Back

Digital Mental Health Support Reduces Intended Use of Clinical Services: An Observational Study of the Mindstep Platform

Citron, E.; Selim, H.; Lin, A.

2025-06-23 public and global health
10.1101/2025.06.23.25330119 medRxiv
Show abstract

BackgroundThe demand on mental health services is rapidly increasing, outpacing the capacity of traditional treatment systems like the NHS. Digital interventions offer a scalable means of addressing this gap, but real-world evidence of their impact on service utilisation is limited. AimsThis study examines whether use of the Mindstep platform, a digital mental health tool offering assessments and video-based self-guided support, is associated with reduced intended use of clinical services and user-perceived usefulness. MethodsA retrospective observational study was conducted using user survey data (N = 254) from individuals with at least 30 days of Mindstep exposure. Participants reported past-year healthcare use, rated the perceived usefulness of Mindstep, and estimated their future clinical service use both with and without Mindstep access. The primary outcome was changes in intended use of public and private clinical services; the secondary outcome was perceived usefulness. ResultsMindstep use was associated with a 58.1% average reduction in intended clinical service use across all categories, with the largest reductions for private consultants and GPs. Users with low prior healthcare engagement reported the greatest reductions, while those with high prior usage rated Mindstep as most useful (mean usefulness score = 0.40 on a 0-1 scale). ConclusionsEngagement with the Mindstep platform was associated with substantial reductions in intended clinical service use and moderate perceived usefulness, particularly among underserved and high-need user groups. While intention may not directly translate to behaviour, these findings suggest that digital mental health platforms may play an important role in alleviating demand on healthcare systems and improving access to early support.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.