Service disengagement is a major problem for “Early Intervention in Psychosis” (EIP). Understanding predictors of engagement is also crucial to increase effectiveness of mental health treatments, especially in young people with First Episode Psychosis (FEP). No Italian investigation on this topic has been reported in the literature to date. The goal of this research was to assess service disengagement rate and predictors in an Italian sample of FEP subjects treated within an EIP program across a 2-year follow-up period.

The goal of this research was to assess service disengagement rate and predictors in an Italian sample of FEP subjects treated within an EIP program across a 2-year follow-up period.

All patients were young FEP help-seekers, aged 12–35 years, recruited within the “Parma Early Psychosis” (Pr-EP) program. At baseline, they completed the Positive And Negative Syndrome Scale (PANSS) and the Global Assessment of Functioning (GAF) scale. Univariate and multivariate Cox regression analyses were carried out.

489 FEP subjects were enrolled in this study. Across the follow-up, a 26 % prevalence rate of service disengagement was found. Particularly strong predictors of disengagement were living with parents, poor treatment adherence at entry and a low baseline PANSS “Disorganization” factor score.

More than a quarter of our FEP individuals disengaged the Pr-EP program during the first 2 years of intervention. A possible solution to reduce disengagement and to facilitate re-engagement of these young patients might be to offer the option of low-intensity monitoring and support, also via remote technology and telemental health care.

None Declared

Our patients often come to us through wait-lists, but when does the wait really start? Early intervention psychosis programs have introduced policies and benchmarks aimed at minimizing the time between referral and first contact. While programs that meet these targets are often celebrated for their success in reducing untreated psychosis, these targets leave out a critical piece of the puzzle: the time elapsed from symptom onset. Understanding how patients and caregivers experience the wait for care in its entirety is critical to further reducing treatment delays in psychosis.

This presentation aims to first examine a disconnect between early intervention wait-list policies and patients’ experiences. Next, it will explore how patient and family member narratives reflect the complex and nuanced nature of waiting for care. Finally, it will propose clinically relevant solutions for reducing the delay between symptom onset and appropriate treatment.

We conducted individual semi-structured interviews with patients and caregivers accessing early intervention psychosis services across Canada. For this presentation, three interviews from different geographic and socio-cultural regions were selected for their distinct perspectives. We performed a two-reviewer narrative inquiry to derive emergent narratives about waiting for services.

Patient and caregiver experiences revealed two distinct waiting periods. Aside from the “wait-list” period between referral and first contact that is addressed by early intervention policies, participants noted experiencing a much longer initial waiting period, with narratives beginning at symptom onset. Participants described this period as an active, dynamic, frustrating, and often traumatic process that involves multiple ER visits and attempts at receiving care.

We propose formally distinguishing between two forms of waiting for services: passive waiting, which is the state of being on a wait-list, and active waiting, which begins at symptom onset and includes the complex struggle to receive stable care. Early intervention programs’ efforts to reduce passive waiting are important, but the high burden of active waiting suggests a need for larger efforts such as clinician education and systemic changes in how patients access healthcare. Reducing active wait times could truly transform how first episode psychosis is managed and improve outcomes for those in urgent need.

None Declared

Adolescents are at increased risk of developing suicide-related behaviour (SRB). Varying contributing factors may play an important role across the different forms of SRB. Perfectionistic concerns may become a cognitive moderator influencing volitional moderators (e.g., non-suicidal self-harm, NSSH) and suicidal ideation escalation.

To identify profiles of perfectionistic concerns in a community sample of adolescents. Also, to study the relationship between suicide-related outcomes and NSSH, according to perfectionism profile.

A sample of 1,526 adolescents (54.3% female; M= 13.81 years, SD= 1.28) participated in our study. A wide range of SRB and motivational and volitional risk factors were evaluated in school settings. Six types of perfectionistic concerns, assessed by the Frost Multidimensional Perfectionism Scale (FMPS), were used to identify perfectionism profiles, through latent profile analysis.

Almost one in five adolescents (19.5%) showed SRB risk and more than one in three adolescents (35.1%) engaged in NSSH in the last year. Five profiles of perfectionism were identified (Figure 1). The profile featured by higher concerns across perfectionistic domains (7.2% of participants) showed significant relationship with SRB risk (OR = 2.84) and suicidal ideation (OR = 1.22), in comparison to the minimal concern profile. On the other hand, the profile featured by high parental concerns (18.2% of adolescents) was associated with increased risk of ideation (OR = 2.75) and NSSH (OR = 1.51).

Image 1:

Perfectionism may constitute a key risk factor for NSSH and SRB development. The promotion of prevention programmes to enhance cognitive regulation skills may help prevent suicide in adolescents.

None Declared

Suicide is a global public health issue. According to the World Health Organization (WHO), more than 700,000 people die by suicide worldwide each year. In 2019, it was the fourth leading cause of death among those aged 15 to 29 globally. In Spain, suicide has been the leading cause of external death in recent years, which has motivated in some regions the implementation of preventive strategies, such as Codi Risc de Suïcidi (Suicide Risk Code).

This study examines the epidemiology of suicidal behavior in the Catalan population between 2010 and 2019, exploring geographical disparities and the influence of different sociodemographic and clinical variables on the incidence of suicidal ideation and suicide attempt.

