How the Change to ICD-10 Codes Will Improve Healthcare Analytics
The upcoming (October 1, 2014) move from the current set of ICD-9 to ICD-10 codes has much of the healthcare industry in an uproar. It’s understandable too.
For one thing, change can be hard. We get comfortable with a methodology or a set of tools, and changing our way of working can interrupt our rhythm, creating anxiety and diminishing productivity in the short term.
On top of that, ICD-10 promises to be a hefty change as the number of diagnosis codes increases from the current 13,000 in ICD-9 to roughly 68,000 –more than a 5X increase. It does this by expanding the length of the codes from three to four numbers in ICD-9 to seven to 8 alphanumeric characters in ICD-10.
Since there is little direct mapping between the two (only 24 percent of the codes map one-to-one, see the table below, from the American Medical Association Preparing for the ICD-10 Code Set: Fact Sheet 7) it means healthcare providers have a lot of hard work and late nights ahead of them in terms of preparation, implementation, testing, and training staff. All within less than a year. With the recent Patient Protection and Affordable Care Act (PPACA) enrollment delays fresh on everyone’s mind, it’s no wonder providers are concerned.
Yet as someone who has spent a good deal of her career focused on medical terminology, I’m here to tell you it’s a good thing. To really comprehend why, though, it helps to understand how we got to where we are.
Evolution of the International Classification of Diseases (ICD) Coding System
The International Classification of Diseases (popularly referred to as ICD) was originally developed and published in 1900 by the World Health Organization (WHO) as a way to standardize healthcare reporting across countries. The goal was to be able to roll up data easily from countries around the world in order to look at macro trends in morbidity and mortality. Keep in mind this was nearly 100 years before computers became ubiquitous in healthcare, so standardization was critical if there was to be any chance of making sense out of so much international data.
Along the way, there have been multiple revisions to ICD. Early on, new codes were published roughly every 10 years; more recently, the interval has been closer to 20 years. In addition, each country has the option to create its own Clinical Modifications (CMs) in order to capture data not requested by WHO.
Here’s where it gets interesting. While ICD’s original purpose was to identify health trends, in the U.S., it has been co-opted to become a centerpiece of billing systems for payers. Most clinicians find that ICD-9 codes don’t provide sufficient granularity for them to pass along information to the next doctor, so they tend not to rely on them.
This is especially true of inpatient facilities, where physicians are more removed from billing procedures, though the coding & billing department relies on them. ICD-9 codes are used more in outpatient facilities, but even there they use other notes to convey medical information.
Healthcare payers, however, are very much on board with ICD. They insist these codes be used, and if anything is classified incorrectly or isn’t clear, they will kick the invoice back for clarification, delaying payment and disrupting cash flow.
The Change from ICD-9 to ICD-10
This is why healthcare providers should be embracing ICD-10, no matter how disruptive it may seem at first. As mentioned earlier, the number of codes in ICD-10 increases to around 68,000. That improvement will allow billing departments to be far more specific when submitting claims, increasing the number of first-pass claims that are accepted.
For example, suppose a patient comes to an emergency care facility with a burn on his right forearm. When the claim is submitted under ICD-9, the code indicates a burn on the forearm but doesn’t specify which one.
The following week, the patient returns with a burn on his left forearm. The facility submits another claim with the code for “burn on forearm.” The payer’s office sees this and questions whether this is a second submission for the first injury and sends the claim back for more explanation. This creates extra work (and overhead) on both sides, and payment is delayed until the clarification is submitted – and the payer is satisfied.
Under ICD-10, the billing department will be able to specify that the first injury was to the right forearm and the second to the left. Clearly, these are two different claims, which should result in prompt payment.
Another difference is the ability to specify whether a claim is for an initial visit or a subsequent visit for the same condition. Consider a patient who breaks her leg skiing. To a physician, and to ICD-9, a broken leg is a broken leg. It doesn’t matter whether this is the first or fourth visit to treat it. That information is very important to the payer, however, because (again) they don’t want to pay twice for the same visit. Under ICD-10, billing departments will be able to provide that explanation when the claim is submitted. This greater level of detail will also be captured in electronic health records (EHRs), providing a more complete picture of the patient’s overall history.
Healthcare Analytics that Improve Patient Care
Billing isn’t the only area where ICD-10 will have an impact, however. Going back to WHO’s original intention, classifying healthcare issues more precisely will help enhance the healthcare analytics used for population-based health initiatives as well.
Health Catalyst is already seeing this in the data we’re pulling into our healthcare enterprise data warehouses (EDWs) and advanced analytics applications. The information we’re gathering through ICD-10 can be much more specific and helpful, in line with the healthcare analytics adoption model, which will ultimately assist us in helping our clients improve patient care.
ICD-9 isn’t actively updated anymore, and ICD-10 is effectively frozen until 2015. This “freeze period” is essentially a concession to make it easier for organizations to make the changeover; you wouldn’t want to be trying to implement an entirely new system and updates to it at the same time. Once 2015 comes along, though, there will be a backlog of changes that will allow WHO and individual healthcare providers to spot new trends, identify new epidemic outbreaks, and view the latest in world healthcare from a broader perspective.
ICD-11 on the Horizon
While the move to ICD-10 is already a lot to grasp, it’s important to remember that ICD-10 is not the destination but merely a stop on the journey. WHO started work on ICD-11 in 2007 and plans to release it in 2015. It is expected that the U.S. will finish making its CMs around 2020, with adoption mandated sometime after that. The focus on ICD-11 is to get the codes to function better in an EHR environment, although as WHO reviews the CMs made in various countries, many of those will likely be incorporated as well.
The key date right now, however, is October 1, 2014. That’s when the change to ICD-10 becomes mandatory.
Rather than fighting or dreading it, healthcare providers need to embrace it. While the transition is certain to come with some pain, in the long run it has a lot to offer in terms of spotting trends, reducing overhead, and improving cash flow. By getting out ahead of it now, especially in terms of testing, healthcare organizations will be able to minimize the disruption and realize the benefits.
The American Medical Association (AMA) is a valuable resource for learning more about how to prepare for the shift to ICD-10 (expand the “AMA Educational Resources” tab near the bottom of the page).
Learn about the development of terminology standards or see my guide for knowing which terminology standard to use.
Now it’s your turn. How prepared are you for ICD-10? What are your concerns?