Opinion: Are ICD codes really necessary?

January 5, 2016

After much angst and several delays, the healthcare industry recently completed the ICD-10 transition, but was it really necessary?

After much angst and several delays, the healthcare industry recently completed the ICD-10 transition. In fact, despite last-minute pleas to delay the update again, we now have roughly one full quarter of working with ICD-10 behind us.

The good news is U.S. healthcare did not experience any major disruptions once the October 1, 2015 deadline passed. Sure, there were some glitches, but that happens any time something new is introduced.

Now for what some might consider the bad news. The U.S. was so woefully behind in moving to ICD-10 that ICD-11 is already on the horizon. In fact, the World Health Organization (WHO) is already looking for beta participants, with the final revision currently due in 2018. 

Not to worry. If history has shown us anything it will still be several years before U.S. healthcare providers and payers are expected to move to the new standard. After all, the WHO completed its work on ICD-10 in 1992, and the U.S. clinical modification (ICD-10-CM) was available beginning in 1999.

The bigger question, though, is whether we even need ICD for claims anymore. To answer that question we must first look at its history.

How we got here

The original intent of ICD coding was to monitor diseases throughout the world. Introducing a standardized classification for those diseases made it easier to build statistical models, especially in the precomputer days when all patient data was contained in paper records. The first classification was published in 1893; in 1946 the WHO took over administration of ICD codes.

Note that nowhere in that history does it mention using ICD codes to pay insurance claims. That came later, as several countries realized that having a ready-made system of classification for healthcare procedures would provide a good structure for all of the data being gathered from healthcare providers.

Rather than having to read through individual charts or notes, submitting claims based on ICD codes would greatly expedite claims review. This meant payers could reduce their administrative burden and providers could get paid faster.

Next: Changing world

 

 

Changing world

Here is the issue today. Between the original approval of ICD-10 and today, the world in general and healthcare specifically, changed significantly thanks to technology. Finding a computer in a physician’s office or a patient room in a hospital was a rarity in 1992. In 2016 it’s the norm. In fact, many healthcare workers not only use PCs but smartphones and tablets in their daily work flows.

While there are still a few holdouts left, the vast majority of healthcare encounters are now captured via electronic health records (EHRs). As of the end of 2014,

had adopted at least a basic EHR system, and nearly 97%reported having a certified EHR technology, according to the Office of the National Coordinator for Health Information Technology. Ambulatory providers are demonstrating similar numbers. These figures are expected to continue to grow, especially as penalties for failure to meet the meaningful use provisions of the Health Information Technology for Economic and Clinical Health (HITECH) Act become more onerous.

The result is much of the data about healthcare services is already being captured in a standardized format and is available electronically. With improvements in natural language processing and cognitive computing, we are even adding the capability of easily working with data being captured in unstructured fields.

All of this begs the question: Why force clinicians to go through the extra step of assigning another code to it?

Next: Exploring the need for ICD-10

 

 

In today’s world it just seems like duplicated, wasted effort. Instead of going through all the machinations of converting from ICD-10 to ICD-11 down the road, wouldn’t it make more sense instead to start working on the ability to submit electronic health records directly for claims processing? Think about what that would mean.

All of the detail about a particular healthcare encounter would already accompany the claim. If everything is submitted correctly, and the enabling technology is in place, the claim could be administered automatically, no human hands would have to touch it. That would do for today’s organizations what ICD codes originally did back in the day: speed up reimbursement and reduce administrative overhead.

If a claim does require human review, payers would be able to eliminate all the back-and-forth with providers to obtain background information. The full record with the diagnosis and services delivered would already be there, available for the payers’ on-staff medical professionals to review. Concerns could be expedited and resolved, reducing the burden on both payers and providers. With the healthcare industry’s move toward bundled payments and other risk-sharing strategies, having all the information about an encounter in one place will become even more important.

Having access to this deeper level of data could even help payers use analytics to improve benefit design. Rather than simply looking at frequency of occurrence, they could dig below the surface to develop meaningful benefits that better serve members, especially as we enter the age of personalized medicine.

What about the WHO’s original purpose of tracking disease worldwide? That could be accommodated by developing algorithms that recognize the data in EHRs and automatically convert it to the correct code. This data can then be reviewed and corrected, where necessary, by coders at the payers to ensure the WHO is receiving quality information.

Next: Technology update

 

 

Technology update

The bogey in all of this is whether payer back office technologies developed in the pre-Internet days (i.e., early to mid-1980s) will have the capacity for this level of data processing. It may require a technology refresh to a more agile relational database model. But that technology refresh is long overdue anyway. It will become one more justification to make the investment so payers are prepared for the 21st century, and beyond.

Making the transition to a new ICD version is costly and time-consuming. And in today’s world it may also be unnecessary. Before we get too far down the road with yet another angst-filled transition, perhaps it’s time to stop repeating the problems of the past and begin looking at more sensible alternatives.

Craig Kasten is chairman of SKYGEN USA, a collection of benefit solution companies that brings together a distinguished mix of next-generation benefit management and technology tools that help healthcare organizations be market-leading and reform-ready. Craig can be reached at craig.kasten@skygenusa.com.