A study from Vanderbilt University further suggests that mapping codes between ICD-9 and ICD-10 could be a problem.
The research, published in the Journal of the American Informatics Association, involves records from 100 patient visits to the Vanderbilt Adult Primary Care Clinic. Those records were manually coded in both ICD-9-CM and ICD-10-CM, then compared with general equivalence mappings (GEMs) and reimbursement mappings (RMs).
The Centers for Medicare & Medicaid Services and the Centers for Disease Control and Prevention created the GEMs and RMs to help organizations make the transition to ICD-10. They're designed for consistency in record-keeping both going forward and backward.
The analysis found that while the GEMs/RMs were consistent with manual encodings more than 80 percent of the time in both directions, a significant fraction of manual mappings were inconsistent with the GEMs and RMs. Of those mismatched mappings consisting of only one ICD-9 and one ICD-10 code, at least half did not match the GEMs due to subtle differences within the same hierarchical category, according to the research.
"These results should reinforce the notion that no true crosswalk exists between ICD-9 and ICD-10. Whether this consistency rate is sufficiently high to avoid significant changes in reimbursements or public health statistics is beyond the scope of this study," the authors say.
They say its vital for healthcare organizations to perform their own analysis with their most commonly used codes to determine how that will affect reimbursement and care statistics.
A study from the University of Illinois at Chicago suggested pediatricians could take a financial hit in the switch to ICD-10. It found that 26 percent of ICD-9 codes used in that specialty are convoluted in their mapping to ICD-10.
In addition, University of Illinois at Chicago researchers warned that the switch to ICD-10 poses the risk of under-reporting and over-reporting adverse events due to mismatches in codes for Patient Safety Indicators.
To learn more:
- check out the study