Effect of a generative artificial intelligence digital scribe on pediatric provider documentation time, cognitive...

To assess the effect of a digital scribe among pediatric providers on documentation time, cognitive load, burnout, and caregiver satisfaction.

Six-month pilot of digital scribe use in pediatric generalists and subspecialists between June 27, 2024 and February 22, 2025. We included all pilot providers with EHR record logs. Using pretest-postest and interrupted time series analysis (ITSA), we analyzed associations between digital scribe use and documentation time, NASA task load index, mini-Z burnout inventory, and caregiver satisfaction. We conducted thematic analysis of provider feedback.

The digital scribe was used to generate notes for 69.5% of encounters (31 931/45 914) across 84 providers during 1454 provider-weeks. Compared with before scribe adoption, there was a mean time savings of 2.8 (95% CI, 2.1-3.5) minutes in the EHR notes activity per appointment following digital scribe adoption, averaging 1.5 (95% CI, 1.1-1.8) hours per provider-week, totaling 2143 (95% CI, 1607-2678) hours during the pilot. ITSA demonstrated 20.9% (95% CI, 17.2-24.6) time savings with digital scribe adoption. In provider surveys, there were significant reductions in all task load components (all P < .001). Burnout decreased from 54.9% (28/51) to 33.3% (17/51), P = .01. Provider-specific caregiver's likelihood-to-recommend increased from 92.3% (IQR, 88.1%-95.5%) to 94.3% (IQR, 89.7%-100.0%), P = .01. Thematic analysis of provider feedback frequently identified decreased cognitive burden and time savings.

Digital scribe use was associated with decreased documentation time, task load, provider burnout, and improved caregiver satisfaction. Further study is needed to determine if pilot results are scalable.