The COVID-19 pandemic has been described as a prolonged societal trauma providing new understanding of long-term post-traumatic stress reactions, both generally and in specific at-risk populations.

The present study examined the longitudinal course of post-traumatic stress disorder (PTSD) symptoms within one of the most high-profile risk groups (i.e. healthcare staff).

The sample comprised 439 healthcare staff who completed the Northern Ireland longitudinal COVID-19 Staff Wellbeing Survey on a minimum of 3 out of 4 distribution time points. The survey was administered repeatedly over 4 years, spanning both peri- and post-pandemic periods (2020–2023), and contained the Impact of Event Scale-Revised, as well as bespoke items on COVID-19, demographics, occupational issues and support factors.

Three distinct classes emerged from a three-class, latent class growth analysis model. A ‘resilient’ group (74%) displayed symptoms that remained below cut-offs for clinically significant moderate–severe post-traumatic stress throughout the pandemic, whereas a ‘recovering’ group (23%) exhibited moderate–severe symptoms during the pandemic, which then decreased to subthreshold levels post-pandemic. A key at-risk group was the ‘chronic’ class (4%), which had moderate–severe post-traumatic stress symptoms peri-pandemic that continued to increase post-pandemic. Significant predictors of the ‘recovering’ and ‘chronic’ classes included perception of poor communication within the healthcare organisation; increased exposure to COVID-19 outside their work; and increased personal health risk factors for COVID-19.

Post-pandemic PTSD monitoring and support for healthcare staff may be warranted alongside the development of internal communication strategies within healthcare systems to protect staff and services going forward.

Patient satisfaction is an important indicator of mental healthcare quality. The Client Satisfaction Questionnaire-8 (CSQ-8) is widely used, but its psychometric performance and contextual relevance have not been examined in humanitarian settings.

To assess the reliability and validity of the CSQ-8 in a sample of displaced and conflict-affected Syrian mental health and psychosocial support (MHPSS) service users.

Structured telephone interviews were conducted with Syrian MHPSS service users in North-West Syria. Psychometric analysis assessed the internal consistency and construct validity of the CSQ-8. Convergent validity with therapeutic alliance was examined using the Working Alliance Inventory Short Revised. Item comprehension and contextual relevance were examined using open-ended qualitative responses.

Among 337 participants, the CSQ-8 demonstrated strong internal consistency (ω = 0.92), but suboptimal model fit (comparative fit index 0.922; Tucker–Lewis index 0.890; root mean square error of approximation 0.132). Convergent validity with therapeutic alliance confirmed the importance of service user–practitioner relationships to satisfaction. Item 4 (‘Would you recommend the service?’) and item 8 (‘Would you return to the service?’) showed large ceiling effects and weaker psychometric properties. Qualitative data indicated responses to these items were shaped by community stigma, low expectations and a lack of alternative services, rather than satisfaction. A secondary analysis of the CSQ-4 showed good reliability (ω = 0.81) and improved model fit (comparative fit index 0.993; Tucker–Lewis index 0.980; root mean square error of approximation 0.065).

In this humanitarian setting, items about a willingness to recommend and intent to return to a service may lack conceptual validity and should be used cautiously. Shorter forms such as the CSQ-4 may be a useful alternative.

Anxiety and depression are prevalent among medical students worldwide, especially in low- and middle-income countries like Pakistan, where mental health stigma remains a significant challenge. This study aimed to develop an explainable machine learning-based screening tool for anxiety and depression using non-stigmatizing lifestyle and sociodemographic factors in academic settings. We hypothesized that explainable machine learning models can use non-stigmatizing lifestyle and sociodemographic factors to screen for anxiety and depression in medical students.

A cross-sectional survey was conducted among 1,630 undergraduate medical students in Islamabad, Pakistan. Participation was voluntary; informed consent was taken and confidentiality assured. The study protocol was reviewed and approved by the institutional ethics committee (Approval No. 00009 IHSA/P\D-2022). Data collection was guided by extensive stakeholder engagement involving students, faculty, and mental health professionals to ensure contextual relevance and acceptability. Sociodemographic variables (e.g., age, gender, year of study, socioeconomic indicators) and lifestyle factors (e.g., sleep patterns, physical activity, academic workload, and social factors) were collected. Anxiety and depression were assessed using validated self-report instruments (GAD-7 and PHQ-9).