All residents in Catalonia who attended the public health system from 2010 to 2019 were included in the study. Data were obtained from the Program d’analítica de dades per a la recerca i la innovació en salut (PADRIS) of the Agència de Qualitat i Avaluació Sanitàries de Catalunya (AQuAS). Data on geographical, sociodemographic, and clinical variables were collected for subsequent statistical analysis. Statistical significance was set at p<0.05.

A total of 1,421,510 individuals were included. Overall, 6921 cases of suicidal ideation (0.5%) and 1143 suicide attempts (0.1%) were registered, which an accumulated incidence of 487/100,000 inhabitants for the first outcome and 804/1,000,000 inhabitants for the second. From the whole sample, 83,592 individuals (5.9%) had a severe mental illness (SMI), whereas the proportion of patients with a SMI in the group of suicide attempters was 9.6%. The majority of individuals who attempted suicide were women (64.1%). The highest proportion of individuals with suicidal thoughts or attempts was found in the age range from 15 to 19 years. The presence of a somatic illness, low socioeconomic status, tobacco use and the presence of a severe mental illness were significantly associated with suicidal ideation and attempt. Those patients with suicidal behavior had a higher number of emergency visits during the 10-year period, and also a higher probability of being prescribed antidepressants, antipsychotics, benzodiazepines or lithium. Geographical disparities in the incidence of suicidal behavior were found, with specific regions showing higher rates. In a logistic regression analysis, all associations remained significant.

This study highlights the association between sociodemographic and clinical factors, which should have been assessed in clinical practice, with suicidal behavior. Geographical disparities may be attributed to various factors, including access to mental health services and socioeconomic factors. This study underscores the importance of identifying specific profiles at a higher risk for committing suicide, and also to implement policies and prevention programs in order to address this public health issue comprehensively.

None Declared

Registry-based studies are efficient to investigate population mental health and suicide risk, but are largely absent in Europe outside of Scandinavia and the UK.

To investigate suicide risk associated with mental disorders in the Catalan population (7.6 million), stratified by gender and history of psychiatric hospitalization.

Population-representative retrospective registry-based cohort study including 764,938 Catalan residents in the period 2014-2019. Data sources included suicide mortality, electronic health registries from five healthcare settings, and administrative data. Suicide deaths were identified through judicial death registers using ICD-10 codes X60-X84. ICD-9CM and ICD10CM codes from all inpatient and outpatient healthcare contacts were used to categorize 109 mental disorders. Age-sex standardized mortality ratios (SMRs) were calculated using indirect standardization, with expected deaths based on official general population mortality rates in Catalonia.

Suicide risk was significantly elevated among Catalan residents diagnosed with any mental disorder (SMR [95%CI] = 1.6 [1.3-1.9] for females; SMR = 1.8 [1.6-2.0] for males). In females, suicide risk was highest for sedative or hypnotic abuse (SMR = 46.1 [3.7-88.5]), cocaine abuse (SMR = 42.8 [9.0-76.6]), borderline personality disorder (SMR = 33.0 [10.7-55.3]), polysubstance abuse (SMR = 32.9 [2.1-63.8]), and mental disorder not otherwise specified (SMR = 24.9 [11.7-38.1]). In males, risk was highest for obsessive-compulsive disorder (SMR = 20.2 [10.7-29.8]), acute and transient psychotic disorders (SMR = 17.9 [1.1-34.8]), mental disorder not otherwise specified (SMR = 17.2 [10.1-24.3]), paranoid schizophrenia (SMR = 16.8 [9.4-24.1]), and opioid abuse (SMR = 16.1 [1.6-30.6]). Suicide risk was substantially elevated in individuals with a history of psychiatric hospitalization (SMR = 18.3 [15.5-21.2]] for females; SMR = 13.4 [12.0-14.8] for males). In females with psychiatric hospitalization history, risk was highest for dependence on stimulants other than cocaine (SMR = 105.4 [11.6-199.3]), attention deficit hyperactivity disorder (SMR = 86.2 [19.2-153.2]), polysubstance abuse (SMR = 66.0 [33.6-98.3]), opioid abuse (SMR = 60.1 [2.1-118.1]), and cocaine abuse (SMR = 57.2 [32.6-81.8]). In males with psychiatric hospitalization history, risk was highest for obsessive-compulsive disorder (SMR = 45.7 [30.6-60.8]), schizoid personality disorder (SMR = 36.9 [13.5-60.2]), unspecified disorders of adult personality and behaviour (SMR = 35.0 [9.6-60.3]), schizotypal disorder (SMR = 34.3 [6.7-61.9]), and histrionic personality disorder (SMR = 32.5 [1.2-63.9]).

Risk of suicide in the Catalan population varies substantially by mental disorder type, gender, and psychiatric hospitalization history, highlighting the need for targeted and diversified prevention strategies.

