Serendipity has driven many of psychiatry’s most important treatments, yet contemporary systems may undermine clinicians’ ability to notice and develop unexpected therapeutic effects. This selective narrative review synthesises landmark discovery stories, conceptual accounts of serendipity and contemporary case examples to clarify how chance observations become robust advances.

Across historical and modern examples, serendipitous discoveries consistently reflected the interaction of unexpected events with prepared observers working in supportive institutional and research systems. We identify current barriers created by standardised care, funding and trial structures, and professional fragmentation, and outline a multi-level framework for cultivating serendipity through phenomenological training, technology-enabled detection of anomalous responses, flexible funding and innovative designs such as adaptive platform and rapid-fail proof-of-concept trials.

Deliberately creating pathways that move rare, surprising responses from bedside observation to formal evaluation could accelerate more precise, personalised treatments while preserving rigor in psychiatric care.

This study investigated the prevalence of advance directives among patients receiving community care within the East London NHS Foundation Trust (ELFT), and to identify factors associated with their clinical application.

We analysed data from electronic health records of 4807 patients (aged 18–75 years), managed under the Care Programme Approach (CPA) in ELFT during 2021–2022. Demographic, clinical, service-level and patient-reported measures were analysed (binomial logistic regression).

A total of 31.2% of patients on the CPA had an advance directive. Black ethnicity, treatment in the forensic service or Newham, Luton and Bedfordshire localities, housing in socio-therapeutic facilities, diagnosis of personality disorder, ten or more previous admissions and engagement with DIALOG+ were positively associated with having an advance directive. DIALOG+ is the first approach that has been specifically developed to make routine patient–clinician meetings therapeutically effective. It is based on quality of life research, utilising the DIALOG scale (a patient-reported outcome measure), concepts of patient-centred communication, IT developments and components of solution-focused therapy, and is supported by an app.

This study highlights a complex interplay of cultural, social and systemic factors that influence advance care plan status. Structured communication, stronger therapeutic relationships and staff facilitation are likely to encourage advance care planning.

Involvement in clinical mental health research can be a challenge for services isolated from academic institutions, limiting opportunities for patients to receive innovative interventions and for clinicians to explore interest in research. We aimed to increase mental health research capacity in Somerset NHS Foundation Trust via a range of initiatives from collaboration between a senior clinician and research and development colleagues.

Over the course of the project, the number of participants recruited to National Institute for Health and Care Research-adopted mental health and dementia research projects quadrupled over a 2-year period, from 57 to 232, and the number of projects hosted rose from 9 to 23. A total of 165 clinicians signed up to receive information about ongoing studies.

We found considerable appetite for becoming involved in research among mental health clinicians, and were able to provide opportunities for research experience as well as access to innovative studies for local patients.

Artificial intelligence ambient voice technology (AI AVT), which uses a large language model to summarise clinical dialogue into electronic notes and GP letters, has emerged. We conducted a mixed-methods, pre–post (manual versus AVT-assisted documentation) service development pilot to evaluate its use in a child and adolescent out-patient clinic.

The median administration time per clinical encounter reduced from 27 min (manual) to 10 min (AVT) (P < 0.001). On average, AVT-assisted documentation required only 45% of the time for manual documentation (P < 0.001). Clinician-rated accuracy, quality and efficiency were significantly higher for AVT-assisted documentation. Patient acceptance was high, with 97% reporting that clinicians were not distracted by note-taking. Thematic analysis from focus groups identified positive effects derived from AVT (improved productivity and clinician well-being), but was balanced by barriers (technological limitations).

Integration of AVT into clinical workflows can significantly alleviate documentation burden, reduce cognitive strain and free up clinical capacity.

There is a pressing need for standardised and comprehensive research frameworks to evaluate both the therapeutic benefits and risks of dependency and misuse associated with medical cannabis. We address the issue of abuse potential or misuse liability by examining research data and what amounts to existing clinical dogma for validity and generalisability. We undertook a broad scoping approach to exploring recent literature, focusing on clinical studies that investigated abuse/misuse and dependence among users of medical and/or recreational cannabis and cannabinoids. The search provided over 350 articles, mostly (∼90%) dated post 2002. Abstract analysis narrowed the selection to under 50 papers, according to their relevance as judged by the authors, an experienced clinician and a public health toxicologist.

We identified and commented upon some broad assumptions within the abuse liability literature that included (a) a standard cannabis formulation; (b) standard routes of administration and standard potency dosing; (c) a standard pattern of use; (d) a standard user or patient; and (e) a standard vulnerability to misuse or dependence.

Unpacking and questioning these assumptions leads to the conclusion that far more rigorous language and research design are needed to address the question of cannabis abuse definitively but, based on the best available evidence, it appears that the abuse liability of medically supervised cannabis is comparable to any other class of widely used and well-regulated pharmaceutical agents.