INTRODUCTION:: We aimed to review the accuracy of current International Classification of Diseases and Related Health Problems (ICD) coding practices at our institution which affect the delivery complication rate. METHODS:: A retrospective review of vaginal and cesarean delivery complications with respective ICD-10 codes was performed (September 2017 to June 2018) at Robert Wood Johnson Barnabas Health. After review of 3 months of data, codes used erroneously were identified. Re-education of both coders and providers was performed. Subsequent coding rates were tracked for the remainder of the study period. RESULTS:: Based on the Joint Commissions list of reportable complications, 67 cesarean and 177 vaginal delivery complications were identified over a 10-month period. Ten cesarean and 26 vaginal delivery complications were found to be coded erroneously; recoding decreased complications by 14% for both. Between September and January, recoding rates for cesarean and vaginal delivery complications were 26% and 15% respectively. After re-education in January, rates dropped to 3% (cesarean) and 14% (vaginal). Interim review in April identified codes persistently being used inappropriately with respect to vaginal delivery complications, despite re-education. Restrictions requiring supervisory review and physician query were placed on codes in question, further decreasing vaginal delivery complications. We decreased our complication recoding percentage from 26% to 3% and 15% to 11% for cesarean and vaginal deliveries respectively. CONCLUSION:: Our findings suggest detailed chart review by medical providers in addition to re-education of coding staff and providers can improve accuracy rates for obstetrics coding.