P. Mortier Grant / Research support from: Miguel Servet CP21/00078 (ISCIII co-funded by the ESF+); project 202220-30-31 (Fundació la Marató de TV3); PI22/00107 (ISCIII, cofunded by the European Union); AC22/00006 (ISCIII and the European Union NextGenerationEU, Mecanismo Para la Recuperación y la Resiliencia), L. Latorre-Moreno: None Declared, F. Amigo: None Declared, M. López: None Declared, A. Portillo-Van Diest Grant / Research support from: PFIS FI23/00004 (ISCIII co-funded by the ESF+), L. Ballester: None Declared, J. Alonso Grant / Research support from: AGAUR 2021 SGR 00624; CB06/02/0046 (CIBERESP - ISCIII), G. Vilagut: None Declared

Suicidal behavior is a complex phenomenon that affects all demographics, with children and adolescents being particularly vulnerable. It is associated with multifactorial conditions that must be considered for the development of more effective prevention strategies. The use of machine learning (ML) models to predict suicide attempts is becoming widespread, as they allow for the simultaneous testing of numerous factors, their complex interactions, and non-linearity in predictive model creation. The Millennium Cohort Study (MCS) is an observational, multidisciplinary cohort study that encompasses a wide range of dimensions, including psychological, genetic, biological, familial, social, and economic factors, as well as traumatic life events, family history, and medical history. This allows for the exploration of their relationship with suicidal behavior throughout individual development using ML models.

The aim was to develop a statistical method that applies ML models to retrospectively predict suicide attempts using structured tabular data from an adolescent cohort defined by the MCS.

The sample consists of 9,824 MCS participants (age 17) who were asked if they had ever purposely hurt themselves in an attempt to end their life. Of these, only 7.4% (725) responded affirmatively. Before starting the modeling phase and fine-tuning any algorithm, several stages were completed: data cleaning, feature extraction and engineering, and feature scaling and selection. We used a wide range of algorithms, from low-complexity (linear regression) to high-complexity (neural networks), while tracking their effectiveness, robustness, generalization, sensitivity, and accuracy.

Even though overall accuracy ranged from 0.83 to 0.87, we generally obtained low f1-scores (˜0.45-0.55) for the targeted class (suicide attempt) and high f1-scores (˜0.95) for the control class. Similar results were observed for precision scores; however, the recall scores were good for both classes, ranging from 0.67 to 0.87. The best performing models were logistic regression and neural networks.

These preliminary results shows that ML models trained with multidimensional data from a young cohort are sensitive in classifying individuals who have attempted suicide. We aim to improve the f1-score and area under the curve (AUC) metrics for the target class through several techniques: over/under-sampling, target encoding, class weight adjustments, ensemble methods, and various neural network architectures.

None Declared

Deliberate self-harm (DSH) is a strong predictor of future suicide attempts (SAs), highlighting the need to evaluate interventions that reduce recurrence risk.

This systematic review and meta-analysis aimed to evaluate pharmacological and non-pharmacological interventions for preventing repeated DSH in individuals recently engaged in DSH. We also examined how participants’ characteristics and trial inclusion criteria impacted interventions’ effectiveness.

Randomized controlled trials (RCTs) with participants engaged in DSH within one month before enrollment were included. Studies with participants showing only suicidal ideation or older DSH were excluded. We searched five databases (PubMed, Embase, PsycINFO, WHO ICTRP, and ClinicalTrials.gov) with no language restrictions through September 2023. Two reviewers independently selected studies and resolved discrepancies through discussion. Outcomes were assessed at three time points: T1 (0–6 months), T2 (6–12 months), and T3 (>12 months). The primary outcome was repeat DSH; secondary outcomes included suicidal ideation, suicide deaths, all-cause mortality, anxiety, depression, hopelessness, and quality of life. Interventions were categorized by type and delivery format (individual or group, remote or in-person).

The PRISMA flowchart is shown in Image 1. Sixty-one trials involving 125 comparisons and 21,147 participants were included in this review. Average age was 31.5 years, with most being female (66.58%), unmarried (60.82%), and white (65.49%). Participants’ history of self-harm was noted in 48 studies, with at least 56% having engaged in another act before inclusion. In 42 studies, suicidal intent was either not investigated or not required. The most common types of interventions were Treatment As Usual (TAU) (46 arms), Psychotherapy (36), and Active Contact and Follow-up (25). Interventions were primarily applied individually and in person (84), with Cognitive Behavioral Therapy (16) being the most frequent psychotherapy framework. Psychotherapies were associated with a reduced risk of new self-harm acts within the first six months (T1) compared to TAU (k = 12; N = 2110; OR = 0.69; 95% CI 0.48–0.99). Active contact and follow-up interventions outperformed TAU in the longer-term follow-up (T3) (k = 4; N = 3724; OR = 0.83; 95% CI 0.70–0.99).

Image 1:

The superiority of psychotherapies at six months was not sustained at twelve months. However, active contact and follow-up interventions may offer benefits in reducing the risk of self-harm recurrence after one year.

None Declared

Social determinants of health (SDOH) have been linked to disparities in suicide rates across various demographics, including racial/ethnic groups, sex, age, and geography in the U.S. However, most studies have focused on individual or selected SDOH, rather than examining comprehensive, multi-dimensional SDOH factors. A more nuanced understanding of how clusters of SDOH contribute to suicide disparities across counties is needed to inform targeted prevention strategies.

To identify multi-dimensional SDOH county clusters and estimate their geographic and temporal associations with county-level suicide rates.

This study used national SDOH data from 3,109 U.S. counties over three time periods (2009, 2014, and 2019), matching them with county-level suicide rates from the National Vital Statistics System aggregated into three-year periods (2008-2010, 2013-2015, and 2018-2020). A total of 284 county-level SDOH variables, spanning six domains (social context, economic context, education, physical infrastructure, healthcare context, and natural environment), were analyzed using unsupervised machine learning algorithms to identify SDOH clusters. Associations between SDOH clusters and county-level suicide rates were estimated using negative binomial and LASSO regression.

