Quality Control Issues of an E-Code Medical Records Review

Background: In 2007, the percentage of hospital discharge records with e-codes was 58%. By December 2008, the percentage had increased to 61.54%. The purpose of this pilot study was to examine issues related to non-documentation of e-codes in Indiana. Methods: Five hospitals were selected to participate in the project. Medical records with an injury code between 800-989.9 without an e-code were obtained from the hospital medical informatics divisions. (N=1967) Stratum one included review of all medical records without an e-code with a principal diagnosis was between 800-989.9.(n=1167) Stratum two included review of randomly selected charts with a principal diagnosis outside the injury range, but another diagnosis between 800-989.9. Results: Seventy-three percent of the charts reviewed in stratum one were first-time injury admissions. Motor vehicle collisions and falls comprised the largest percentage (61%). Of the 193 charts reviewed during stratum two, 30 (15.5%) had a first time admission that was directly related to an injury. The most common reasons for lack of e-code were absence of documentation of injury mechanism by medical provider, or failure of the coder to assign an e-code. An additional 5.8% of the medical records contained e-codes in diagnosis fields 18 to 59 which were excluded from the hospital discharge data base. Conclusion: Several issues with e-coding were identified. Based upon the issues identified during the pilot study, the state department of health is developing a quality improvement plan.