Abstracts

Death in Children with Seizures and Epilepsy in the Pediatric Epilepsy Learning Healthcare System

Abstract number : 1.374
Submission category : 16. Epidemiology
Year : 2021
Submission ID : 1826464
Source : www.aesnet.org
Presentation date : 12/4/2021 12:00:00 PM
Published date : Nov 22, 2021, 06:54 AM

Authors :
Zachary Grinspan, MD - Weill Cornell Medicine; Erika Axeen - MD, University of Virginia; Anne Berg - MD, Lurie Children's Hospital; Sonam Bhalla, MD - Emory University; Jason Coryell, MD - Oregon Health and Sciences University; Scott Demarest, MD - Children's Hospital Colorado; William Gaillard, MD - Children's National; Howard Goodkin, MD - University of Virginia; Tobias Loddenkemper, MD - Children's Harvard; Belinda Marquis, MD - Weill Cornell Medicine; Lindsey Morgan, MD - Seattle Children's Hospital; Anup Patel, MD - Nationwide Children's Hospital; Scott Perry, MD - Cook Children's Hospital; Renee Shellhaas, MD - University of Michigan; Nilika Singhal, MD - University of California San Francisco; Dmitry Tchapyjnikov, MD - Montana Children's Specialist; Elissa Yozawitz, MD - Montefiore Medical Center

Rationale: Electronic health record (EHR) data include the date of death for some individuals, and thus may be a valuable source of information about mortality for children with seizures and epilepsy. The Pediatric Epilepsy Learning Healthcare System (PELHS) is a consortium of US pediatric epilepsy programs that collects and evaluates EHR data to improve outcomes for children with epilepsy. We report the demographic characteristics of children with seizures and epilepsy and a date of death.

Methods: PELHS extracts administrative data from the EHRs from centers with pediatric epilepsy specialists for individuals with a diagnosis code of seizures (ICD9 780.3x or 779.0; ICD10 P90 or R56.9) or epilepsy (ICD9 345.x or ICD10 G40.x). We provide descriptive statistics for individuals with a date of death in 2019 (“decedents”), including available EHR data from the 30 days prior to death. We focus on individuals with epilepsy with a date of death prior to age 25 years. We compared the demographics of decedents to the baseline population of children with seizures or epilepsy within each EHR extract, by site.

Results: The PELHS coordinating center includes data from 16 sites. Ten had EHR data extracts with dates of death. In 2019, there were 358 decedents with the following demographics: 166 (46%) female, 180 (50%) White, 92 (26%) Black, 57 (16%) Hispanic. Black race was overrepresented among decedents at one site (Figure, top panel). Hispanic ethnicity was not different among decedents at any site (data not shown). The age distribution was: median 7.4 years, range 2 days to 24.7 years, interquartile range 1.9 to 15.2 years. There were 15 (4%) neonates, 47 (13%) infants, 103 (29%) preschool age (1 - 5 years old), 69 (19%) school age (6 - 12 years), 91 (25%) teenagers (13 - 18 years), and 33 (9%) young adults (19 - 24 years). Decedents were significantly younger than the baseline population at four sites (Figure, bottom panel). 260 (73%) had at least one healthcare encounter in the 30 days prior to death, including 187 (52%) with an inpatient admission, 114 (32%) with an emergency department visit, 116 (32%) with an outpatient visit, and 248 (69%) at which medication was administered or prescribed.

Conclusions: Multisite EHR data can provide substantial information about death in children with seizures and epilepsy, across all pediatric age ranges, genders, races, and ethnicities. The over-representation of Black children among decedents is consistent with known healthcare disparities for Black Americans and merits further investigation, although the variation among centers suggests that center or regional differences may explain some of these disparities. The young age of decedents suggests interventions are needed to reduce mortality in early life. For three-quarters of decedents, there are healthcare encounters in the month before death, suggesting medical record review will yield information about the circumstances of death. In ongoing work, PELHS investigators are reviewing records for additional information about these deaths, including social determinants of health and potentially modifiable risk factors.

Funding: Please list any funding that was received in support of this abstract.: BAND Foundation
Pediatric Epilepsy Research Foundation; Weill Cornell Medicine.

Epidemiology