Three distinct SDOH clusters were identified (Figure 1):

• • Cluster 1 (“REMOTE”) included rural counties with elderly, marginalized populations and substandard housing.

• • Cluster 2 (“COPE”) represented counties with complex family dynamics, overburdened health systems, poverty, and extreme heat challenges.

• • Cluster 3 (“DIVERSE”) encompassed densely populated areas with immigrants, racial/ethnic minorities, environmental challenges, and economic inequality.

Geographically, REMOTE was more common in North and Central U.S., COPE in the South and Central U.S., and DIVERSE along the coasts. Suicide rates were highest in REMOTE counties, especially among men. COPE counties had elevated suicide rates among Whites, while DIVERSE counties saw higher rates among women and Black/Hispanic populations. Most counties (70%) remained within the same cluster over time, with stable suicide rate associations.

This study identified three multi-dimensional SDOH clusters that were associated with varying suicide rates across U.S. counties. These clusters offer insight into the social and environmental conditions contributing to suicide risk. Future prevention strategies should focus on addressing the distinct challenges within each cluster, such as housing inadequacies, healthcare access, and economic inequality, to reduce overall suicide rates and related disparities.

None Declared

Previous results show that women who suffered from perinatal depression (PND) have a higher risk of suicidal behavior. However, clinical tools for predicting suicidal behavior among patients with PND are lacking.

To develop a prediction model for suicidal behavior after PND using machine learning.

Leveraging nationwide Swedish register data, we included 57,848 women with PND diagnosed during 2001-2018 and identified 2,303 events of suicidal behavior up to 5 years after PND diagnosis. Based on our previous association studies, 16 predictors containing information on demographics and pregnancy characteristics were included after multiple imputation. The sample was randomly split into 80% as a training set and 20% for testing. Classification and Regression Tree (CART), Random Forest (RF), Naïve Bayes (NB) and Logistic regression (LR) were used to establish prediction models with area under the curve (AUC) assessed for prediction performance. 10-fold cross-validation was used to evaluate the algorithms on unseen data.

In the prediction models, the CART yielded the best performance for suicidal behavior within 5 years after PND diagnosis (AUC 0.78, 95% CI 0.76-0.81). The population that scored highest (17%) in CART model had 84% risk of suicidal behavior. LR also had a comparable performance (AUC 0.78, 95% CI 0.76-0.81), whereas RF (AUC 0.74, 95% CI 0.72-0.76) and NB (AUC 0.70, 95% CI 0.68-0.72) had relatively poor performance. Notably, suicide history was a main contributor in all four models. Other predictors like household income, gestational age and education level were also important indicators of suicidal behavior risk.

The machine learning models have promising prediction performance for suicidal behavior after PND. Yet, further improvement is needed before clinical implementation.

None Declared

IMPORTANCE Changes in suicide rates after a nationwide trauma may be different from changes in psychiatric symptoms or general distress after such events. However, very few studies have examined short-term suicide-related reactions after such an event.

To evaluate the short-term outcome of the events in Israel on October 7, 2023, a large-scale terror attack and unfolding war, on changes in suicidality as reflected in percentages of suicide-related calls in relation to all calls to a national mental health first aid helpline, the Israeli Association for Emotional First Aid (ERAN).

DESIGN, SETTING, AND PARTICIPANTS The data included all interactions via the various ERAN helpline services between January 1, 2022, and December 31, 2023.

EXPOSURES The October 7, 2023, terror attack on Israel.

MEAN OUTCOMES AND MEASURES Changes in the numbers of overall calls and suicide-related calls to the ERAN helpline using an interrupted time-series analysis.

RESULTS. Analysis indicated that overall calls increased significantly on October 7. However, the number of suicide-related calls in the 3 months before October 7 was 1,887, whereas 1,663 suicide-related calls were registered in the 3 months after. The percentage of suicide-related calls decreased significantly on October 7 and gradually increased in the following period.

CONCLUSIONS AND RELEVANCE The findings suggest that although short-term emotional distress increased after national trauma, the percentage of suicide-related calls decreased. These results support previous studies suggesting that suicidality is not one of the immediate reactions to such traumas

None Declared

Suicide remains a major risk factor for individuals suffering from Schizophrenia. Recent studies have established that patients in the prodromal Schizophrenia state (i.e. Ultra High Risk for Psychosis) also experience significant rates of suicidal ideation and behaviour. However, less is known about the prevalence of suicidality among the adolescent and youth UHR population, a demographic particularly vulnerable to the psychosocial and environmental risk factors of psychosis.

This review aims to synthesise the existing literature on the prevalence of suicidal ideation and behaviour in the adolescent and youth at Ultra High Risk for Psychosis (UHR), and the relevant associations between suicidality and its correlates.

The databases PsycINFO, PubMed, Embase, Cochrane Library, Web of Science and Scopus were accessed up to July 2024. References within selected journals were further hand-searched for eligible articles. Keywords and controlled vocabulary used consisted of: (‘Ultra High Risk’ OR ‘At Risk Mental State’ OR ‘Clinical High Risk’) AND (‘Schizophrenia’ OR ‘Psychosis’) AND (‘Self-Harm’ OR ‘Suicide’ OR ‘NSSI’) AND (‘Adolescent’ OR ‘Youth’). Articles that included participants with an established diagnosis of schizophrenia or intellectual disability, history of frank psychotic episodes and extended use of antipsychotics were excluded. A meta-analysis of prevalence was subsequently performed for lifetime suicidal ideation, lifetime non-suicidal self-injury, lifetime suicidal attempt and current suicidal ideation. A narrative review was also carried out for the correlates of suicidality amongst the adolescent and youth UHR population.

