Accuracy of institutional orthopedic trauma databases: a retrospective chart review

INTRODUCTION: Academic trauma institutions rely on fracture databases as research and quality control tools. Frequently, these databases are populated by trainees, but the completeness and accuracy of such databases has not yet been evaluated. The purpose of this study is to determine the capture rate of a resident-populated database in collecting extremity fractures and to determine the accuracy of assigned Orthopaedic Trauma Association (OTA) classifications. MATERIALS AND METHODS: A retrospective study was performed at a level 1 trauma center of all adult patients who underwent treatment for extremity fractures after an emergency department or inpatient consultation. A 20% random sample was taken from these entries and compared to a resident-populated fracture database designed to capture the same patients. For all matching records containing a resident-assigned OTA classification, relevant imaging was blindly reviewed by a trauma fellowship-trained orthopedic attending surgeon for fracture pattern classification. Resident OTA classifications were compared to this gold standard to determine overall accuracy rate. RESULTS: Three hundred eighteen (80%) out of 400 entries were captured by the resident-populated database. Two hundred thirty-one of these 318 entries contained an OTA classification. One hundred fifty-three (66%) of these 231 entries demonstrated concordance between resident and attending assigned OTA classifications. On subgroup analysis, 133 (70%) of the 190 lower extremity classifications were accurately identified as compared to just 20 (49%) of the 41 upper extremity classifications (p = 0.009). Seventy-nine (65%) of the 121 end segment fractures showed agreement versus 42 (67%) of the 63 diaphyseal injury patterns (p = 0.85). Accuracy of classification did not significantly vary by resident year of training (p = 0.142). CONCLUSION: Trainee generated databases at academic institutions may be subject to incomplete data entry and inaccurate fracture classifications. Quality control measures should be instituted to ensure accuracy in such databases if efforts are invested with the expectation of useful information.