Electronic Health Records

Policy & Regulation News

ICD-10 coders can regain significant productivity with practice

By Jennifer Bresnick

Medical coders working with ICD-10 for more than a year were able to gain back nearly a third of their lost productivity after switching to the new code set, revealed researchers from the University of Texas School of Biomedical Informatics (UT-SBMI) at Houston. While coders proficient in ICD-9 experienced a frightening 82% decrease in productivity after trying ICD-10 with little training, the same coders were able to mitigate that decrease by 35% after one year of practice with the significantly more complex coding system.

Susan Fenton, PhD, RHIA, FAHIMA, Associate Professor and Associate Dean for Academic Affairs at UT-SBMI, Mary Stanfill, MBI, RHIA, CCS, CCS-P, FAHIMA, and Kang Lin Hseih, a doctoral student working under Dr. Fenton, originally asked coders to complete ICD-9 and similar ICD-10 claims in 2012 with some ICD-10 education but a minimal amount of coding practice. The coders were able to complete an ICD-9 claim in 29.45 minutes and an ICD-10 claim in 53.5 minutes, an 82.67% decrease in productivity.

However, when the coders were asked to complete the same cases two years later in the first quarter of 2014, they were able to finish the assignment in 43.3 minutes, which was a 35% increase over their 2012 times. But the study also found that coders who had very high proficiency with ICD-9, likely due to memorizing many of the most common codes, struggled significantly with the changes in ICD-10 that rendered much of that knowledge and experience useless.

“We think it could be due to clinical documentation inadequacy, or it could be due to the complexity of the codes,” said Fenton, adding that even with practice and robust training, the coders were unlikely to regain their former speed. “You can’t have [every ICD-10 code] memorized. You’re just not going to be that fast. When we look at international studies in Canada and Australia, they never reached ICD-9 levels again. I think it’s reasonable to expect that we don’t fully rebound with our standard ways of coding. I stress that’s with our standard methods. As with other things, if we bring some additional technology to bear on it, we can see benefits.”

Agreement among the professionals on the right codes also increased from 2012 to 2014, and even improved from their agreement on the correct ICD-9 selections, the researchers said. “You can’t think about speed without thinking about accuracy. They’re two sides of the same coin. You absolutely need to balance them both, and we had to think about that in this study,” Stanfill said.

The team found that when coders collaborated on their work and were able to share ideas and knowledge, they improved more quickly and learned more. Coders who worked longer on complex ICD-10 problems without external help were more likely to have lower accuracy scores, Fenton said. “When are you going to say, ‘Stop spinning your wheels and go talk to someone’?’ It doesn’t increase their quality to sit there and be frustrated. They’re not making progress. They need to talk to someone else and get some help.”




Sign up to continue reading and gain Free Access to all our resources.

Sign up for our free newsletter and join 60,000 of your peers to stay up to date with tips and advice on:

EHR Optimization
EHR Interoperability

White Papers, Webcasts, Featured Articles and Exclusive Interviews

Our privacy policy

no, thanks

Continue to site...