Fourteen studies were included in this meta-analysis. Meta-analysis revealed a high prevalence of lifetime suicidal ideation (56%), lifetime non-suicidal self-injury (37%), lifetime suicidal attempt (25%) and current (2-week) suicidal ideation (58%). Narrative review revealed that personal transition to psychosis and positive family history of psychosis was strongly associated with suicidal attempt, while CAARMS (Comprehensive Assesment of At Risk Mental States) severity score was associated with suicidal ideation. Depression was strongly associated with both suicidal attempt and suicidal ideation, whereas positive symptoms and self-disturbance were associated with self-injury.

The prevalence of suicidal ideation and behaviour among UHR adolescent and youth is high and comparable to the general UHR population. Effective detection and management of suicide risk will be especially crucial in the adolescent UHR population. Existing measures that mitigate suicide risk in the general UHR population should be adopted for the youth and adolescent context.

None Declared

Due to the potentially severe side effects of antipsychotics, overtreatment is an important concern. Previous research focussed on antipsychotic polypharmacy and excessively high doses.

In this study, the aim is to map trends in potential overtreatment, antipsychotic polypharmacy, total antipsychotic dose and the subjective side effect burden. Moreover, the association of the total antipsychotic dose and antipsychotic polypharmacy with the subjective side effect burden will be investigated.

Data from a large (n>5000) naturalistic longitudinal cohort was used (PHAMOUS, 2013-2021). Potential overtreatment was defined as a total antipsychotic dose equivalent to >5mg risperidone, in combination with a high subjective side effect burden. Mixed effect models were used to investigate trends in potential overtreatment, antipsychotic polypharmacy, total antipsychotic dose and the subjective side effect burden. A mixed effect model was used to assess the association of total antipsychotic dose and antipsychotic polypharmacy with total subjective side effect burden.

Overall, 15,717 observations nested in 5,107 patients were used. About one-third of patients were potentially overtreated, with no change over time. The prevalence of a dose above the equivalent of 5 mg risperidone decreased over time, while antipsychotic polypharmacy prevalence increased. The total subjective side effect burden slightly decreased. A higher dose and antipsychotic polypharmacy was associated with a higher subjective side effect burden.

The subjective side effect burden did decrease the last decade. This might be caused by lower doses and more adequate use of polypharmacy. Still, the overtreatment rate is about one-third and the subjective side effect burden is still high. To reduce the subjective side effect burden and overtreatment, addressing inappropriate antipsychotic polypharmacy remains prudent.

None Declared

Functional outcome is a central clinical concern in inpatient psychiatry. Neurocognition is known to be an important factor in achieving a good functional outcome.

We have previously investigated whether cognitive dysfunction improves over the course of inpatient treatment, where acutely admitted patients are offered a combination of pharmacological treatment and cognitive remediation (Maihofer et al. J Clin Med 2024, 13, 4843). We now investigate the extent to which the functional outcome of patients with psychotic and affective disorders is associated with cognitive function over time.

Adult inpatients aged 18-66 years (female = 57.9%, male = 42.1%) were assessed with the Screen for Cognitive Impairment in Psychiatry (German version, SCIP-G: Sachs et al. Schiz Res Cogn 2021, 25, 100197; Sachs et al. Schiz Res Cogn 2022, 29, 100259). According to ICD-10 research criteria, 83 patients received an F2 diagnosis (schizophrenia, schizoaffective and delusional disorders), 61 patients met the criteria for bipolar disorder or mania (F30/F31) and 90 for depression (F32/F33). All patients received state-of-the-art pharmacotherapy and cognitive remediation using the COGPACK® software package version 6.06. Functioning was assessed using the Global Assessment of Function (GAF).

SCIP scores at baseline correlate significantly with SCIP scores at time point two (r=.74, p<.001). The SCIP at baseline is significantly correlated with patients’ functional level (r=.32, p=.01). The higher the baseline SCIP score, the higher the GAF score (r=.33, p=.01). The higher the GAF score at baseline, the higher the SCIP score at time 2 (r=.26, p=.039). The higher the SCIP score at time 2, the higher the GAF score at time 2 (r=.42, p<.001).

During their stay in hospital, acutely admitted patients improved in function and neurocognition, regardless of their diagnostic classification. Functionality as measured by the GAF correlates significantly with cognitive ability as assessed by the SCIP-G.

None Declared

Schizophrenia is a chronic mental disorder marked by positive symptoms such as hallucinations and delusions, and negative symptoms such as social withdrawal and apathy. While traditional pharmacological treatments effectively manage positive symptoms, they often fall short in addressing negative symptoms and social functioning. Yoga has emerged as a complementary therapy that may help improve both. However, its overall impact remains uncertain.

This review aims to synthesize evidence on yoga’s effectiveness in reducing positive and negative symptoms of schizophrenia and enhancing social functioning.

A systematic search was conducted in Scopus, Web of Science, and PsycINFO in September 2024, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Included studies were peer-reviewed randomized controlled trials (RCTs) assessing yoga’s effects on symptom severity and social functioning in schizophrenia. A fixed-effects or random-effects model was applied, with subgroup analyses performed. Standard mean differences (SMDs) and mean differences (MDs) were used for effect size estimation.

