Seizure Patients Often See Swift Return to Hospital

Nearly 11% of U.S. patients hospitalized with seizures were readmitted within 30 days — most commonly for convulsions or epilepsy — a retrospective cohort study showed.

While readmitted patients were more likely to have multiple medical comorbidities, inpatient adverse events also were significantly associated with 30-day readmission, reported Leah Blank, MD, MPH, of the University of Pennsylvania in Philadelphia, and co-authors in Neurology.

This study is the first to describe a nationally representative 30-day readmission rate for seizure discharge, Blank said. “This is important because hospital readmissions are increasingly being used to assess hospital — and, perhaps in the future, provider — performance,” she told MedPage Today.

Medicare payments are already docked for centers with poor performance on 30-day readmissions after an initial admission for a number of diagnoses, including heart attack, she added. “This list of penalized diagnoses is likely to expand and we think seizure is likely to be the first neurologic disease to be included.”

Readmission metrics are monitored as markers of quality and cost-effective care, noted Nathalie Jette, MD, of the Icahn School of Medicine at Mount Sinai in New York, and colleagues in an accompanying editorial.

“Current health care regulations allocate monetary incentives directed at enhancing quality of care that is also cost saving,” they wrote. “Although epilepsy is not currently within the list of diseases being penalized for readmission within 30 days by the Centers for Medicare & Medicaid Services, this will undoubtedly change as it is considered by many an ambulatory care sensitive condition.”

On a national population-based level, little was known regarding risk factors for readmission after seizure-related discharges, Jette and co-authors added.

In this study, Blank and her group studied emergently hospitalized adults with a primary discharge diagnosis of seizure or epilepsy, sampled from the Healthcare Cost and Utilization Project’s 2014 Nationwide Readmissions Database. They excluded elective hospitalizations, hospitalizations in which the patient left against medical advice, or hospitalizations occurring near the end of the year due to insufficient follow-up.

The researchers identified 139,800 hospitalizations that met inclusion criteria for 30-day readmission. Of these, 15,094 patients (10.8%) were readmitted within 30 days. The most common primary reasons for readmission were epilepsy or convulsions (17%) and sepsis (7%).

In multivariate logistic regression models, several variables were tied to 30-day readmission:

• Higher Elixhauser comorbidity score (OR 2.28 for 4+ vs ≤1 condition, 95% CI 2.01-2.58)
• Longer index admission stay (OR 1.78 for 7+ days, 95% CI 1.63-1.95)
• Public insurance (compared with private insurance: OR 1.48 for Medicare, 95% CI 1.34-1.62, or OR 1.39 for Medicaid 95% CI 1.26-1.54)
• Discharge to an inpatient care facility (OR 1.32 vs with discharge home, 95% CI 1.23-1.42)
• Documented adverse events during the index admission (OR 1.17, 95% CI 1.06-1.30)

When adverse events were analyzed in more detail, medication adverse events no longer were statistically significant (OR 1.12, 95% CI 0.99-1.26), but medical or surgical events were (OR 1.34, 95% CI 1.10-1.62).

“The most common reason for readmission was epilepsy or convulsion, which suggests that there is an opportunity for improvement as patients are not being readmitted for unforeseeable complications,” Blank said.

Adverse medical events need to be examined on granular level to better understand the health care structure — neurology ICU, continuous EEG, and neurology nursing, for example — and the processes and protocols they are associated with, she added: “Although documented adverse events were uncommon, readmitted patients were more likely to have had an adverse event during their original admission, which again suggests we can continue to improve the care that we deliver to seizure patients.”

Outpatient data was not available in this dataset, so the researchers could not study relationships between outpatient follow-up and readmission. They also could not assess epilepsy severity or information about social support, presence of a care partner, marital status, education, or anti-seizure medication adherence.