October 27, 2014

ICD-10: Not One But Two Deadlines Coming in 2015

By D’Arcy Guerin Gue

We all know that the official deadline for implementing ICD-10 has been moved to Oct. 1, 2015. Any healthcare organization that singles out that date as its primary focus, however, is placing itself at grave financial risk. While there are some portions of ICD-10 that simply cannot be implemented in production early, like sending ICD-10 codes on claims, there are others that can – and simply must – be implemented earlier.

Meeting another deadline – April 1, 2015 – will be a critical factor. By that date, six months prior to final implementation, organizations should have increased staffing to meet the challenges of the ICD-10 transition. Similarly, by then their systems should be ready to support the staff impacted by ICD-10 as they begin to use the coding system on actual patient records.

Why so early? ICD-10 will affect almost the entire spectrum of healthcare, but the broadest impacts will come in the area of productivity. This will be a result of the increased complexity of ICD-10 coding support and the general unfamiliarity with the new requirements. The implementation of ICD-10 will impact the productivity of any staff member who documents, determines, records, or uses an ICD-10 diagnosis or procedure code. Nowhere will the impact be more obvious than in the health information management (HIM) department, where a coder’s main job function is to determine which of these codes to use. Mitigating that impact will require time and ample preparation.

How Significant Will the Productivity Impact Be?

Making precise estimates of exactly how great ICD-10’s productivity impact on coders will be is difficult. A simultaneous replacement of diagnosis and procedure codes on this scale has never been attempted. Fortunately, studies published during Canada’s transition to ICD-10 and more recent efforts in the United States provide a good basis for building a model of these anticipated productivity impacts. 

Impact of ICD-10 Coding on Coder Productivity

Study

Coding
Productivity

Time of measurement

Humber River Regional Hospital, Ontario Canada

50%

3 months after start date

80%

1 year after start date

AHIMA ICD-10-Field Testing Project

53%

Impact at start date

HIMSS/WEDI ICD-10 National Pilot Program

50%

Impact at start date

These studies and other data suggest that coding productivity will initially decrease by 50 percent and will not start returning to normal for at least six months. At best, some organizations are likely to gradually achieve coding levels of 80 percent of pre-ICD-10 levels. Combining this productivity decrease with the overall expansion of healthcare as the baby boomers age is a recipe for permanent productivity loss.

The Looming Staff Shortage in Medical Coding

Simple math tells us that a 50-percent decrease in productivity will call for double the coding resources. Similarly, a permanent 20 percent decrease would call for about 25 percent more resources. Because good, experienced coders are already not easy to find, and being as every hospital will face the burden of needing more resources, a staffing shortage is essentially inevitable. The fallout is expected to include:

  • A labor market in which experienced coders will be lured away from organizations by other facilities. Ironically, this is a tactic of some of the same coding vendors you plan on working with to meet the staffing shortfall, in addition to other healthcare providers (I know of at least one hospital that has contracted with several large coding vendors, both to meet their productivity demands and to get no-hire clauses to prevent poaching of current staff.)
  • Rising costs to satisfy higher coder salaries needed to retain your staff.
  • Increased costs to pay third-party vendors to augment your coding staff.
  • Increased denials and rework due to decreased coding accuracy. Also expect to spend more on additional supervisory resources and quality assurance time needed to improve coder accuracy.
  • An increased number of queries for completion of documentation, which will slow coders (and AR) even more.

Using At Least Six Months to Mitigate the Risks

Many leading hospitals have identified that they will need at least six months of increased staffing before the ICD-10 transition. This requirement will impact systems as well, demanding six months of system readiness to support the coders, documenters, and other impacted personnel as they use ICD-10 in the context of actual patient records. It’s also important to realize that the requirements for staffing will come even before the deadline, as training requirements and new coder onboarding will impact productivity months before the transition date arrives. All of this adds up to the need to recruit and train coders now, not at the last minute.  

This six-month span – starting on April 1, 2015 – will require system readiness in order to accept ICD-10 codes and to enable coders to perform dual coding (coding portions of their workload in both ICD-9 and ICD-10). Benefits will include:

  • Mitigation of the expected initial productivity losses by enabling coders to become comfortable (and fast) with ICD-10 coding. The Humber River analysis, referenced in the table above, indicates that it may take six months of coding to begin seeing these benefits.
  • Actual data for internal testing, which will provide a broader range of ICD-10 data that is often identified in test cases, which in turn will improve the quality of testing. This internal testing should be as fully integrated as possible and include as much documentation, coding, and claims submission systems as are available in a test environment.
  • Advance actionable data for assessments of coder productivity and accuracy, which will facilitate early, focused remedial training. This precise data will enable better predictions of the quantity of coding resources that will be required after the transition to ICD-10.   
  • Actionable data for documentation readiness assessments, which will be used to identify problem areas and address shortcomings in provider documentation.
  • Real data for reimbursement analysis to take the place of general (and risky) assumptions about the revenue impacts of the transition, particularly as it applies to DRG grouping of documented codes that arise from an organization’s particular case mix.

Dual coding also will provide live data for end-to-end testing with payers, which is critical to the reduction of revenue cycle risk providers are facing from ICD-10. This is primarily because of the breadth of the ICD-10 changes and the ICD-10 reimbursement structures that are based on a normalized case mix and ICD-9 converted data (not data that was natively coded in ICD-10). Some early testers have noted that their ability to produce claims with live data was impaired because of poorly remediated interfaces. 

Also of note are the results of the Centers for Medicare & Medicaid Services (CMS) nationwide claims acceptance testing (http://www.cms.gov/eHealth/ListServ_ICD10AckTestWeek.html) conducted with providers in March 2014. During testing, CMS accepted 89 percent of the test claims, which compared with normal claim acceptance rates of 95-98 percent. However, it was noted that some testers might have included flawed data as part of their submissions. Additionally, early reports regarding end-to-end testing from Aetna (https://www.wedi.org/forms/uploadFiles/36BEE00000542.toc.7.31_Payer_Combined.pdf) indicated a 10-percent variance in DRGs initially, although this reduced to 3-4 percent after coding and data entry errors were eliminated.

We continue to learn that many hospitals and other healthcare organizations are far from ready to start coding with ICD-10. Others have made much progress in the last two years. Without a doubt, the deadline delay until Oct. 1, 2015 represents an opportunity for all to ensure that they can convert on time. Perhaps just as important, the delay has provided the precious six months to help organizations do more to reduce productivity-related and related revenue risks that will come with the conversion.

Start working now to enable impacted staff to work with upgraded systems and actual patient records by April 1, 2015 – and plan for plenty of practice, results assessments, retraining, and refining.

About the Author 

As a co-founder of Phoenix Health Systems, D’Arcy has had leadership roles in the growth of the company. Currently, she leads overall corporate administration, marketing and industry relations, services development, human resources, and knowledge management. She has led various strategic initiatives, including the development of ICD-10 services, HIPAA-based security and privacy compliance tools, and online education programs.

Comment on this Article