AI-Powered Documentation for Mental Health Providers: Retrospective Observational Mixed Methods Study

TL;DR

This study shows that AI-powered documentation tools like Smart Notes are widely accepted by mental health providers and help reduce their administrative workload without affecting note quality.

Contribution

The study presents the first observational evaluation of AI-powered documentation tools in mental health care, focusing on adoption, satisfaction, and productivity.

Findings

Smart Notes was used by 94% of full-time and 72% of contractual mental health providers weekly.

97.7% of full-time and 98.4% of contractual providers rated the note quality positively.

Mental health providers reported increased productivity and reduced administrative burden.

Abstract

Mental health providers (MHPs) face a significant administrative burden from documentation, which can contribute to burnout and reduce time available for direct patient care. Although artificial intelligence (AI)–powered scribes have shown promise in general medical settings, their utility has not been well explored in the specific context of mental health care. This study describes the development and preliminary observational evaluation of Smart Notes, a generative AI tool designed to assist MHPs with documentation on a commercial virtual mental health platform.

This study aimed to examine MHPs’ use and uptake of a generative AI documentation tool, including use patterns, perceptions of note quality, feedback and satisfaction, and changes in productivity.

Smart Notes was developed using a HIPAA (Health Insurance Portability and Accountability Act)-compliant Azure OpenAI infrastructure to securely generate session summaries from individual therapy sessions. The tool was rolled out to MHPs in a phased approach, and its use required MHP and client consent as well as a mandatory review and edit by the MHP. We conducted a 1-year retrospective observational evaluation of the feature, examining MHP use, MHP-rated note quality, MHP feedback and satisfaction, and MHP productivity (weekly working hours, session completion rate, and client caseload).

Over 1 year, 162 full-time and 1366 contractual MHPs used Smart Notes to generate over 286,000 clinical notes. Use of the feature was high and stable, with nearly all (averaging 94% each week, SD 1%) full-time MHPs and most (averaging 72% each week, SD 6%) contractual MHPs using it weekly for eligible individual therapy sessions after the full launch. MHP-rated note quality was overwhelmingly positive, with 97.7% (9980/10,219) of feedback ratings among full-time MHPs and 98.4% (19,977/20,300) of feedback ratings among contractual MHPs being a “thumbs-up.” Qualitative feedback from MHPs was also largely positive, praising the tool for saving time and easing administrative burden. Finally, both full-time and contractual MHPs demonstrated changes in productivity.

Our findings suggest that AI-powered documentation tools, such as Smart Notes, are a feasible and acceptable approach to supporting MHPs. The high adoption rate and favorable MHP feedback indicate the potential utility of these tools without compromising note quality. This study provides preliminary data regarding the application of AI documentation in the mental health care context, highlighting a promising path for future research and development in digital mental health.

Figures