Sixteen RCTs were included, involving 862 participants. Yoga (n = 393) was compared to three control groups: treatment-as-usual (n = 152), other physical activities (n = 124), and waitlist (n = 193). Yoga significantly reduced overall symptom severity relative to control, as shown by a decrease in PANSS scores (MD = -6.61, 95% CI -13.21 to -0.0, p = 0.05, I² = 82%). It significantly improved positive symptoms relative to waitlist (SMD = -0.87 [-1.70, -0.03]) and treatment-as-usual (SMD = -0.65 [-1.21, -0.09]), with effects comparable to those observed with physical activity (MD = -1.30 [-3.09, 0.49]). There were no significant effects on negative symptoms, with SMDs of -1.59 [-4.18, 1.01] for the waitlist and -1.16 [-2.41, 0.09] for treatment-as-usual, and a MD of -1.32 [-3.60, 0.97] when compared to physical activity outcomes. Additionally, yoga did not significantly impact social functioning, showing SMDs of -0.44 [-1.56, 0.68] for the waitlist and -0.44 [-2.29, 1.41] for treatment-as-usual, and a MD of 2.30 [-0.74, 5.34] for physical activity.

This review shows that yoga is effective in reducing overall symptom severity compared to controls and improves positive symptoms relative to treatment-as-usual or waitlist, but not compared to physical activity. There were no significant differences in alleviating negative symptoms or enhancing social functioning. These findings suggest yoga may be a promising adjunctive treatment for positive symptoms of schizophrenia, especially when traditional treatments are insufficient. Further high-quality RCTs with standardized protocols are needed to confirm these results and establish optimal treatment parameters.

None Declared

Schizophrenia, affecting about 1% of the global population, is marked by positive, negative, and cognitive symptoms, leading to significant disability and a shortened lifespan due to physical health issues. Early intervention is crucial, as untreated psychosis can persist. This study examines changes in mRNA expression in the NRG-1/PI3K-AKT and DRD2 pathways in treated and untreated patients, hypothesizing that expression decreases and varies with treatment.

We aim to assess the impact of antipsychotic treatment on the expression levels of the PI3K/AKT/m-TOR signaling pathway, linked to the neurodevelopmental hypothesis of schizophrenia, and associated with the dopaminergic DRD-2 and glutamatergic NRG-1 systems. This study will examine mRNA expression of DRD-2, NRG-1, and PI3K/AKT/m-TOR in leukocytes from female schizophrenia patients before and after treatment. The findings may enhance our understanding of schizophrenia’s pathogenesis, identify potential genetic markers, and clarify the molecular effects of antipsychotic drugs.

Our study includes 25 healthy female volunteers and 25 female schizophrenia inpatients from Bakırkoy Prof. Dr. Mazhar Osman Mental Health and Neurological Diseases Training and Research Hospital. These inpatients met the experiment criteria, had not received treatment in the last 3 months, and were diagnosed with schizophrenia spectrum disorder according to DSM V criteria. After obtaining written consent, a sociodemographic form was completed. The Positive and Negative Symptom Scale (PANSS) was administered to eligible patients. Blood samples for DRD-2, NRG-1, and PI3K/AKT-1/mTOR mRNA expression analysis were collected from patients with PANSS scores of 95 or above before treatment and after scores decreased to 58 or below. Expression levels were determined by RT-PCR.

In untreated schizophrenia patients, NRG-1 mRNA expression was significantly lower, while PI3K and mTOR mRNA expressions were significantly higher compared to the healthy control group, with no change in AKT-1 mRNA expression as shown in Table 1. Compared to healthy controls, treated patients showed increased NRG-1 mRNA, unchanged PI3K and AKT-1 mRNA, and decreased mTOR mRNA as shown in Table 2. In treated patients, NRG-1 mRNA expression was significantly higher, and PI3K and mTOR mRNA expressions were significantly lower compared to untreated patients, with no change in AKT-1 as shown in Table 3. DRD-2 mRNA expression was undetectable in all groups.

Image 1:

Image 2:

Image 3:

NRG-1, PI3K, and mTOR may contribute to schizophrenia pathogenesis, with NRG-1 and mTOR serving as potential genetic biomarkers. Antipsychotics affect molecular pathways, but not AKT-1, and DRD-2 is not expressed in immune cells.

None Declared

Cognitive dysfunction has been identified as a key mediator of functional impairment in major depression and a contributing factor to antidepressant resistance. Theta burst stimulation (TBS) is a novel form of transcranial magnetic stimulation (TMS) that has shown greater efficacy and efficiency than conventional TMS in the treatment of treatment-resistant depression (TRD). However, its effects on cognitive symptoms in depression remain largely unstudied.

The objective of this study is to evaluate the impact of TBS on neurocognition in unipolar and bipolar TRD patients treated at a public hospital. Additionally, clinical, demographic, and treatment predictors of cognitive change were explored.

