JMIR Publications

ABSTRACT

Background:

Mental disorders are the leading cause of disability in young people (aged 10-24-years), and their incidence constitutes a major health crisis. Primary youth mental health services are struggling to keep up due to overwhelming demand, the complexity and severity of young people presenting for care and a shortage of qualified mental health professionals (MHPs). Artificial Intelligence (AI) tools have potential to facilitate necessary improvements to diagnosis, triage and care planning for young people with emerging mental disorders.

Objective:

The objective of the current study was to examine beliefs and attitudes underlaying MHP acceptance of AI tools in youth mental health services.

Methods:

Fifty-seven MHPs (Mage = 35.35, 54% Male) with experience working with youth populations (age 12 to 30) completed online surveys about the acceptability of two novel AI prototypes. Attitudes towards the use of AI in youth was also explored in one-hour semi-structured interviews. Fifteen MHPs also participated in online one-hour semi-structured interviews. Quantitative data was interpreted using descriptive statistics, and qualitative analysis followed the Thematic Analysis approach.

Results:

MHPs were more likely to agree than disagree that AI will improve youth mental health care overall (e.g. 64.4% somewhat or strongly agree that the field of mental health will improve with AI). Despite voicing concerns regarding data security and privacy, MHPs also acknowledged a need for AI to improve the “signal-to-noise ratio” in services and address delays to care for those with severe and complex problems. Such problems were seen as pervasive across the youth mental health system and emphasise the serious costs of delaying development and implementation of novel tools. All participating MHPs discussed the potential negative impacts of not adopting novel tools.

Conclusions:

MHP acceptance and uptake of novel AI tools in youth mental health services will be driven by a more complex cost-benefit analysis of both adopting and not adopting, rather than solely on their design. The costs of delay are clear and so researchers and MHPs have a shared imperative to develop useful and meaningful clinical tools, and to work jointly on integrating them into practice. These findings should inform the future design and implementation of such tools.