Clinical Notes for Just $24.99 /Month

Electronic health records (EHRs) have transformed clinical documentation but often at the expense of time, efficiency, and patient interaction. The rise of artificial intelligence promises potential alternatives to traditional charting, but prior studies have suffered from inconsistent standards, poor controls, and the use of immature technology. This study evaluated the performance of an ambient AI notetaking tool compared to manually written notes. Six physicians conducted interviews with three simulated patient profiles, producing 36 total notes, one manual and one AI generated, for each of three encounters. Notes were randomized, blinded, and reviewed by a panel of judges using an adapted eight question version of the Physician Data Quality Instrument (PDQI-9) alongside additional measures of error rate, harm likelihood, and severity of said harm. A total of 64 evaluations were returned. On average, AI-generated notes scored 7.1 points (24.5% higher) on the PDQI scale (36 vs. 28.9/40) and contained 13.5 times fewer errors (0.129 vs. 1.75/5). Additionally, errors in manual notes were rated three times more likely to cause harm, with the severity of said harm also rated 3.1 times higher. Physicians also rated AI notes as higher in quality and ease of use. Overall, ambient AI documentation produced notes superior in nearly all measured metrics, suggesting that advanced AI charting tools can improve documentation quality, safety, and experience.