This is a follow-up study of n=64 patients with TRD (unipolar=48, bipolar=16) who received daily adjunctive TBS for 6 weeks. Cognitive performance was assessed before and after TBS using different versions of the Screening for Cognitive Impairment (SCIP-S), which measures immediate verbal learning, working memory, verbal fluency, delayed verbal learning, and processing speed. Cognitive performance in each domain was compared using paired t-tests. Global cognitive change was assessed by quantifying pathological domains at baseline and at the end of TBS. Differences in neurocognition between clinical responders (50% reduction on the HDRS) and non-responders to TBS were compared using ANOVA models. Additionally, possible predictors of cognitive change in each domain were explored using correlation analyses and multiple linear regressions, which included factors such as age, diagnosis, number of TBS sessions, treatment type, TBS modality (left unilateral vs. bilateral), clinical response, and baseline severity.

Patients treated with TBS achieved significant clinical response (68%) and cognitive improvement in the domains of immediate verbal learning, working memory, verbal fluency, and processing speed (mean differences of 2.16, 2.31, 1.30, and 1.05, respectively). Neurocognitive improvement was independent of clinical response. The percentage of improvement on the HDRS was only associated with improvement in the verbal fluency domain (p = .007). Left unilateral TBS and bipolar diagnosis predicted better global cognitive improvement in SCIP in regression models (p = .042, p = .037, respectively).

The results support the utility of TBS in treating cognitive dysfunction associated with TRD. Further larger studies are needed to clarify clinical and treatment predictors of cognitive improvement.

None Declared

Evidence regarding metabolic alterations associated with maternal antenatal depression (AD) is limited, and their role as potential biomarkers improving the prediction of AD and adverse child birth, neurodevelopmental, and mental health outcomes remains unexplored.

To identify metabolic measures associated with AD.

To test whether the metabolic measures associated with AD increased the amount of variance explained in AD over its risk factors.

To test whether the identified metabolic measures increased the amount of variance explained in child gestational age and weight at birth, developmental milestones at ages 2.3-5.7 years and any mental or behavioral disorder by the ages of 13.1-16.8 years over AD, sex, and age.

To replicate the findings in an independent cohort.

In a cohort of 331 mother-child dyads, we applied elastic net regression to study associations between AD (history of medical register diagnoses and/or Center of Epidemiological Studies Depression Scale score during pregnancy≥20) and 95 metabolic measures analyzed three times during pregnancy. Child birth and mental health outcomes were extracted from national registers and child neurodevelopmental outcomes were mother-reported.

Elastic net regression identified 15 metabolic measures that collectively explained 25% (p<0.0001) of variance in AD, including amino and fatty acids, glucose, inflammation, and lipids. These metabolic measures increased the variance explained in AD over its risk factors (32.3%,p<0.0001 vs. 12.6%,p=0.004), and in child gestational age (9.0%,p<0.0001 vs. 0.7%, p=0.34), birth weight(9.0%,p=0.03 vs. 0.7%, p=0.33), developmental milestones at the age of 2.3-5.7 years(21.0%,p=0.002 vs. 11.6%,p<0.001) and any mental or behavioral disorder by the age of 13.1-16.8 years(25.2%,p=0.03 vs. 5.0%,p=0.11) over AD, child sex and age. These findings replicated in the independent cohort.

AD is associated with alterations in 15 metabolic measures, which collectively improve the prediction of AD over its risk factors, and birth, neurodevelopmental and mental health outcomes of the child over AD. These metabolic measures may become biomarkers identifying at-risk mothers and children for personalized interventions.

None Declared

Major depressive disorder (MDD) is one of the most prevalent and disabling mental health conditions globally. Approximately one-third to half of MDD patients, suffer from difficult-to-treat depression (DTD), a condition marked by persistent symptoms that do not respond to multiple standard treatments. DTD is often associated with comorbid physical and psychiatric conditions leading to chronic disability and a reduced quality of life. The complexity of DTD poses a considerable challenge in clinical practice, underscoring the need for innovative treatment options.

Despite the demonstrated efficacy of esketamine in controlled clinical trials, real-world evidence is limited. This study aimed to address this gap by assessing the effectiveness of esketamine in routine clinical practice in DTD patients.

This prospective, naturalistic, open-label, observational study was conducted at Marqués de Valdecilla University Hospital in Cantabria, Spain. It included 33 patients diagnosed with DTD, comprising both unipolar and bipolar depression, as well as persistent depressive disorder (see table 1). Esketamine was administered intranasally in doses ranging from 56 mg to 84 mg, across three phases: induction, consolidation, and maintenance. Treatment effectiveness was measured using the MADRS, Clinical Anxiety Scale (CAS), and Clinical Global Impression (CGI). Functionality and quality of life were assessed with the Brief Functioning Scale (FAST) and the EQoL-5D (ESH). Assessments were conducted at baseline, 1-month, and 3-month. Data analysis was performed using SPSS v26, with repeated measures ANOVA and Pearson’s χ² tests employed to evaluate changes over time.

The final analysis included 33 patients with DTD and long-lasting current MDD episodes (table 2). Baseline MADRS scores indicated severe depression (39.12 ± 6). Significant reductions in MADRS scores were observed at both one month (21.61 ± 11.15) and 3-month (19.70 ± 11.65) compared to baseline. At 3-month, 54.5% of patients achieved a ≥50% reduction in MADRS scores, and 30.3% reached remission (MADRS < 12). Similar improvements were seen in anxiety (CAS) and health status (ESH) scores, with significant reductions noted over time. However, functional improvements (FAST) were not statistically significant (see table 3).

