Researchers develop mapping tool for ICD-10 analytics

One of the many changes that will occur along with the transition from ICD-9 to ICD-10 involves the data analytics within the health care industry. The study of coding will become much more complex for analysts, as translating ICD-9 codes to ICD-10 can be very time-consuming and complicated.  

Potential data analysis issues following ICD-10 transition
Andrew Boyd, professor of medical informatics at the University of Illinois-Chicago, published a report in the Journal of the American Medical Informatics Association to highlight and clarify a few of the data analysis problems that are likely to occur following the implementation of ICD-10 codes. The report suggests mapping tools and metrics that will enhance the comparison between ICD-9 and ICD-10 coding systems. 

"The most important finding that we have about the ICD-10 transition that people need to know is that the general equivalent mappings that the government has provided only map back to 70 percent of the ICD-9 codes," Boyd told EHR Intelligence. "Without looking at both directions at the same time, you will miss 30 percent of your historical data."

Boyd used the example of the Type 2 Diabetes Mellitus Adult Onset, Unspecified Type, Uncontrolled without Mention of Complication code – for adults with diabetes who do not experience complications or take medications – which does not get mapped backward within the mappings issued by the government. Although this code is used frequently across the country, no patients will map to the code when using the backward mapping. 

When looking at the historical data over the course of the last 30 years, it is necessary to use a tool to get up to 99 percent of coverage for that data. Boyd and his colleagues from the university conducted a study that developed mapping tools to provide guidance on these ambiguous and complicated coding translations. The tools would then reveal where reports or analyses may be challenging, or even impossible.

Boyd noted that it will be difficult to find a solution that reduces payment delays after the implementation of ICD-10 codings. However, Boyd and the researchers created an online mapping tool that explains why insurance companies could possibly deny a claim. It also shows what the mapping is with ICD-9 and ICD-10 codes both forward and backward in what Boyd describes as a comprehensive approach. 

Around 1 to 5 percent of the mappings are incorrect when they are based off of past studies that are not entirely accurate. Delay in payment could be prevented if practices visualize the mappings ahead of time when submitting their claims. They will get a clearer understanding of the reasoning behind an insurance company's potential refusal of a claim if it is using a clinically incorrect map to deny the claim.

New mapping tool impacts data analytics
There is also another feature of the tool that helps providers with the process of ensuring a successful ICD-10 transition. 

"One easy way is to take the top 25 or 50 codes they encounter on a daily or weekly practice, put them into the system, and either look at the visualization or the online table and you will get a rough past of all of the ICD-10 codes that are related to the diagnoses you frequently encounter," Boyd explained to EHR Intelligence. "If you focus on the [codes] you most frequently encounter, you will be able to look at the increased detail in ICD-10 that you need to document in clinical notes or for diagnosis."

One important aspect of the ICD-10 coding system is that there is much greater detail compared to ICD-9. This should assist providers in improving their diagnoses, ultimately enhancing health outcomes. The mapping tool will help physicians manage and organize patient health outcomes through its ability to compare the patients seen while using ICD-9 to those they will see after the ICD-10 transition. This is significant, as there is currently no other way to get estimates from the previous year's data on how many cases of the flu or other conditions a hospital or practice will see.

The mapping tool makes it possible to compare ICD-10 codes to the 30-year historical data in ICD-9 or a three-year moving average, which will greatly improve patient outcomes and management. This also offers a much needed solution to the challenge of running reports that have always been run on a daily, weekly or monthly basis.

Although comparing the same data after the ICD-10 transition will not be possible, health care providers will be able to see more definitive analytics based on ICD-10 codes within a three-year timeframe. Most importantly, however, the tool will assist health care professionals in understanding the increased specificity of the ICD-10 coding system. As the majority of insurance companies are looking for as much specificity as possible, the tool ensures that providers have all the options of the codes associated with ICD-9.