Abstracts

Rationale: Rationale: Diagnostic codes of the International Classification of Diseases, 9th Revision (Clinical Modification: ICD-9-CM) for epilepsy were revised twice: in 1996, one category of convulsions (780.3X) was reclassified as seizures (780.39), and in 2006, some nonspecific seizures were reclassified as epilepsy (345.XX). We want to examine how the revisions for epilepsy and seizures affected hospitalization trends in subcategories of epilepsy, seizures, and associated primary procedures.Methods: Methods: We tracked the annual rate of hospitalizations (per 100,000 population) with primary subcategories of epilepsy or seizure and their associated primary procedures. We calculated proportions of primary to any-listed diagnoses to examine attribution of non-specific seizures to tractable (345.X0) or intractable (345.X1) epilepsy diagnoses.Results: Results: Hospitalization rates of most epilepsy diagnoses decreased before 1996. Thereafter, average rate in intractable and tractable epilepsy slightly increased until suddenly rising after 2006 for tractable epilepsy only. The proportion of primary to any-listed diagnoses was significantly higher and less sensitive to the reclassification in intractable (20-year average: 66.3% (95% CL (confidence interval) =64.9%-67.7%)) than tractable epilepsy (45.4% (95% CI=43.7%-47.0%)). The relative proportion of primary procedures was less sensitive to the reclassification in those with primary intractable epilepsy than in those with primary tractable epilepsy. Computerized assisted tomography head scans decreased, and video EEG monitoring (89.19) increased after 1996 till 2006 in those with tractable and intractable epilepsy. Use of electroencephalograms and spinal taps declined before 1996, stayed level between 1996 and 2006, and rose after 2006. Brain lobectomies (01.53) were not affected by either revision.Conclusions: Conclusion: Revisions of epilepsy/seizure diagnostic codes in 1996 and 2006 showed different effects on trends of hospitalization rate for epilepsy/seizure, and a greater effect on the proportion of primary to any-listed diagnoses and associated procedures in tractable than intractable epilepsy. Researchers should consider the effects when assessing epilepsy hospitalization trends.