How Reliable Is the G41 Discharge Code for Status Epilepticus?

Abstract

ABSTRACT Introduction Medico‐administrative databases are increasingly used to study the epidemiology of status epilepticus (SE), targeting hospitalizations with the SE G41 ICD‐10 code. However, the positive predictive value (PPV) of the G41 code, which measures the percentage of true cases among those identified by the code, is unknown. Methods We identified all hospitalizations with a primary or secondary diagnosis coded as G41 in five different hospitals. Medical reports for each hospitalization were reviewed to classify the stays as really related to SE or not, using two distinct approaches (sensitive and specific). The clinical characteristics of SE cases were also extracted. Results Among the 797 hospitalizations identified, the PPV ranged from 85.7% using the sensitive approach to 70.6% with the specific approach. Hospitalizations coded with G41 as the main diagnosis had the highest PPV, whereas codes G411 and G418 showed the lowest PPV. Of the 400 hospitalizations with a G410 (generalized convulsive SE) code, 72.7% were classified as generalized convulsive SE, while 76.5% of the 149 hospitalizations with a G412 (focal SE) code were classified as focal SE. Conclusion Our findings highlight that PPV varies by G41 subtype and diagnostic position. Studies requiring a higher PPV should exclude certain codes or hospitalizations with G41 code only as an associated diagnosis. Further studies are needed to estimate the sensitivity and specificity of G41 code.

Keywords

• medico‐administrative databases
• positive predictive value
• status epilepticus