PROJECT ABSTRACT Every year 2.4 million CT scans are performed to evaluate for pulmonary embolism
(PE) in Emergency Departments (ED) in the United States. Approximately one-third of studies are avoidable,
with each scan resulting in unnecessary exposure to radiation and increasing risk for iatrogenic lung cancer
and heart disease. The use of clinical prediction rules to assess pre-test probability before CT reduces testing
by 25% without any missed PEs. Several professional society guidelines recommend their use. However,
providers do not use these tools, or use them incorrectly, in up to 80% of patients. Clinical decision support
(CDS) that incorporates well-validated clinical prediction rules can improve the use of evidence-based
diagnostic testing strategies however impact is often limited by low provider adoption. Integrating insights from
behavioral economics, including nudges, into CDS design is a novel approach to improving adoption of CDS
and evidence-based practices that minimizes negative impacts on clinical workflow and cognitive load. Nudges
are defined as positive reinforcement and indirect suggestions which have a non-forced effect on decision
making. For example, ‘opt-out’ options for organ donation consent led to striking differences in enrollment. The
central hypothesis of this proposal is that a CDS tool incorporating a theory-informed “nudge” will address
identified behavioral barriers to use and improve provider adoption of guideline-concordant CT ordering for PE.
Our team developed, pilot tested, and demonstrated preliminary efficacy of a CDS tool + Nudge. The tool
supported PE risk stratification by leveraging a well-validated clinical prediction rule and included a nudge
designed to address identified barriers to adoption. Adoption rates for the CDS tool + Nudge were almost
double the adoption rates for the CDS tool alone. We propose to conduct a large, multisite, hybrid type III trial
to evaluate CDS tool + Nudge. The CDS tool’s innovative technological architecture enables its seamless
implementation across sites. The tool was developed on an electronic health record (EHR)-agnostic web-
based platform, designed for dissemination to work with any EHR using the open communication protocol Fast
Healthcare Interoperability Resources (FHIR). We will use a stepped-wedge, cluster randomized design across
three health systems and 10 EDs. Specific aims: 1) Examine the impact of the nudge implementation strategy
(CDS tool + Nudge) on adoption of guideline-concordant CT ordering, 2) Determine the impact of the nudge
implementation strategy on identified behavioral barriers to adoption of guideline-concordant CT ordering, and
3) Confirm the clinical effectiveness of guideline-concordant CT ordering across diverse patient populations.
The proposed work is significant in its efforts to enhance health by preventing iatrogenic lung cancer and heart
disease caused by overtesting. It is innovative in its use of state-of-the-art technological methods and
assessment of causal mechanisms of nudge in CDS. Achieving the project’s objectives will advance the
science of implementation in emergency medicine and the application of nudges to provider decision-making.
