Project Summary/Abstract
Red blood cell (RBC) transfusions are commonly administered in the pediatric intensive care unit (PICU) with
the goal of increasing oxygen delivery; however, storage of RBCs may limit efficacy to prevent or reverse
oxygen debt. RBC transfusions have been independently associated with increased PICU morbidity and
mortality and impart additional risk to patients due to infectious and non-infectious serious hazards of
transfusion. Evidence-based guidelines were developed to restrict transfusion to those patients most likely to
benefit. Liberal transfusion practices persist; however, placing nearly one quarter million children annually
admitted to the PICU at risk for harm from transfusions. Threshold-based guidelines exist for hemodynamically
stable patients with a hemoglobin (Hb) >7 g/dL where evidence is robust, whereas clinical judgement is
recommended when Hb is 5-7 g/dL or in hemodynamically unstable patients with a Hb > 7 g/dL.
Qualitative study of barriers and facilitators to implementation of the transfusion guidelines and intervention
mapping informed the development of a bundle of implementation strategies directed at addressing specific
contextual barriers to routine adoption of the transfusion guidelines in the PICU. The bundle of strategies,
informed by the Consolidated Framework for Implementation Research (CFIR) includes building consensus,
identifying, and empowering champions, educating providers, use of integrated computerized clinical decision
support (CCDS) tools, and providing quantitative metric-based feedback to providers.
The proposed pilot study aims to address two critical gaps in operationalization and evaluation of the
implementation bundle across institutions that are necessary prior to execution of a multi-center type 2 hybrid
effectiveness and implementation trial to formally evaluate the implementation bundle. These include the need
for 1) validated, electronic cohort characterization of patients at-risk for unnecessary RBC transfusion and 2)
optimization and integration of CCDS tools, which are anticipated to pose the greatest operational challenges
in a multi-site study. Thus, we propose the following specific aims:
Specific Aim 1. Create and validate a computable phenotype for a dynamic cohort of PICU patients at-risk for
unnecessary transfusion, including children eligible for threshold-based and/or clinical judgment-based CCDS
tools
Specific Aim 2. Deploy and assess implementation of Computerized Clinical Decision Support (CCDS) tools
as one component of the transfusion implementation bundle. Evaluation of the CCDS tools using a customized
version of the RE-AIM evaluative implementation framework for clinical informatics will facilitate further
optimization of the tools.