City academic authors Nature Medicine article

A mental health academic from City, University of London has published an article in the leading journal Nature Medicine.

The commentary explores a new paper by Habicht et al. that was published in the same edition of the journal. The study found that an AI-enabled chatbot increased self-referrals to psychological therapies for common mental health disorders.

But despite the promising results, the author Dr Jacqueline Sin, Professor of Mental Health Nursing in the School of Health & Psychological Sciences at City, says that further research is needed to ensure that better access to talking therapies translates into quality treatment experiences and outcomes for everyone.

Using AI bots for referrals

According to the charity Mind, one in six people report experiencing a common mental health problem, such as anxiety and depression, in any given week in England. These common disorders can cause marked emotional distress and debilitate personal and social functioning.

To help address these conditions, psychological therapies, in particular cognitive behaviour therapy (CBT), can be used to effectively treat anxiety and depression. Indeed, according to Professor Sin, the UK NICE clinical guidelines recommend CBT, rather than pharmacological anti-depressants, as the initial and mainstay of treatment for mild and moderate cases.

In the Nature Medicine article, Professor Sin – a mental health nurse-researcher who has previously developed and evaluated digital interventions for family carers supporting individuals with psychosis (www.cope-support.org) – discusses a new study in the same issue of the journal that has explored how a personalised AI-enabled chatbot named ‘Limbic’ has helped people refer themselves for talking therapies.

She writes that in a multi-site study including nearly 130,000 patients, Limbic increased self-referrals by 15% when compared to the pre-implementation period, while standard webform-filling increased self-referrals by 6%. Patients who were assisted by the AI bot took less time to complete a self-referral and better recognition of their treatment needs. Most importantly, the enhanced self-referral rates were particularly pronounced for individuals from ethnic minority backgrounds (29% increase) and those who identified as non-binary (179% increase), according to Professor Sin.

“Anxiety and depression are a major cause of disability and sickness absence from work. However, the very nature of these conditions – such as feeling low and growing withdrawn – means that many individuals do not reach out for treatment. For these people, being referred by GPs or other healthcare professionals – who often lack both time and skills to diagnose, let alone to refer onward – is a barrier to treatment,” said Professor Sin. “Early treatment leads to better prognosis and recovery, therefore self-referral to talking therapies can be a crucial first step, leading to a meaningful trajectory-change for the individual, their family and community. Assisting self-referrals to access evidence-based psychological treatment not only enables citizens to exercise their right to seek treatment, but the process of self-referral – including completion of self-reported symptom questionnaires – can also help individuals recognise their own treatment needs. This is one of the key mechanisms behind the Limbic effect.”

Improving access for minorities

The article also mentions Limbic’s notable success in improving diversity of access to ethnic minority and non-binary individuals, against the backdrop of long-standing ethnic disparities in mental health service access in the UK. But despite positive results, Professor Sin also urges caution:

Although these are positive results from Habicht et al., we are not reaching psychological therapy utopia yet; far from it. Although more than two-thirds of talking therapy referrals are online self-referrals, we must not forget those who cannot or would not do so. These include the most marginalised groups – people experiencing higher symptom severity, social and economic disadvantages, or lacking skills or means to use the internet. Despite Limbic’s success, the pre-existing health inequalities and inequities for those most in need persist.