Impact of an Artificial Intelligence-Based Solution on Clinicians’ Clinical Documentation Experience: Initial Findings Using Ambient Listening Technology

We administered a voluntary web-based survey to a convenience sample of 117 clinicians at Emory Healthcare, a large, urban integrated academic medical institution, at the time of pilot onboarding and 60 days after. The four survey questions addressed three domains: usability, clinician wellness, and patient experience. The follow-up survey contained additional questions on likelihood of recommending this technology, impact on productivity, and intentions for future use. Usability was assessed with a Likert scale with scores ranging from 1 to 5. Well-being and patient experience were assessed by comparing the proportion of negative responses. Statistical analysis was performed in Stata 14 (StataCorp, LLC, College Station, TX) comparing means of the pre- and post-intervention groups using paired-sample two-tailed t -tests, with an alpha of 0.05 determining significance. This study was not deemed to be human subjects research by the Emory University Institutional Review Board. The onboarding survey received 117 unique responses. The follow-up survey received 55 unique responses, with an overall response rate of 47%. In total, 31 participants completed both the onboarding and follow-up surveys with 58.1% identifying as male, 64.5% as white, 48.4% between the ages of 20 and 39, mean tenure as 8.4 years, and 32.3% being a primary care clinician (Table 1 ). For participants who completed both surveys, when asked if their current documentation process meets their requirements as a provider, 41.9% responded positively compared to 71% post-intervention ( p = 0.034). Concerning the ease of their current documentation process, 32.3% responded positively compared to 48.4% post-intervention ( p = 0.023). When asked about the impact of their current documentation process on their well-being, 71% responded negatively compared to 38.7% post-intervention ( p = 0.010). When asked about the impact of their current documentation process on the patient experience, 35.5% responded negatively compared to 6.5% post-intervention ( p = 0.005) (Fig. 1 ). Additionally, 35.5% of participants responded they would be highly likely to recommend this documentation solution to a colleague, 58.1% agreed that it increases their productivity, and 29.0% responded they intend to use it for the majority of their notes. Mean comparison of paired participant survey responses. “My current documentation process meets my requirements as a provider” and “my current documentation is easy for me to complete” were assessed with a Likert scale with scores ranging from 1 to 5. Well-being and patient experience were assessed by comparing the proportion of negative responses. Improvements in each area are indicated by an increase in the score. Statistically significant differences are noted with one asterisk for p < 0.05 and two asterisks for p < 0.01.