Esketamine demonstrated substantial effectiveness in reducing both depressive and anxiety symptoms in DTD patients over three months. More than half of the patients achieved a significant reduction in depression severity, with nearly a third reaching remission. The presence of late responders suggests that esketamine may benefit those initially unresponsive to treatment. These findings support esketamine as a valuable therapeutic option for DTD in real-world clinical settings.

None Declared

Major depression not only impairs individual health and labour market performance but also imposes significant economic burdens on society. It is linked to substantial costs through healthcare use, lost productivity and absenteeism. Recent genome-wide association studies have identified genetic markers associated with major depression, offering new insights into its genetic risk factors. However, the potential association of these genetic risks with educational attainment and career outcomes remains underexplored. Understanding this connection is crucial for addressing the broader public health and socio-economic implications of depression risk beyond clinical populations.

This study aims to investigate the relationship between genetic risk for depression and individual career performance in the general population of Finland from 1992 to 2017.

We utilised pooled data from the Finnish Finrisk (1992-2012) and FinHealth (2017) studies, which together include a population representative sample of individuals aged 25-64 (N=20,121). Genetic, survey and socio-economic registry data were integrated for this analysis. Using probit and semi-structural regression models, we examined various career performance indicators, with polygenic scores for depression (Howard et al. Nat Neurosci 2019; 22 343-352) as the main explanatory variable. Socio-demographic characteristics and genetic principal components were included as controls.

Our study revealed a negative association between higher genetic risk for depression and the likelihood of attaining higher education—an essential predictor of career success. Additionally, our study provides novel insights into how elevated polygenic risk for depression was linked to employment and self-employment rates, both directly and via educational pathways.

These findings highlight that genetic predispositions for depression can adversely affect career prospects in the general population, suggesting that the economic burden of depression extends beyond those clinically diagnosed. As effect sizes are modest, our results imply that supportive measures and compensatory behaviours could mitigate some of the educational and career disadvantages associated with higher genetic risk for depression.

A. Hazak: None Declared, J. Liuhanen: None Declared, K. Kantojärvi: None Declared, S. Sulkava: None Declared, T. Jääskeläinen: None Declared, V. Salomaa: None Declared, S. Koskinen: None Declared, M. Perola: None Declared, T. Paunio Consultant of: Idorsia Pharmaceuticals and Biogen (unrelated to the present work)

Major depressive episodes (MDEs) occur in mood disorders such as major depressive disorder (MDD) and bipolar disorder (BD), affecting nearly one in four U.S. adults over their lifetime. The neurotrophic hypothesis suggests that disruptions in growth factor signaling contribute to MDEs. While brain-derived neurotrophic factor (BDNF) is well-studied, vascular endothelial growth factor (VEGF) may also play a crucial role, though evidence of its association with MDEs is inconsistent. Understanding VEGF is important for identifying predictors of treatment outcomes, such as those related to electroconvulsive therapy (ECT). This study explores the relationship between VEGF and MDEs, focusing on implications for ECT effects.

To consolidate evidence from studies evaluating the association between VEGF and ECT outcomes in patients experiencing an MDE.

A systematic search for published clinical trials and cohort studies was conducted on August 13, 2024, using keywords including ECT, VEGF, MDE, and mood disorders, with no language or publication date restrictions. We selected studies enrolling patients in a current MDE related to MDD or BD, excluding those focused on manic episodes. A fixed-effects or random-effects model was applied. Subgroup analyses were performed to investigate the data further.

Seven studies involving 621 participants (61.9% female; mean age: 50.2 years) were preselected. Six studies measured plasma VEGF levels; one assessed cerebrospinal fluid (CSF) levels. Plasma VEGF levels did not differ significantly between healthy controls and MDE patients, either before (SMD = 0.02 [−0.17; 0.21], p = 0.84, I² = 0%) or after ECT (SMD = 0.11 [−0.21; 0.44], p = 0.50, I² = 0%). Of the five studies reporting post-ECT VEGF levels, three found a significant increase from baseline. A significant correlation was observed between baseline plasma VEGF levels and depression response to ECT (r = 0.34, Z = 4.92, p < 0.0001, I² = 0%). Of the five studies examining increased VEGF levels after ECT and symptom reduction, only one found a significant association. A sensitivity analysis indicated substantial heterogeneity when including the CSF study.

Plasma VEGF levels were not significantly different in MDE patients compared to healthy controls, either before or after ECT. Baseline plasma VEGF levels positively correlated with ECT treatment response, suggesting they may provide neurotrophic support and predict outcomes. Despite robust findings and minimal heterogeneity, this analysis was limited by the low number of studies and small sample sizes. Further research is needed to explore the association between MDEs and VEGF, especially in the CSF, and to clarify the role of baseline VEGF in ECT treatment response.

I. Borja De Oliveira: None Declared, G. A. M. Alves: None Declared, D. Soler Lopes: None Declared, A. de Vasconcelos: None Declared, A. Stephany: None Declared, F. Reis Soares: None Declared, M. Kraide Piedade de Abreu: None Declared, M. Carbajal Tamez: None Declared, J. Quevedo Shareolder of: Instituto de Neurociencias Dr. Joao Quevedo, Grant / Research support from: LivaNova; and receives copyrights from Artmed Editora, Artmed Panamericana, and Elsevier/Academic Press, Consultant of: EMS, Libbs, and Eurofarma, Speakers bureau of: Myriad Neuroscience and AbbVie., M. Teranishi: None Declared