Abstracts

Rationale: Children with infantile spasms often require extensive diagnostic work-up to determine etiology, expensive medications for treatment, and hospitalization during the initiation of certain treatments. The common diagnostic studies and therapies have evolved over the last several decades. The purpose of this study was to determine trends of treatment for infantile spasms at children's hospitals, including cost and initial diagnostic workup. Methods: A retrospective cohort study was designed using the Pediatric Health Information System (PHIS), a database that contains inpatient, emergency department, ambulatory surgery and observation data from 43 not-for-profit, tertiary care pediatric hospitals in the USA. Patients were included if they were less than or equal to 2 years of age, admitted as an inpatient from 2004-2014, and had an ICD-9 code of infantile spasms (345.6). Data fields collected included: patient demographics, hospital length of stay, ICU admission, mortality, hospital admission cost, use of various diagnostic studies (e.g. lumbar puncture, brain MRI, EEG), and medications used for infantile spasms (antiepileptic drugs, corticotropin, and steroids). Cost data, as a ratio of cost to charges, were collected and adjusted to 2014 dollars. Results: A total of 6,183 patients met study criteria with an average age of 9 months at diagnosis. Three-quarters of all patients had an EEG, 57% had brain imaging, and 17% had a lumbar puncture performed. Medications were initiated during inpatient hospitalization in two-thirds with 33% starting corticotropin, 29% topiramate, and less than 10% of patients on an oral or IV steroid, zonisamide, or vigabatrin. Use of corticotropin decreased over time, whereas use of oral steroids trended upwards. Length of stay was a mean of 5.8 days with an overall adjusted cost of $18,348. Over time there was an increase in cost, from a mean of $12,534 in 2004 to $23,391 in 2014. This correlated with an increase in mean length of stay over time from a mean of 5.2 days in 2004 to 6.9 days in 2014. The multivariable model, clustered around hospitalization, demonstrated the following statistically significant (p < 0.05) associations with increased cost: length of stay and days of corticotropin therapy, whereas lumbar puncture and zonisamide demonstrated decreased cost. Conclusions: Variability exists in the diagnostic studies obtained as well as pharmacotherapy for infantile spasms, which may lead to a difference in cost. Although there has been an increase in cost over time, the length of hospitalization and use of corticotropin contribute significantly to this increase. Thus measures to decrease length of hospitalization or decrease the cost of corticotropin may help control costs associated with initial treatment of infantile spasms. Funding: N/a