Abstract
Objective: To evaluate the accuracy of ICD-10 codes for
placenta accreta spectrum (PAS). Design: Validation study.
Setting: Quaternary obstetric center in the U.S.
Population: Patients who had a delivery or dilation and
evacuation between October 2015 and March 2020. Methods: We
reviewed medical records of patients who were (1) assigned an ICD-10
code for PAS (O43.21-O43.23), (2) had a histopathologic diagnosis of
PAS, and/or (3) underwent a cesarean delivery with subsequent review of
placental pathology. Main Outcome Measures: Sensitivity,
specificity, positive predictive value, and negative predictive value of
the ICD-10 PAS code assignments. Results: Among 22,345
patients, 104 (0.46%) had an ICD-10 code for PAS and 51 (0.23%) had a
histopathologic diagnosis of PAS. ICD-10 codes for PAS had a sensitivity
of 88% (95% CI: 76-96%), specificity of 97% (95% CI: 91-99%),
positive predictive value of 43% (95% CI: 34-53%), and negative
predictive value of 100% (95% CI: 96-100%). The sensitivities of the
individual ICD-10 codes for placenta accreta, increta, and percreta were
72% (95% CI: 47-90%), 47% (95% CI: 21-73%), and 67% (95% CI:
41-87%), respectively. Based on chart review, primary reasons for code
misassignment included code assigned at referral for PAS ultrasound
evaluation (28%) and code assigned based on antenatal imaging alone
(27%). Conclusion: The sensitivity, specificity, and negative
predictive value of ICD-10 codes for PAS were overall high, but the
positive predictive value was moderate. These findings suggest that
ICD-10 codes may be useful for research and surveillance of PAS, but
opportunities may exist to improve accuracy.