Following data cleaning, imputation, and encoding, the dataset was split into training (80%) and testing (20%) subsets. Separate Random Forest classification models were developed for anxiety and depression due to their capacity to model complex, non-linear relationships. Hyperparameters were optimized using cross-validation. Model performance was evaluated using accuracy, sensitivity, specificity, and AUC-ROC. Explainability was enhanced through SHapley Additive exPlanations (SHAP), enabling both global and class-specific interpretation of predictors.

The study reported that the prevalence of depression and anxiety was 57.8% and 46.4%, respectively. The anxiety model achieved 84.36% accuracy, while the depression model achieved 81% on the test dataset. SHAP analysis identified academic performance, sleep patterns, and physical activity as the strongest predictors of anxiety and depression, demonstrating non-linear and context-specific relationships. The anxiety model showed comparatively stronger performance, potentially reflecting differences in symptom structure and their associations with lifestyle variables.

We concluded that explainable machine learning models based on non-stigmatizing data can effectively support mental health screening among medical students. This approach offers a scalable, ethically transparent decision-support tool to inform targeted preventive interventions, such as sleep hygiene initiatives and workload management, advancing precision mental health in resource-constrained educational settings.

Israel’s 2015 mental health reform transferred responsibility for mental health services from the government to community health maintenance organisations (HMOs), aiming to improve accessibility and equity and reduce stigma. However, its implications for different populations remain significantly underresearched.

To examine trends in mental health service utilisation and psychiatric medication issue rates among Arab and Ultra-Orthodox Jewish populations before and after the 2015 mental health reform.

A retrospective cohort study using data from Israel’s four HMOs (2012–2019), including all individuals with mental/behavioural disorder diagnosis, psychiatric medication or a community-based mental health service encounter (N = 2 847 184). Participants were classified as either general (non-Ultra-Orthodox Jewish), Ultra-Orthodox Jewish or Arab. The study analysed annual community visits to mental health professionals and dispensed psychiatric prescriptions.

Between 2012 and 2019 the number of mental health service users increased across all populations, most prominently among Ultra-Orthodox Jewish and Arab populations (1.94- and 1.91-fold, respectively), compared with 1.55-fold in the general population. Annual community visits increased 2.4-fold in the general population, 2.2-fold in the Ultra-Orthodox Jewish population and 3.7-fold among Arabs, who had the lowest baseline utilisation. Psychiatric medication issue rates increased most in the Ultra-Orthodox Jewish population (2.5-fold), compared with 1.7-fold in the Arab and 1.5-fold in the general population. Women consistently used more services than men, with the greatest relative increase among Arab women.

The findings suggest improved access to mental health services over the study period spanning the 2015 mental health reform, particularly in previously underserved populations. Continued progress requires culturally sensitive, community-tailored interventions for Arab and Ultra-Orthodox Jewish populations.

Although all patient deaths affect clinicians, it remains unclear how the impact of suicides differs from other deaths. Is the trauma of losing a patient by suicide qualitatively distinct, or are the emotional, professional and organisational consequences of suicidal and non-suicidal deaths more similar than assumed?

To investigate the impact of patient suicide compared with other patient deaths on clinicians’ psychological well-being, clinical practice and career. To explore clinicians’ perspectives on how current support systems do, or do not, meet their needs.

A mixed-methods approach was used. An online survey with two subsets of questions (one for suicidal and one for non-suicidal patient deaths) was circulated to clinicians across South London and Maudsley NHS Foundation Trust.

A total of 122 responses were collected: two-thirds of respondents had experienced a patient suicide, with 53% reporting moderate and 12% reporting severe impact versus 36.6% reporting moderate and 4.2% severe for non-suicidal deaths. Non-suicidal death was associated with significantly lower impact (odds ratio 0.14, 95% CI [0.05, 0.41], p < 0.001) and less disruption to clinical practice. Blame emerged as a key factor shaping clinicians’ responses: 98% of respondents rated suicide as <60% predictable in secondary care, and 69% rated the ‘zero-suicide’ policy as unachievable.

Patient suicide has a heavier impact on clinicians, qualitatively distinct from other patient deaths. Blame shapes defensive responses in suicides, and internal questioning in non-suicidal deaths. The low-risk paradox and perceived unachievability of zero-suicide policies call for re-evaluation. Acknowledging predictability limits and clinicians’ support needs can help systems navigate the complex impact of patient suicides.

Recent decades have seen a steady increase in antidepressant prescribing, but little is known about prescribing trends during and following the COVID-19 pandemic.

This preregistered systematic review, following Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines, aimed to investigate antidepressant prescribing trends for adults in the UK and Republic of Ireland during and after the pandemic. It also compared prescriptions by drug and location.

We searched six databases: APA PsycInfo, CINAHL, MEDLINE, Scopus, medRxiv and Preprints.org. The review included primary research articles reporting trends in antidepressant prescriptions, including at least one time point after March 2020 in the UK and Republic of Ireland. This review has been preregistered on PROSPERO (ID: CRD42024498503).

We identified 7,320 studies, of which ten met the search criteria for the review. Studies were grouped on the basis of time period (2020: n = 5; 2021: n = 3; 2022: n = 2), location (England, Scotland, Northern Ireland, Republic of Ireland, UK) and drug type (serotonin–noradrenaline reuptake inhibitors, selective serotonin reuptake inhibitors, tricyclics, and others (e.g. monoamine oxidase inhibitors)). Most studies (eight of ten) demonstrated increased antidepressant prescribing over time. Two studies highlighted a decrease between March and May 2020. Demographic variables reflected higher rates of prescribing for women, and the modal group receiving antidepressants comprised middle-aged adults.

The commonly reported increase in antidepressant prescribing corroborates pre-pandemic trends and may suggest further, increased demands for mental health support to meet the unique challenges of the pandemic. Future research is required to evaluate the appropriateness of treatment decisions and to explore psychosocial factors that influence individual prescribing choices.

Students, due to their specific academic and psychosocial conditions, are at higher risk of suicide compared with the general population, and suicide is one of the leading causes of death among students worldwide.

To investigate the prevalence of suicidal ideation and suicide attempts among Iranian university students.

A systematic search was conducted in international and national databases, including Scopus, Web of Science, PsycINFO, PubMed and Magiran, up to February 2025. Title and abstract screening was performed by a single reviewer. Two reviewers independently undertook full-text screening (study selection) and data extraction. Data were analysed using Stata 16. The heterogeneity of studies was tested with Cochran’s Q and quantified with the I2 statistic. To explore the sources of heterogeneity, we performed sensitivity analyses and meta-regression. The protocol was registered in the International Registration of Systematic Reviews (no. CRD42023471340).

We included 28 studies in this research. The pooled prevalence of suicidal ideation, 12-month suicide attempts and lifetime suicide attempts among Iranian students was 17% (95% CI: 13–21%), 3% (95% CI: 2–4%) and 8% (95% CI: 6–10%), respectively, with substantial heterogeneity (I2 = 94.85, 91.16 and 93.46%, respectively).

This study highlights the high prevalence of suicidal ideation and suicide attempts among Iranian university students, underscoring the need for effective preventive strategies and further research.

Non-suicidal self-injury (NSSI) is a prevalent behavioural health concern during adolescence and has been consistently linked to emotionally invalidating interpersonal environments.

This study tested whether distress tolerance and negative urgency prospectively explained the link between invalidating environments and adolescent NSSI.

The study used a 2-wave longitudinal design with a 6-month interval; 300 Iranian adolescents (70% female, aged 13–17 years) participated.

Cross-lagged panel modelling showed that invalidating environments at wave 1 predicted higher NSSI at wave 2, when we controlled for baseline NSSI, and were associated with increases in negative urgency. Although both negative urgency and distress tolerance were related to later NSSI, only negative urgency significantly mediated this relationship; distress tolerance and the hypothesised serial pathway did not show significant associations. Sensitivity analyses indicated developmental differences across age groups.

The findings highlight impulsive response under distress as a central mechanism linking environmental invalidation to adolescent NSSI and identify negative urgency as a key intervention target.

Clozapine is the only antipsychotic with demonstrated efficacy in treatment-resistant schizophrenia, but remains underutilised worldwide, with limited data from the Middle East and North Africa.

We examined clozapine initiation, delays and outcomes in a large United Arab Emirates (UAE) cohort.

We conducted a retrospective cohort study of patients prescribed antipsychotics at a tertiary psychiatric hospital in the UAE (2018–2025). Treatment resistance was defined as two or more distinct sequential adequate antipsychotic trials, each for at least 6 weeks and at a drug-specific therapeutic dose (defined daily dose method). Outcomes were clozapine initiation, time to initiation, guideline concordance (within 3 months) and hospital admissions.

Of 6256 patients with schizophrenia spectrum diagnoses prescribed antipsychotics, 1738 (27.8%) met treatment-resistance criteria. Clozapine was initiated in 129 (7.4%; 95% CI 6.3–8.8%), comparably between expatriates and nationals (7.7 v. 7.1%; p = 0.72). Among 76 delayed initiators, median delay was 13.2 months (interquartile range 2.2–32.5) and 26.3% were guideline-concordant. The number of adequate trials most strongly predicted initiation (adjusted odds ratio 1.98, 95% CI 1.70–2.30; p < 0.001). Delayed initiation was associated with higher hospital admission rates (adjusted incidence rate ratios 2.5–2.6 for delays up to 24 months; p < 0.05), although an exploratory propensity-weighted analysis was underpowered (β = −2.57; p = 0.12).

Clozapine remains underutilised in this setting, with fewer than 1 in 12 treatment-resistant patients receiving it and only a quarter of delayed initiators achieving guideline-concordant timing. These findings highlight an evidence-to-practice gap and support strategies for earlier clozapine access.

The National Health Service in England has encouraged the development of specialised services for patients with high-intensity use of emergency departments: liaison psychiatry services in emergency departments and community services.

To evaluate the effect of different types of high-intensity use services on attendances at emergency departments, for those with 5 (HU5) or 10 (HU10) or more attendances in the previous 12 months, and the subset of this group with probable mental health problems, versus emergency departments without locality high-intensity use services.

A retrospective longitudinal study was conducted using nationally sourced data from 13 hospitals in England, spanning 1 April 2016 to 1 January 2020. An interrupted time-series analysis assessed the impact of three high-intensity use service models on weekly emergency department attendances: emergency department-based services with and without designated staffing, and community based services. Patients and clinicians contributed throughout the project.

A total of 2 938 510 emergency department attendances were analysed. Statistically significant associations were identified between the implementation of community-based services and reductions in weekly attendances by HU10 patients (rate ratio 0.83, 95% CI 0.71–0.97), and HU5 and HU10 patients with mental health-related attendances (HU5 rate ratio 0.82, 95% CI 0.75–0.91; HU10 rate ratio 0.70, 95% CI 0.59–0.84). Emergency department-based services with designated staff were associated with a reduction in mental health-related attendances among HU10 patients (rate ratio 0.88, 95% CI 0.80–0.98).

Services targeting patients with high-intensity use of emergency departments in England demonstrate promising potential in reducing emergency department attendances among this patient group. Community-based interventions show the most substantial impact, whereas hospital-based services often run by liaison psychiatry teams also contribute to reductions in those with high-intensity use with mental health issues. These findings support further development and expansion of tailored high-intensity use services across the country.

Recently, there has been a significant rise in potency (% tetrahydrocannabinol (THC)) of cannabis products globally. As such, there is a need for a better understanding of the relationship between cannabis potency and mental health outcomes, especially in a developmentally vulnerable population such as adolescents and young adults.

The objective of this scoping review was to summarise existing literature investigating the potency of cannabis products as it relates to mental health outcomes in adolescents and young adults aged 14–25.

Systematic searches of MEDLINE, Embase, CINAHL and PsycINFO were conducted for relevant manuscripts up to October 2025. Following PRISMA-ScR guidelines, retrieved studies were then screened and data extracted by two independent reviewers.

Out of 11 225 studies identified by our searches, 71 were included in the review after screening. Compared with low-potency cannabis, our findings suggest that high-potency cannabis is more strongly associated with severe mental health issues, such as cannabis dependence, psychosis and cognitive deficits.

Overall, it was found that high-potency cannabis use (>15% THC) was associated with a great number and magnitude of adverse mental health outcomes. As such, the potency of cannabis products should be measured in future cannabis research that investigates short- and long-term outcomes. Additionally, the potency of cannabis products should be a consideration in any future cannabis regulatory policy discussions.

The Mental Health Crisis Assessment Centre (MHCAS) at St Charles Hospital is a newly established acute psychiatric service within CNWL NHS Foundation Trust. Initial clinical and operational processes relied on parallel documents across multiple applications, highlighting the need for a safer and more efficient solution in a high-pressure clinical environment.

Aim 1 – Achieving Safe Digital Implementation: Successfully design and implement a safe, standardised digital handover tool using existing NHS IT infrastructure.

Aim 2 – Achieving Efficiency & Productivity: Demonstrate measurable improvements in clinical efficiency, workflow productivity, and multidisciplinary communication.

Aim 3 – Achieving Sustainability & Governance: Embed the digital handover tool into clinical governance processes to achieve sustainable “business as usual” practice.

The MHCAS Digital Handover Tool was developed using three applications within the NHS Microsoft 365 infrastructure: SharePoint Lists, Power Automate, and Power BI. Share Point Lists replaced three legacy handover documents with a single ‘master’ patient list serving as the central operational record. Power Automate automated routine administrative tasks. Power BI dashboards drew live data from the patient list to provide real-time visibility of operational activity, patient flow, and internal performance indicators, creating a unified digital handover and monitoring system.

Evaluation focused on system adoption, data quality, efficiency, and operational performance. Quantitative measures included weekly patient entries, number of active users, documentation update frequency, estimated administrative time savings, and referral-to-arrival times. These measures were continuously captured and visualised through live Power BI dashboards throughout the first quality improvement cycle. Qualitative feedback on staff experience, communication, and perceived value was collected at cycle completion.

Six-week outcomes (15/12/2025–25/01/2026) included:

• Weekly patient entries increased from 15 to 61.

• 27 active multidisciplinary users.

• Documentation updates increased from 10.5 to 12.6 per patient per day.

• Data completeness >98%.

• All legacy handover documents retired within one week.

• 30% reduction in administrative workload.

• Median referral-to-arrival time reduced from 15 to 6 hours (60% reduction) despite increased referrals.

• 100% (9/9) staff reported improved communication and reduced administrative burden.

The MHCAS Digital Handover Tool demonstrates that meaningful digital transformation in acute mental health services can be safely achieved using existing NHS infrastructure and governance frameworks. The intervention delivered measurable efficiency gains, improved operational oversight, enhanced clinical safety, and improved staff experience. Live operational data supports continuous quality improvement, while early stakeholder engagement facilitated rapid adoption. The approach is scalable and replicable, with discussions underway regarding application across inpatient, crisis, and liaison services. Beyond providing a more efficient and modern alternative to the current handover tool, the project highlights the largely untapped potential of automation already available and applicable to NHS systems.

Understanding the economic cost of self-harm is essential for evaluating intervention cost-effectiveness and guiding funding allocation and service planning.

To estimate the cost associated with self-harm presentations to hospital emergency departments and investigate key predictors of cost.

Data on presentations to hospital for self-harm in all Irish emergency departments were analysed for 2018 and 2019. Costs of hospital treatment following self-harm were identified (in 2019 euros) using top-down and bottom-up approaches. The perspective taken was that of the health service. Factors associated with costs were investigated using generalised linear models.

There were 25 053 self-harm presentations from 2018 to 2019. The average annual cost of self-harm was approximately €26.5 million; almost half of the total cost was due to repeat self-harm presentations (47.3%). The mean cost per presentation was €2117 (s.d. €1845), which incorporates acute hospital costs (mean €2067, s.d. €2127) and those of initial aftercare (mean €50, s.d. €69). Psychiatric and medical admissions were associated with highest costs, three times that of presentations resulting in emergency department discharge (incidence rate ratio (IRR) 3.01, 95% CI 2.72–3.36 and IRR 2.88, 95% CI 2.72–3.36, respectively). Other factors associated with higher costs included older age, emergency department medical assessment unit admission, receiving a psychosocial assessment and self-harm involving a firearm. Demographic and clinical predictors of cost varied according to care pathway.

Significant costs associated with repeat attendances and hospital admission provide evidence for investment in emergency department services providing comprehensive care for those presenting with self-harm, as well as in community-based mental health services.

People with schizophrenia spectrum disorders (SSD) experience high rates of type 2 diabetes (T2D), mainly due to antipsychotic medication side-effects and lifestyle factors (e.g. suboptimal nutrition and physical inactivity). Digital technologies may reduce T2D risk by complementing face-to-face and pharmacological treatments, through the provision of flexible and personalised psychoeducation and behavioural prompts tailored to end-users.

This study tested the preliminary efficacy of the Schizophrenia and diabetes Mobile-Assisted Remote Trainer (SMART), a co-designed text message-facilitated intervention, designed to reduce the risk and/or improve self-management of T2D, along with its acceptability and feasibility.

Using an uncontrolled pre–post design, 29 out-patients of an endocrinology mental health clinic and two community-based rehabilitation mental health facilities used SMART for 12 weeks. The primary outcome was patient activation, measured using the Patient Activation Measure. Secondary outcomes were combined objective cardiometabolic and self-reported health and mental health indicators. Pre–post changes were analysed with a linear mixed model, accounting for within-participant variation.

Significant improvements (p < 0.05) were detected in patient activation, confidence in diabetes self-management and general health management, health literacy and mental health recovery. High levels of acceptability and feasibility were confirmed, with recruitment, retention and adherence rates of 67.4, 92.9 and 93.0%, respectively.

SMART is a world-first digital intervention aimed at improving metabolic health in individuals with SSD. This study provides evidence of its preliminary efficacy in self-management of metabolic health while confirming its high acceptability and feasibility, supporting expansion towards a sufficiently powered controlled trial to assess its clinical effectiveness.

Memory deficits are among the key risk factors in mental health problems, and are associated with dysfunctional cognitive factors.

The present study explored the relations between dysfunctional attitudes, rumination and mind-wandering, and emotional and non-emotional memory, respectively, aiming to inform the mechanism underlying the negative association of these cognitive factors with memory.

A total of 123 undergraduate university students completed self-report measurements, including Dysfunctional Attitude Scale Form A, Rumination Responses Scale, Mind Wandering Frequency Scale and the Center for Epidemiological Studies Depression Scale. Additionally, they were invited to participate in an experiment consisting of two tasks (n = 111 completed the task for emotional memory and n = 110 for non-emotional memory). Pearson correlation analysis and a linear regression model examined the relationship between cognitive factors and memory performance.

Based on correlational analyses, rumination, dysfunctional attitudes and depression are negatively correlated with the accuracy of negative emotional face memory (emotional memory, all P<0.05). The regression models suggest that dysfunctional attitude significantly predicts the accuracy of negative face memory (β = –0.205, P = 0.037). Digit memory accuracy (non-emotional memory) was associated with rumination and mind-wandering (all P<0.05), in which rumination was the significant predictor of accuracy (β = –0.231, P = 0.021).

The study indicated that dysfunctional attitudes are related mostly to negative emotional memories, and that rumination is mostly associated with non-emotional memories. Dysfunctional attitude and rumination represent potential treatment targets for emotional and non-emotional memory deficits, respectively, related to psychopathology.

Medical students experience higher rates of mental health problems compared with the wider student population. Stigma and fitness-to-practise concerns can deter help-seeking, with students often turning to peers. Mental Health First Aid (MHFA) is a psycho-educational programme developed for the general public, designed to support recognition of mental health difficulties and responses to people in distress. We present an innovative, co-constructed mixed-methods study exploring medical students’ experiences of MHFA training,delivered as a student-selected component within a medical school curriculum. We aimed to explore the motivators, enablers and barriers to student engagement, and students’ perceptions of the value and relevance of MHFA in relation to mental health knowledge, self-care, help-seeking and supporting others. The study also sought to inform the development of a bespoke course tailored to the needs of medical students.

MHFA was delivered in four half-day face-to-face sessions to Year 1 and 2 medical students at Kent and Medway Medical School. Data were collected via post-session questionnaires, in addition to post-course student-led focus groups and optional semi-structured individual interviews. Questionnaire data were analysed using descriptive statistics, and qualitative data were analysed thematically.

Twenty-five students participated; 21 attended all four sessions (range was 21–25 students). In post-session questionnaires, students rated MHFA content positively in relation to self-care and supporting others, and noted areas of overlap with the medical curriculum, including psychosis, suicide and communication skills. Analysis of qualitative data from four focus groups (n=22) and nine individual interviews identified common themes. Students described generally positive experiences of MHFA training and valued having an adaptable framework for guiding conversations about mental health in professional and personal contexts. An additional theme related to suggestions for adapting the course to better suit the needs of medical students. These included a faster-paced format, case scenarios that more closely reflected medical student experiences, greater integration of skills practice, and increased focus on issues specific to medical students, such as disclosure of mental illness and fitness to practise, exam-related anxiety, and inclusion of doctors’ lived experience of mental illness.

MHFA training was generally experienced positively by medical students and perceived as relevant to self-care and supporting others. Students also identified limitations in how well the course aligned with the specific contexts and pressures of medical training. Together, these findings support the development of a bespoke mental health support course for medical students, building on MHFA approaches while addressing discipline-specific needs.

Understanding prescription trends and costs of psychotropic drugs is essential to inform mental healthcare policy and practice. However, comprehensive national evidence across psychotropic drug classes is limited, and little is known on how prescription volumes changed during the COVID-19 pandemic.

We examined prescription trends and costs of psychotropic drugs in the UK and assessed changes in prescription volumes following the onset of the COVID-19 pandemic.

We analysed national Prescription Cost Analysis data from 2013 to 2023, for six major psychotropic drug classes. Annual prescription volumes, total costs and average costs per drug were estimated. Linear regression was used to assess temporal trends and changes in prescription volume during the pandemic years. Costs were adjusted for inflation to 2023 prices.

Between 2013 and 2023, psychotropic prescriptions grew 4% (95% CI 3.9–4.2) annually. Prescriptions for antidepressants, antipsychotics, stimulants and anti-dementia drugs increased, but not for hypnotics and anxiolytics and drugs used in substance dependence. In contrast, total costs declined by approximately £328 million, and the average cost per drug dropped from £11 to £5.9. Prescription volume remained unchanged during the pandemic years, except for anti-dementia medications, which showed an annual reduction of 17.2% (95% CI 7.8–26.6) relative to the underlying temporal trend during the study period.

Our observed rising prescription trend is consistent with global evidence on availability and use of psychotropic drugs. Furthermore, we found a decrease both in total drug costs and average costs per drug.

Antipsychotic agents play a paradoxical role in obsessive–compulsive phenomena, augmenting selective serotonin reuptake inhibitors in treatment-resistant obsessive–compulsive disorder yet also inducing or worsening obsessive–compulsive symptoms, mainly in psychotic but also in mood disorders. How strongly individual agents are implicated has not been compared.

To map the distribution and density of evidence on antipsychotic-associated obsessive–compulsive symptoms across all agents, including those without induction evidence, and to synthesise causality indicators and management strategies.

A scoping review following Joanna Briggs Institute methodology and Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews. PubMed/MEDLINE and Scopus were searched from inception to December 2025, supplemented by two regulatory pharmacovigilance databases (United States FDA Adverse Event Reporting System and World Health Organization VigiBase). Prespecified operational definitions, a four-dimension triangulation framework and a standardised dose metric were applied. Registered on the Open Science Framework (DOI: 10.17605/OSF.IO/R8ZFA).

Eighty-eight primary studies were included. Clozapine carried the densest, most consistent evidence (de novo prevalence 5.9–44.3%; total up to 77.9%); in the one controlled cohort with a drug-free comparator, the number needed to harm for clozapine (at least 6 months) was approximately 4 (95% CI: 2–8). Olanzapine and risperidone were intermediate, and predominantly antidopaminergic agents showed minimal literature. Pharmacovigilance placed aripiprazole, not clozapine, highest, discordant with its sparse clinical literature, indicating that the gradient partly reflects research concentration. Genetic associations supported a serotonergic–glutamatergic framework.

Evidence for inducing obsessive–compulsive symptoms varies across antipsychotics, being most consistent for clozapine and least for predominantly dopamine-blocking agents. Brief routine screening and a pragmatic response pathway may be considered in clozapine services, with vigilance for other agents as evidence accrues.