A Five Fold Decrease in Admissions for Uncomplicated Vaso-Occlusive Crisis and Other Benefits from Care in Infusion Clinics: Results from the Escaped Trial

• Published in: Blood 2018-11, Vol.132 (Supplement 1), p.853-853
• Main Authors: Lanzkron, Sophie, Little, Jane, Field, Joshua J., Haywood, Carlton, Varadhan, Ravi, Saheed, Mustapha, Proudford, Marc, Robertson, Derek, Kincaid, Adrienne, Burgess, Lorri, Green, Charles, Wang, Hang, Seufert, Rebecca, Brooks, Jasmine, Piehet, Allie, Griffin, Brandi, Arnold, Nicole, Frymark, Steven, Huang, Chiung- Yu, Wallace, Marcus, Abu Al Hamayel, Nebras, Segal, Jodi
• Format: Article
• Language: English
• ISSN: 0006-4971; 1528-0020
• DOI: 10.1182/blood-2018-99-111484

Background: Patients with sickle cell disease (SCD) suffer from frequent vaso-occlusive crises (VOC), the leading cause of hospitalization and ED visits for these patients. Infusion centers (IC) are alternatives to ED care and may provide patients with a better care experience. We previously demonstrated in a single center that the use of such centers provides more rapid pain control and can decrease the risk of a hospital admission. For the management of VOC the NHLBI's guidelines recommend that patients receive analgesic therapy within 60 minutes of registration and that pain be reassessed every 15-30 minutes until it is controlled. In the ESCAPED study, we aimed to learn whether the ED or IC more rapidly provides analgesia to patients who present with an uncomplicated VOC. Methods: The ESCAPED study is a prospective cohort study that recruited subjects at four sites (Baltimore, Cleveland, Milwaukee and Baton Rouge). The Baltimore and Milwaukee sites have dedicated ICs that treat only adults with SCD; the Cleveland and Baton Rouge sites have ICs that treat a diverse group of patients requiring infusion services. Patients were enrolled during regular outpatient visits between 4/2015 and 12/2016. Uncomplicated crisis was defined as an acute episode of pain with no known other cause and required treatment with parenteral pain medications. Upon study entry, participants completed surveys to collect demographic data; chart abstraction was done for information on comorbidities. Data was extracted after the acute visits for the primary endpoint: time to first dose of pain medication. We also extracted data on reassessment of pain after the first dose of parenteral pain medication and admission or discharge status. Patients' comorbidities were updated through chart abstraction after each visit. Each participant was followed for 18 months and data from visits for acute, uncomplicated VOC were collected. All acute visits were recorded and complete data was collected for each patient from the first visit each month to each site of care for uncomplicated crisis. Time-varying propensity scores estimated by a random effects model was used to balance covariates in the two arms. The treatment effects were estimated using a sub-classification approach and standard errors were computed by nonparametric bootstrapping at the individual level. Results: 483 subjects were enrolled and 444 completed 18 months of follow up (29 withdrew, 10 died). The median follow-up was 8.2 mon