Abstract 19611: Are Magnet® Hospitals Associated With Improved Outcomes in Children With Critical Illness?
Introduction: Little is known about the association of magnet recognition with outcomes among children with critical illness.
Hypothesis: We hypothesize that children treated at magnet hospitals will be associated with improved outcomes as compared to children treated at non-magnet hospitals.
Methods: Patients ≤ 18 years admitted to an intensive care unit (ICU) at a Pediatric Health Information Systems (PHIS) participating hospital were included (2004-2014). Patients with both cardiac and non-cardiac diagnoses were included. We used inverse probability of treatment weighting (IPTW) method to create a sample where treatment selection was independent of measured baseline characteristics. As a sensitivity analysis, we compared outcomes using multivariate models after adjusting for baseline covariates.
Results: A total of 454,616 patients (55.2%) of the patients were treated in 23 magnet hospitals, and 369,018 patients (44.8%) were treated in 19 non-magnet hospitals. Resource utilization (such as mechanical ventilation, nitric oxide, inotropes, and ECMO) was higher in magnet hospitals. After adjusting for patient and center characteristics, there was no difference in mortality between the two groups (Table). In adjusted models, we did not find any difference in hospital length of stay, duration of mechanical ventilation, or hospital charges. We found similar results among the cardiac and non-cardiac subgroups for all study outcomes by primary analysis using IPTW. The sensitivity analysis using multivariable models demonstrated similar results for all study outcomes, except for lower mortality in cardiac subgroup cared for in magnet hospitals.
Conclusions: This study calls into question the utility of magnet recognition program among children with critical illness. However, results from sensitivity analysis among cardiac patients leave some ambiguity in the interpretation of results.
Author Disclosures: K. Shah: None. M. Rettiganti: None. J. Gossett: None. P. Gupta: None.
- © 2016 by American Heart Association, Inc.