Engagement phenotypes for a sample of 102,684 AI mental health chatbot (ash) users and dose-response associations with clinical outcomes

Conversational AI chatbots are emerging as scalable mental health tools, but little is known about real world engagement or its relationship to clinical outcomes. To characterize engagement phenotypes among users of Ash, a purpose-built AI mental health chatbot, and examine associations with clinical change and working alliance. K-means clustering across eight behavioral features identified engagement phenotypes among 102,684 users. Subsamples completed the PHQ-9 ( n = 298), GAD-7 ( n = 298), and MSPSS (social support; n = 194) at baseline and 3 weeks; 11,437 users completed baseline Working Alliance Inventory (WAI). Five engagement phenotypes emerged: Early Dropouts (52.2%), Power Users (1.6%), Intensive Users (4.1%), Weekly Users (25.3%), and a novel Concentrated User pattern (16.8%); across users, 66.9% had at least one overnight session (9pm-5am). Significant pre-post improvements occurred in depression ( d = -0.51), anxiety ( d = -0.57), and social support ( d = 0.22). An observed dose-response gradient in self-reported depression improvement was replicated in a larger sample using model-predicted PHQ-9 ( n = 23,813), with the largest improvements among high-engagement Power and Intensive Users ( d = -0.57 and -0.50) and the smallest among Early Dropouts ( d = -0.22). Higher working alliance predicted depression improvement and moderated the engagement-social support relationship. Engagement with AI mental health tools is multidimensional, and different clinical outcomes respond to different dimensions of use. Findings caution against treating session counts as a primary engagement metric and offer naturalistic, correlational evidence consistent with clinical benefit of purpose-built conversational AI, pending confirmation in controlled designs.

Authors

Institutions

Publication Details

Journal
Scientific Reports
Published
2026-11-15
DOI
https://doi.org/10.1038/s41598-026-78802-w
Citations
4
Primary Topic
Digital Mental Health Interventions
Type
article
Field-Weighted Citation Impact
11.93
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Engagement phenotypes for a sample of 102,684 AI mental health chatbot (ash) users and dose-response associations with clinical outcomes

Matteo Malgaroli, Emma Catherine Wolfe, Olivier Tieleman, Caitlin A. Stamatis et al.
4 citations
Scientific Reports
Digital Mental Health Interventions
11.93
article

Engagement phenotypes for a sample of 102,684 AI mental health chatbot (ash) users and dose-response associations with clinical outcomes

Matteo Malgaroli, Emma Catherine Wolfe, Olivier Tieleman, Caitlin A. Stamatis, Thomas D. Hull, Ting Su
article en
4 citations

Abstract

Conversational AI chatbots are emerging as scalable mental health tools, but little is known about real world engagement or its relationship to clinical outcomes. To characterize engagement phenotypes among users of Ash, a purpose-built AI mental health chatbot, and examine associations with clinical change and working alliance. K-means clustering across eight behavioral features identified engagement phenotypes among 102,684 users. Subsamples completed the PHQ-9 ( n = 298), GAD-7 ( n = 298), and MSPSS (social support; n = 194) at baseline and 3 weeks; 11,437 users completed baseline Working Alliance Inventory (WAI). Five engagement phenotypes emerged: Early Dropouts (52.2%), Power Users (1.6%), Intensive Users (4.1%), Weekly Users (25.3%), and a novel Concentrated User pattern (16.8%); across users, 66.9% had at least one overnight session (9pm-5am). Significant pre-post improvements occurred in depression ( d = -0.51), anxiety ( d = -0.57), and social support ( d = 0.22). An observed dose-response gradient in self-reported depression improvement was replicated in a larger sample using model-predicted PHQ-9 ( n = 23,813), with the largest improvements among high-engagement Power and Intensive Users ( d = -0.57 and -0.50) and the smallest among Early Dropouts ( d = -0.22). Higher working alliance predicted depression improvement and moderated the engagement-social support relationship. Engagement with AI mental health tools is multidimensional, and different clinical outcomes respond to different dimensions of use. Findings caution against treating session counts as a primary engagement metric and offer naturalistic, correlational evidence consistent with clinical benefit of purpose-built conversational AI, pending confirmation in controlled designs.

Scientific Reports
University of Virginia (US), New York University (US)
Good health and well-being
Openalex Percentile: Top 2%
Digital Mental Health Interventions
11.93
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon