Validation of Pediatric Idiopathic Generalized Epilepsy Diagnoses from the Danish National Patient Register During 1994‒2019
Magnus Spangsberg Boesen1, Melita Cacic Hribljan2, Soren Christensen2, Amalie Klein-Petersen2, Sahla El Mahdaoui3, Malini Sagar1, Emilie Schou3, Anna Eltvedt1, Malene Børresen2, Maria Jose Miranda1, Alfred Peter Born2, Peter Uldall2, Lau Thygesen4
1Herlev Hospital, 2Rigshospitalet, 3Rigshospitalet Glostrup, 4National Institute of Public Health, University of Southern Denmark
Objective:
To validate whether pediatric idiopathic generalized epilepsy can be identified in Danish health registers using International Classification of Diseases, 10th Revision (ICD-10) codes combined with anti-seizure prescriptions (1994–2019).
Background:
The Danish National Patient Register is widely used for epidemiological research, but most validation studies have focused on “any epilepsy” (ICD-10: G40–G41). Whether specific idiopathic generalized epilepsy subtypes can be reliably identified remains uncertain.
Design/Methods:
We reviewed medical records of children coded with idiopathic generalized epilepsy (ICD-10) before age 18, and pediatric neurologists confirmed diagnoses according to International League Against Epilepsy criteria. We estimated the positive predictive value (PPV) and sensitivity of ICD-10 coding alone, as well as in combination with anti-seizure prescriptions and age at first code registration, using medical record–validated diagnoses as the gold standard.
Results:
Of 969 children coded with idiopathic generalized epilepsy, 431 were confirmed (115 childhood absence, 97 juvenile absence, 192 juvenile myoclonic, 27 generalized tonic-clonic seizures alone). The PPV of ICD-10 codes combined with an antiseizure prescription and age at registration was 44% for childhood and juvenile absence epilepsy. Adding ethosuximide prescription before age 8 years improved the PPV for childhood absence epilepsy to 59%, but sensitivity remained low (17%). For juvenile myoclonic epilepsy, the highest PPV was 68% (sensitivity 85%) using G40.3F plus antiseizure prescription and age at registration after 8 years. For generalized tonic-clonic seizures alone, the highest PPV was 31% using G40.3G from 2006–2019 plus antiseizure prescription and age at registration after 5 years.
Conclusions:
The Danish National Patient Register and the Danish Prescription Database are not reliable for identifying idiopathic generalized epilepsy subtypes in children, except juvenile myoclonic epilepsy, which can be identified with acceptable accuracy but should be interpreted with caution. Without chart review, register data can only identify the broad category of “any epilepsy” (ICD-10: G40–G41) rather than specific idiopathic generalized epilepsy subtypes.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.