Accuracy of ICD-10-CM coding for physical child abuse in a paediatric level I trauma centre

Holly Hughes Garza1,2, Karen E Piper1, Amanda N Barczyk1

  • 1Trauma and Injury Research Center, Dell Children's Medical Center of Central Texas, Austin, Texas, USA.

Insights

International Classification of Diseases, Clinical Modification (ICD-10-CM) coding for child abuse is inaccurate. Surveillance using ICD-10-CM data alone may significantly underestimate physical child abuse occurrences.

Area of Science:

  • Medical Informatics
  • Pediatric Medicine
  • Forensic Medicine

Background:

  • Accurate coding of child abuse is crucial for public health surveillance and resource allocation.
  • The International Classification of Diseases, Clinical Modification (ICD-10-CM) is widely used for disease and injury coding.
  • The diagnostic accuracy of ICD-10-CM codes for physical child abuse has not been extensively evaluated.

Purpose of the Study:

  • To evaluate the accuracy of ICD-10-CM coding for physical child abuse.
  • To compare coding accuracy between inpatient and outpatient settings.
  • To determine the sensitivity, specificity, and predictive values of ICD-10-CM codes against a multidisciplinary child protection team's determination.

Main Methods:

  • Retrospective study of 312 patients evaluated for physical abuse between 2016-2017.
  • Used multidisciplinary child protection team (MCPT) abuse determination as the reference standard.
  • Calculated sensitivity, specificity, predictive values, and diagnostic odds ratios (OR) for ICD-10-CM codes, stratified by admission status.

Main Results:

  • Inpatient coding sensitivity was 55.6% and specificity was 78.6% (diagnostic OR=4.58).
  • Outpatient coding sensitivity was 22.2% and specificity was 86.3% (diagnostic OR=1.80).
  • ICD-10-CM coding demonstrated lower sensitivity in outpatients compared to inpatients.

Conclusions:

  • ICD-10-CM coding alone significantly underestimates the occurrence of physical child abuse.
  • Reliance on ICD-10-CM coded data for surveillance purposes may lead to underestimation of child abuse incidence.
  • Improved coding practices and supplementary data sources are needed for accurate child abuse surveillance.