Validity of hospital discharge data for identifying infants with cardiac defects

J Perinatol. 2005 Nov;25(11):737-42. doi: 10.1038/sj.jp.7211382.

Abstract

Objective: To examine validity of the International Classification of Diseases, 9th Edition, Clinical Modification (ICD-9-CM) codes in discharge data for identifying infants with cardiac defects according to surveillance guidelines.

Study design: Retrospective medical record review of infants born in 2001 at one hospital in Minneapolis, Minnesota. Infants were identified using ICD-9-CM codes from hospital discharge data, and keywords in medical records.

Results: Of 2,697 children, ICD-9-CM codes identified 66 infants coded with cardiac defects; physician review confirmed 24 had cardiac defects. Only 35 of 85 (41.2%) ICD-9-CM codes accurately reflected the cardiac defect diagnoses. Additional case finding located four infants with five cardiac defects. Sensitivity of ICD-9-CM codes for identifying these infants was 0.857, predictive value positive was 0.364.

Conclusions: ICD-9-CM codes from hospital discharge data identified most infants with cardiac defects, but many were false positives. ICD-9-CM codes were inaccurate for specific cardiac defects.

Publication types

  • Research Support, U.S. Gov't, P.H.S.
  • Validation Study

MeSH terms

  • Birth Certificates
  • Heart Defects, Congenital / classification
  • Heart Defects, Congenital / diagnosis*
  • Heart Defects, Congenital / epidemiology*
  • Humans
  • Infant
  • International Classification of Diseases*
  • Medical Records / classification*
  • Medical Records / standards
  • Minnesota / epidemiology
  • Patient Discharge*
  • Pilot Projects
  • Population Surveillance / methods