ICD-10 and Clinical Research

When the U.S. healthcare system finally transitioned to ICD-10 on Oct. 1, 2015, the clinical research community was quick to embrace the move.

As the last industrialized nation to implement the greatly expanded code set, its use means that clinical trials for international pharmaceutical manufacturers in particular can finally align on a global basis.

In terms of utilizing the code set itself, the primary change in the research environment was swapping ICD-9-CM V70.7 for ICD-10-CM Z00.6 (examination of participant in clinical trial), typically used in conjunction with Condition Code 30 (non-research services provided to all patients, including managed care enrollees, enrolled in a qualified clinical trial). The transition also means that trials which have been underway since before the change date must operate in a dual environment.

Most agree that the move to ICD-10 is well worth these temporary complications, because it enables research organizations to conduct a true apples-to-apples analysis when data is collected from multiple countries. This is particularly true when patient records cannot be accessed and researchers instead must rely on code analysis.

The more granular detail in ICD-10 also holds the potential to uncover previously hidden relationships or phenomena, which could pave the way to new procedures or therapies capable of saving thousands of lives and billions of dollars.

Identifying & Eliminating Challenges

While clinical research organizations welcomed ICD-10 with open arms, the transition wasn’t problem-free. Nor did it reflect the scale of issues experienced in other countries. In Canada, for example, healthcare organizations experienced a 67% decline in post-ICD-10 productivity.

Some of the most significant – and ongoing – challenges for research sites are lost productivity and staff training. Most found these initial productivity dips to be temporary, with coders returning to their normal efficiencies after just a few short weeks of working with the new coding system on a daily basis.

Another issue with a longer reach is the financial impact of ICD-10 on research sites. Specifically, incorrect linking/mapping within software systems can cause a spike in denials and, subsequently, impact reimbursements and slow the revenue cycle.

Again, steps were taken to lessen the blow. CMS decided against denying or auditing claims just for specificity within the first year of implementation as long as the code is from the correct ICD-10 coding family. The agency also authorized advance payment to physicians if Medicare contractors cannot process claims due to ICD-10-related problems – a real safety net for physician-directed research studies.

ICD-9’s lack of specificity also had ramifications for research organizations that wished to use reimbursement information to compare relative healthcare costs for new and innovative treatments or procedures across the globe. Because U.S. providers were using ICD-9, they were forced to report broader and/or more generic diagnosis codes, which skewed reimbursement and patient cost data when compared to countries using ICD-10.

With the U.S. finally aligned under ICD10, higher quality health cost data reporting is expected to assist with setting reimbursement rates that more accurately reflect the expense involved in treating specific illnesses.

Finally, because ICD-9 diagnosis codes are much broader than ICD-10, it can be more difficult for users to discover and code specific diagnoses. While conversion systems such as general equivalence mappings (GEMS) are available to draw a translation between ICD-9 and ICD-10, the conversion process still contains many pitfalls.

One such example is “One-to-Many” code mapping, wherein a single ICD9 code maps to multiple diagnosis codes in ICD-10. For instance, this may occur when ICD-9 codes a certain disease by locality only, while the more specific ICD-10 system reflects additional factors such as severity and laterality. Research coders will find it necessary to use multiple ICD-10 codes when a single generic ICD-9 code would previously suffice.

While the short-term impacts were in many ways far less than what was originally anticipated, attention must still be paid in order to ensure long-term success in the new ICD10 environment. Solid strategies require the ongoing analysis of coding accuracy to identify problem areas and the provision of tailored training to close any gaps. This includes close tracking of denials to identify any trends or trouble points so they can be corrected with additional training or refresher courses for both coders and clinician staff on the use of ICD-10.

Benefits on a Global Scale

Elimination of the limitations placed on clinical research data by the U.S.’s prolonged use of ICD9 is one of the greatest benefits the nation will see from the transition. Because it had run out of room for new codes, ICD-9 had basically been frozen since June 2012.

That is when the Centers for Medicare and Medicaid Services (CMS) posted the final updates to the code set, meaning something as common as a new flu strain could not be properly coded under ICD-9. This caused great concern for research organizations that were left without a valid means of properly coding or even identifying new or varied diseases and medical technologies.

Further, while research organizations worldwide were able to code using the rich detail available under ICD-10, U.S. sites were left to track and monitor patients using non-specific codes. Ebola, for example, was coded under “other specified diseases due to virus” or “hemorrhagic fever not elsewhere classified.”

For researchers, this meant that from a statistical standpoint, there was no way to differentiate between patients with H1N1 (swine flu) or H5N1 (avian flu) and Middle East Respiratory Syndrome (MERS). Nor could those with Ebola be distinguished from those with a similar disease, making it almost impossible to identify candidates for research studies or adequately track and record progress of any therapies being tested.

In addition, U.S. data submitted to the World Health Organization (WHO) prior to the ICD-10 transition had to first be translated into ICD-10 so it was compatible with the rest of the world. However, because ICD9 lacks the specificity of ICD10, the translated data for tracking global illnesses was necessarily generic and therefore subpar to that submitted by other countries.

The Promise of Better Outcomes

It is imperative that researchers have access to accurate and comprehensive data when tracking potential global epidemics, pandemics and deadly outbreaks, as well as the ability to compare apples-to-apples when studying new drug therapies.

ICD10 enables that level of tracking in part because the U.S. is now compatible with the rest of the industrialized world. But more importantly, it provides a granular detail that enables early and better identification of therapeutic outcomes.

For clinical research trials in particular, the more expansive coding system is expected to lead to better information gathering on comorbidities and serious adverse events. It should also help streamline the patient eligibility process through more accurate medical history reporting, better identification of candidates, and less screening failures.

Finally, the heightened granularity of ICD-10 should lead to better outcome assessment reporting – and perhaps even to a more refined approach to clinical research by enabling the inclusion of more defined patient subsets based on medical history and study indication.

Jessica Krone (JKrone@pfsclinical.com) is Team Manager with PFS Clinical, which offers turn-key administrative solutions that address the key pain points of establishing and running a clinical trial office. Staffed by skilled research professionals, PFS Clinical’s suite of services include coverage analysis, financial management, budget and contract negotiation, claims review, corporate marketing, study advertising, patient recruitment, and business development. For more information, visit www.pfsclinical.com.

About The Author