Lung Ultrasound Implementation in the Management of Patients Hospitalized With COVID-19

Study overview:. The investigators will begin LUS implementation using strategies that target
barriers already identified by hospitalists in prior qualitative work. The investigators will
then proceed to conduct weekly rapid Plan-Do-Study-Act cycles informed by Rapid Iterative
RE-AIM, in order to ensure timely identification of implementation barriers. Rapid Iterative
RE- AIM is an innovative application of the RE-AIM framework in which screening assessments
of RE-AIM outcomes (e.g. level of reach, adoption, implementation) which is performed at
regular intervals during implementation and has been shown to be a feasible means of
assessing interval implementation. For outcomes not meeting target implementation at our
interval assessments, we will perform rapid qualitative interviews of key stakeholders to
identify current barriers. Using the contextual domains of PRISM and rapid qualitative data
from stakeholders we will select new or adapt existing implementation strategies,
facilitating robust implementation. Study outcomes will be measured using the RE-AIM outcomes
from PRISM including Reach, Effectiveness, Adoption, Implementation, and Maintenance.

Study Setting: University of Colorado Hospital is a 620 beds academic tertiary care center in
Denver, Colorado.

Participant Recruitment: Hospitalist participants will be recruited from among the faculty of
University of Colorado Hospital Division of hospital medicine which has over 100 physician
and advanced practice clinicians (APPs).

Study Procedures:

All hospitalist physicians and APPs will be offered training in LUS. The training module to
be used has been shown to be effective at training providers to competency. The investigators
will evaluate acceptability and feasibility of the training module using rapid qualitative
techniques and questionnaires of both completers and non-completers. Once a provider has
completed training, she may begin performing LUS on COVID patients. The images will be
wirelessly uploaded into PACS. The provider's LUS interpretation will be documented in the
EMR. Rapid Iterative RE-AIM will be performed on a weekly basis as a means of iteratively
screening for implementation barriers. The implementation outcomes of both fidelity and
adaptation will be prioritized although all aspects of implementation will be evaluated.
Interval Reach will be screened for using EMR data. For RE-AIM domains with poor
implementation, the investigators will use rapid qualitative methods to obtain stakeholder
perspectives of barriers and suggestions for adaptations to existing strategies or new
strategies.

Measures/Data Collection Study Outcomes:

Reach: The investigators will measure the percentage of patients hospitalized with a positive
COVID test result who received a LUS during the hospitalization. The investigators will
compare the demographics and characteristics of the eligible patients who received a LUS
versus those that did not in order to identify disparities in implementation that can be
addressed by adapting our implementation strategies. The investigators will explore the
underlying reasons for disparities in Reach using rapid qualitative interviews of providers
and other stakeholders as appropriate.

Effectiveness: Using a quasi-experimental interrupted time series design, the investigators
will compare the number of chest x-rays and CTs performed, PPE use and inpatient mortality in
patient hospitalized with COVID before implementation of LUS versus after.

Adoption:

Adoption by clinicians will be measured by calculating the percentage of clinicians who
completed the LUS training module within the UCH hospitalist group. The investigators will
also measure the percentage of clinicians who performed at least 1 LUS exam. The
investigators will evaluate for differences between providers who did and did not complete
training. The investigators will interview hospitalists who did not complete training or did
not perform LUS to understand why LUS was not adopted.

Implementation:

The investigators will focus on collecting implementation outcomes that are particularly
relevant to the evaluation of diagnostic tests. Fidelity will be measured using data
available through the EMR and through qualitative interviews of providers regarding their
clinical decision-making. EMR data will be gathered by random selection of at least 10% of
LUS exams within the EMR over the prior 30 days using the following 3 components to evaluate
fidelity. 1. Image quality: Through the EMR, the investigators will assess whether image
acquisition was adequate using a checklist of image requirements. 2. Image interpretation:
the investigators will review interpretation of POCUS by assessing ultrasound reports in the
EMR using a checklist 3. Clinical Decision-making using the LUS result: Assess appropriate
integration of POCUS findings into the clinical decision-making by review of the LUS report
in the EMR and daily clinician progress notes well as through qualitative interviews of
providers.

Feasibility/Acceptability:

The investigators will measure feasibility and acceptability of the 1) LUS training module 2)
use of LUS for evaluation of patients with COVID using a questionnaire as well as rapid
qualitative interviews of hospitalists and other stakeholders