Closing the accessibility gap to mental health treatment with a conversational AI-enabled self-referral tool

Background

As awareness of mental health grows, more individuals seek support. However, significant barriers, including negative attitudes, stigma and structural issues, still limit timely access, especially for individuals from minority and disadvantaged backgrounds. Recognising and addressing these challenges at the initial help-seeking step is vital for creating an inclusive and fair mental healthcare system. To tackle this problem, we introduced our triage and clinical assessment assistant, Limbic Access, to bridge the accessibility gap. Our aim is to make mental health support more accessible to everyone, irrespective of their background.

Methods

We conducted a multi-site, real-world observational study of 129,400 patients across 28 NHS Talking Therapies services in England to evaluate the impact of Limbic Access on patient referral volume and diversity. Using publicly available NHS Digital data, we compared the referral data from a period before Limbic Access was implemented to a period when Limbic Access was used in services that used Limbic Access and matched control services to identify changes in the number and sociodemographic composition of referrals. In addition, we used Natural Language Processing (NLP) to analyse qualitative feedback about Limbic Access from 42,332 individuals to understand the mechanisms behind the increased access.

Results

The study results show that NHS services that used Limbic Access saw a 9 percentage point stronger increase in the number of referrals after Limbic Access was launched compared to matched control services where other forms of referrals were used in the same time period (15% increase vs 6% increase in control services).

The services that used Limbic Access saw a proportionally higher increase in the number of individuals from minority groups, such as non-binary individuals (179% increase) and ethnic minorities (29% increase), compared to the matched control services. For individuals from ethnic minority backgrounds, we especially saw an increase for Asian or Asian British groups (39%) and Black or Black British groups (40%).

‍

_179%
Increase in referrals from non-binary individuals

_29%
Increase in referrals from ethnic minorities

Qualitative feedback indicated that Limbic Access’s human-free nature and individuals’ self-realisation of their need for support played pivotal roles in enhancing the diversity of access, offering promising evidence for the potential of digital tools in reducing inequality in mental healthcare.

Conclusions

The results of the study show that digital tools and AI can effectively reduce the accessibility gap by addressing several key barriers. With 220,000 patients referred to NHS Talking Therapies services through Limbic Access, Limbic Access has allowed ~33,000 additional individuals to find their way into treatment highlighting that technological solutions can significantly contribute to overcoming inequality in mental healthcare, making support more accessible for all.