Burnout, Well-being at Work
Conditions
Keywords
Ambient AI Scribes, Wellness, well-being outcomes, Burnout
Brief summary
The primary objective of the study is to investigate the impact of an ambient AI scribe on clinicians' wellness and well-being outcomes; additionally, the investigators will also explore how the use of the ambient AI scribe will lead to changes in documentation burden, clinical note characteristics and financial productivity.
Interventions
OTHERAmbient AI Scribe
Clinicians will use an ambient AI scribe as part of routine clinical care. The AI scribe captures the patient-clinician conversation, generates a draft clinical note, and supports documentation in the electronic health record. Clinicians receive training before beginning use of the AI scribe. The intervention is introduced in three sequential waves using a stepped-wedge design, with all participants eventually receiving the intervention.
Sponsors
Washington University School of Medicine
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
SINGLE (Caregiver)
Eligibility
Sex/Gender
ALL
Healthy volunteers
No
Inclusion criteria
* Must be a Clinician (attendings and advanced practice practitioners) who is part of the Emergency Medicine Department * Must be willing to use Ambient AI as apart of their clinical practice work
Exclusion criteria
-Residents who are apart of the Emergency Medicine Department
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Impact of an ambient AI scribe on clinician well-being and professional fulfillment | From enrollment to the end of maintenance phase at 24 weeks | A linear model will be used to describe the effect of Ambient tool introduction on our co-primary outcomes under the intention-to-treat framework with a random effects structure to describe within provider variability. While high or complete survey completion rates is expected, in the event that not all surveys are completed and returned, a primary analysis on completed surveys only will be performed, and perform sensitivity analyses accounting for potentially systematic survey non-response bias using a response weighting strategy, using provider, scheduling, and patient encounter characteristics to create survey response weights (within each survey time period), then reweighting observations to account for non-response patterns. In analyses for both co-primary outcomes, Wald type hypothesis tests for inferences on the overall Ambient treatment effect and compare p-values to 0.05 / 2 to conservatively account for multiple comparisons using Bonferroni's method will be performed. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Assessment of the longitudinal changes in documentation burden | From enrollment to the end of maintenance phase at 42 weeks | The investigators will fit Bayesian generalized linear models with appropriate link functions and likelihood choices (ie, log and Poisson for count data, logit and Bernoulli for binary responses) to for each encounter, while accounting for serial correlation overall through time (indexed to the day) using an autoregressive modeling structure, within provider patterns using a provider level random intercept, weekday, weekend, and holiday effects using fixed effects terms, and specify a flexible interrupted time series treatment effect using a semi-parametric, thin plate spline to describe how the introduction of Ambient AI affects the relationship between providers and notes, and how that relationship evolves over time. In sensitivity analyses, evaluation of how these effects and their evolution may differ by treatment wave will be assessed. All inferences will be based on describing the mean and 95% credible intervals for the treatment effect daily. |
Countries
United States
Contacts
PRINCIPAL_INVESTIGATORThomas Kannampallil, PhD
Washington University in Saint Louis
Outcome results
None listed