Web Exclusive

Time to Let the Air Out of the Tires on ICD-10
By David Winn, MD, FAAFP

Few topics in health care these days seem to generate as much angst as the long-awaited transition from ICD-9 to ICD-10. It was supposed to happen in 2013, then in 2014, and is now scheduled for October 1, 2015. So far.

The latest controversy on the topic has state physician organizations in New York and Texas—two of the states with the most members—encouraging their members to press their representatives in the Senate and the House to sneak a line item onto another must-pass bill putting the transition off two more years to 2017. Which, incidentally, is when the new standards for ICD-11 will be introduced, officially putting the United States two revisions behind.

The rationale is that the cost and burden of conversion, not to mention learning a more complex system, is too high for physicians to bear at this time. The cost may not be the hundreds of thousands to millions of dollars an early study claimed, but it still could fall somewhere between $10,000 and $20,000.

What makes it so difficult and expensive? ICD-9, which is already considered cumbersome by many, uses 13,000 codes. ICD-10 increases that number to 68,000 codes. It also requires physicians and billers to use a much greater degree of specificity when entering the information. And since proper ICD coding is required for reimbursement by payers, the physician organizations reason that problems with the conversion could have serious financial consequences for practices. The number of delayed payments is likely to increase exponentially if physicians aren't properly prepared. This looming threat is often cited as a reason for physicians retiring early or selling out to hospitals.

Still, whether it happens in 2015 or 2017, the amount of work and the degree of difficulty will be the same. Clearly, having two more years to prepare is unlikely to make much of a difference because physician practices have already had two more years to prepare, and they're now being encouraged to ask for two more. They are merely trying to forestall the inevitable.

The real issue with the transition to ICD-10 isn't the level of complexity—it's the complexity of the solutions being proposed. It's like the story of the truck that gets stuck in the underpass on the highway. A group of civil engineers is brought out to examine the situation, and each offers a complex and expensive solution on what can be done to raise the bridge high enough to allow the truck to pass beneath it. Then a small child wanders by and asks, "Why don't you just let the air out of the tires?"

The approaches that have been proposed and tools that have been developed for making the transition to ICD-10 are similar to figuring out how to raise the bridge. They essentially involve two things. The first is assuming physicians (and billing coders) should continue to work the way they've always worked. We all know what Einstein said about doing the same things over and over. The second is determining how to make those 13,000 ICD-9 codes map to the new 68,000-code ICD-10. The math just doesn't work.

What is needed is a fresh approach that not only solves this issue but also makes working with ICD codes in general easier—especially with ICD-11 coming soon.

Think about this: What if we had a tool that allows physicians to enter information using ICD-10 codes and then maps backward to ICD-9 instead? It's always easier to go from more to fewer possibilities rather than fewer to more. Physicians could start learning with it today while having the information they need to submit for reimbursement now. By the time the transition is made in 2015, ICD-10 would be second nature to them.

That still doesn't solve the main problem of those 68,000 codes, though. Scrolling through lists or trying to memorize basic parameters for searching are both time-consuming and not a good use of physician time, especially when physicians already are being pressed to see more patients in a day and complete many new tasks.

Instead, what if the tool helped narrow the possibilities through a series of simple drill-downs. Anyone who has purchased a product at a big-box online retailer has seen this technology in action. You begin with a general category—say, televisions—then start selecting parameters, such as screen size, LCD or LED, certain features, price range, etc. By the time you're finished, the technology has narrowed the field from hundreds of possibilities to two or three selections.

The same technique can be applied to ICD-codes. Suppose the patient presents with a broken arm. The physician sees the patient and diagnoses a fracture of the right ulna. By entering the search term "ulna fx" or clicking on an anatomical illustration, the tool presents all the associated possibilities. By further narrowing the search to the right arm, one-half of the code possibilities are eliminated. Click on a few more parameters related to the injury and the final, correct ICD-10 code is identified—all in a matter of five to 10 seconds (depending on the nature of the injury or illness). If billers have a SuperBill that requires more information for reimbursement, they can use the standalone tool to monitor its progress via a URL link and request that the physician click through a few more choices to provide the needed information. Neat, simple, and no extra training required.

Finally, imagine if, rather than paying thousands or tens of thousands of dollars for this miracle tool, physicians found it was available for free. That might encourage even the most resistant physicians to give it a try and ease their way toward using ICD-10.

It's not the way things are usually done, but it's an approach that makes sense. Instead of figuring out how to raise the bridge, perhaps it's time to think about letting the air out of the ICD-10 tires. In a year, we may just wonder what all the fuss was about.

— David Winn, MD, FAAFP, is CEO of e-MDs.