and 5010 does not allow for units, minutes, or time. These are current requirements in various specialties. There are additional changes in both dental and ambulance billing different from how they are done today. How will payers accommodate these changes to pay claims correctly? 

These are major changes. Electronic claims 835/837 are charged to the clients through clearing houses. While it’s advantageous to have remittances electronic, staff will still have to review them for errors and omissions. In addition, there will be increased IT time performing and maintaining the support of these functions. The proposed rule indicates that a reduction of staff due to less need for eligibility, claim status and authentication will be realized. This will generally not happen. Payers maintain this information on Web sites where billing staff can verify information. This step will not change. Billing staff verifies this information due to 45-90 allowable days for billing. Currently, claims are denied by payers with no exception for missing data both electronically or on paper. Changing the forms will not change payer rules.

The same level of billing staff follow-up work will be required to pay claims in a timely manner. The Cost for testing with vendors, clearing houses, and trading partners is clearly under estimated.

 Smooth the Transition with a Plan of Action

So where do we go from here?  First of all, let’s not panic. Once the final rule is published, each medical practice, payer, hospital, etc. should form a committee to begin planning and budgeting. It is easier to plan with a firm deadline to work with. Hopefully, we will have 3-4 years to prepare, but that may be unlikely. Even with as small practices, a plan and budget for the transition is necessary. Initially, it’s not a good idea to spend a great deal of time and money on training. In-depth training should not begin until the year of ICD-10 implementation, however, it is recommended to become familiar with ICD-10 early in the conversion process.

The AAPC is aware of the burden to its members and will make sure our professional coders are trained appropriately. AAPC will begin offering a series of audio’s in 2009 as the first step to ICD-10 awareness. More training including seminars, workshops, etc will become available prior to implementation. We want all of our members and their employers fully informed and prepared for the transition.


Latest posts by admin aapc (see all)

Leave a Reply

Your email address will not be published. Required fields are marked *