• James Heckman, Kenneth J Mukamal, Adam Christensen, and Eileen E Reynolds.
• Division of General Internal Medicine and Primary Care, Beth Israel Deaconess Medical Center, East Campus, Shapiro Clinical Center, 330 Brookline Ave, Boston, MA, 02215, USA. jaheckma@bidmc.harvard.edu.
• J Gen Intern Med. 2020 Mar 1; 35 (3): 770-774.

BackgroundMedical scribes have been proposed as a solution to the problems of excessive documentation, work-life balance, and burnout facing general internists. However, their acceptability to patients and effects on provider experience have not been tested in a real-world model of effectiveness.ObjectiveTo measure the effect of medical scribes on patient satisfaction, provider satisfaction, and provider productivity.DesignQuasi-experimental difference-in-differences longitudinal design.ParticipantsFour attending physicians who worked with scribes, 9 control physicians who did not, and their patients in a large, hospital-affiliated academic general internal medicine practice.Main MeasuresProvider experience and patient experience using 5-point Likert scale surveys from the AMA Steps Forward Team Documentation Module, and visits and wRVUs per hour during 4 weeks before and 12 weeks after initiation of a practice model that included use of scribes and a shortened visit template.Key ResultsParticipating providers worked a total of 664 clinic sessions and returned 547 (82%) surveys. Average provider experience scores did not differ between providers working with scribes and control providers working without (4.01 vs. 3.40 respectively; p time-by-group interaction = 0.26). Providers with scribes were more likely to agree that work for the encounter would be completed during the visit then controls (3.58 vs. 2.48 respectively; p interaction = 0.04). A total of 6202 visits occurred during the study period. Average patient experience scores did not differ between the experimental and control groups (4.73 vs. 4.75 respectively; p interaction = 0.90). Compared with the control providers, providers with scribes completed more visits per hour (2.29 vs. 1.91; p interaction < 0.001) and generated more wRVUs per hour (3.42 vs. 3.27; p interaction < 0.001).ConclusionsIn this test of a modified practice model, scribes supported greater patient throughput and improved provider perceptions of documentation burden with no decrement in high patient satisfaction.

8 keywords...

hide…