Risk of readmission for suicide attempt after epilepsy hospitalization

The objective of this study was to examine if epilepsy admissions are associated with a higher readmission risk for suicide attempt, independent of psychiatric comorbidity, compared with index admissions for other medical causes.

The Nationwide Readmissions Database is a nationally representative dataset containing data from roughly 15 million hospital discharges. Analysis of International Classification of Disease Clinical Modification 9 (ICD-9-CM) codes in the year 2013 revealed 58,278 index admissions for epilepsy; this group was compared with admissions for stroke (N=215,821) and common medical causes (N=973,078). Ninety-day readmission rates for suicide attempts were calculated. Cox regression tested for associations between admission type and suicide attempt readmissions up to 1year following index admission.

There were 402/100,000 readmissions for suicide attempt within 90days from index admission in the group with epilepsy; 43/100,000 in the stroke group; and between 37 and 89/100,000 in the medical group. Unadjusted hazard ratios (HR) for suicide readmissions within 1year in the group with epilepsy compared with the stroke group were 9.61 (95% confidence interval (CI): 7.69-11.90, p<2.0×10^-16) and 5.02 compared with the medical group (95% CI: 4.40-5.73, p<2.0×10^-16). The HR for readmission in the group with epilepsy, after adjustment for sociodemographic and psychiatric variables, were elevated at 4.91 compared with the stroke group (95% CI: 3.83-6.27, p<2.0×10^-16), and 2.66 compared with the medical group (95% CI: 2.32-3.05, p<2.0×10^-16).

Independent of psychiatric comorbidities, epilepsy admissions may be independently associated with more than a threefold increased risk of hospital readmission for suicide in the year following index admission in comparison with patients recently hospitalized because of stroke or other common medical disorders.