Markku Lähteenvuo

Aims: This study aimed to investigate the incidence of low‐dose quetiapine initiation and associated factors in adults newly diagnosed with depression in primary care. Methods: We obtained medication dispensing records, primary and secondary care diagnoses and employment status for antipsychotic‐naïve individuals aged 16–65 newly diagnosed with depression in primary healthcare from several national Finnish registers in 2016–2022. We then calculated the incidence of low‐dose (≤50 mg) quetiapine initiation from 2 years before to 2 years after the depression diagnosis. Modified Poisson regression models derived adjusted risk ratios (aRRs) and 95% confidence intervals (CIs), assessing the association between covariates and low‐dose quetiapine initiation. Results: Of the 189 115 individuals with depression, 13.4% (25 373) initiated low‐dose. The incidence of low‐dose quetiapine peaked in the first 30 days following the depression diagnosis, constituting 14.8% of all initiations. Initiation had the strongest association with younger age (16–24 vs . 55–65 years: aRR 1.87, 95%CI 1.80–1.96) and psychiatric comorbidities, especially alcohol use disorder (1.44, 1.39–1.50) and other substance use disorders (1.27, 1.19–1.36). Ischaemic heart diseases had a positive association (1.13, 1.04–1.22), while hyperlipidaemia had a negative association (0.91, 0.86–0.97); other cardiometabolic conditions had no significant associations. Among other medications assessed, SSRI and hypnotic antidepressant use had the strongest associations with low‐dose quetiapine initiation (1.81, 1.77–1.86 and 1.81, 1.77–1.85, respectively). Conclusions: Low‐dose quetiapine initiation is common among adults newly diagnosed with depression in primary care. The benefits of this practice should be weighed against cardiometabolic and other adverse effects.

Background: Pharmacotherapies have shown limited efficacy in cannabis use disorder (CUD). This exploratory study examined the real-world effectiveness of commonly used psychiatric medications in reducing the risk of substance use and psychiatric hospitalisations among individuals with CUD. Methods: We conducted a nationwide prospective cohort study using Swedish health and administrative registers. The cohort included 51,947 individuals aged 16–64 years with a first CUD diagnosis (ICD-10 F12) between 2006 and 2021. Medication use was modelled as time-dependent with the PRE2DUP method. Within-individual Cox regression was used to compare hospitalisation risk during medication use versus non-use. The primary outcome was hospitalisation for substance use disorder (SUD). Results: ADHD (aHR: 0.85; 95% CI: 0.81–0.88) and SUD medications (aHR: 0.97; 0.93–0.99) were associated with reduced SUD hospitalisation risk. In contrast, antipsychotics (aHR: 1.12; 1.09–1.15), antidepressants (aHR: 1.09; 1.06–1.11), and benzodiazepines (aHR: 1.14; 1.11–1.17) were associated with increased risk, while mood stabilisers showed no significant effect. Specific agents, including clozapine (aHR: 0.47; 0.31–0.73), naltrexone (aHR: 0.79; 0.71–0.88), buprenorphine (aHR: 0.92; 0.88–0.96), lisdexamfetamine (aHR: 0.74; 0.69–0.79), methylphenidate (aHR: 0.90; 0.86–0.94), and ADHD polytherapy (aHR: 0.85; 0.75–0.96), were associated with reduced risk. Conclusions: Most psychiatric medications were not associated with a reduced risk of SUD in individuals diagnosed with CUD. However, pharmacotherapies for ADHD, opioid or alcohol use disorder, and treatment-resistant schizophrenia were linked to better outcomes, likely reflecting the underlying clinical trajectories of other comorbid psychiatric disorders rather than direct therapeutic effects on CUD.

Introduction: Diabetes, obesity and bipolar disorder often co‐occur and may have shared pathophysiology. Glucagon‐like peptide 1 receptor agonists (GLP‐1RA) treat diabetes and obesity but their impact on bipolar disorder is unknown. In this era of stagnated pharmacotherapy, we examined psychiatric hospitalisation and absence from work due to sick leave as potential measures of relapse in people diagnosed with bipolar disorder who were also prescribed antidiabetic medications, including GLP‐1RA. Methods: The study cohort was derived from the National Swedish Registers and included all people with a diagnosis of bipolar disorder who used any antidiabetic medication, between the years 2009 and 2024. GLP ‐ 1RA individually and as a group were compared with non‐use of GLP ‐ 1RA , and directly with other second‐line antidiabetic medications. A within‐individual design was utilised for all comparisons to reduce confounding. The main outcome was psychiatric hospitalisation for any reason, with secondary outcomes being psychiatric hospitalisation after a relapse of bipolar disorder, and sick leave due to psychiatric reasons. Within‐individual stratified Cox models with adjusted hazard ratios ( aHRs ) and 95% confidence intervals ( CIs ) were used. Results: 14,694 people were included in the study, of whom 5200 used GLP ‐ 1RA . Semaglutide was associated with a 21% ( aHR 0.79, 95% CI 0.69–0.91) lower risk of psychiatric hospitalisation compared with non‐use of GLP ‐ 1RA in that same individual. Liraglutide and dulaglutide were not specifically associated with a lower hospitalisation risk. Semaglutide was also linked to a 17% ( aHR 0.83, 95% CI 0.69–0.99) lower risk of relapse of bipolar disorder. Absences from work due to sick leave were not significantly associated with the specific antidiabetic medications studied. Conclusion: In people with both bipolar disorder and diabetes and/or obesity, the use of semaglutide was linked to lower rates of psychiatric hospitalisation compared with time periods when GLP‐1RA were not used by that individual. Liraglutide and dulaglutide were not associated with reduced psychiatric hospitalisation, implying this is not a class effect. This finding with semaglutide should be further tested in a randomised controlled trial.

Lithium is the gold-standard maintenance treatment for bipolar disorder, yet many patients experience inadequate response or intolerance. Using harmonized nationwide cohorts from Sweden ( n = 105,495) and Finland ( n = 60,045) with an average of 9 years of follow-up, we assessed the comparative effectiveness of mood stabilizers, antipsychotics and their combinations in preventing psychiatric hospitalization. Within-individual analyses revealed that, compared with lithium monotherapy, clozapine plus aripiprazole (adjusted hazard ratio = 0.42; 95% confidence interval, 0.22–0.80), clozapine alone (0.61; 0.49–0.75) and long-acting injectable antipsychotics + lithium (0.70; 0.59–0.84) were associated with lower relapse risk. Among 20,645 lithium discontinuers, long-acting injectable antipsychotics + valproate (0.41; 0.19–0.92), quetiapine + lamotrigine (0.64; 0.51–0.79), olanzapine + valproate (0.65; 0.50–0.84) and risperidone + valproate (0.63; 0.42–0.93) were linked to reduced hospitalization risk. These findings highlight specific regimens that provide potential options for patients in whom lithium is ineffective or discontinued.

Background: Research indicates that first‐episode psychosis (FEP) with cannabis use disorder (CUD) carries a substantial disease burden, but more granular clinical trajectories of cannabis‐related psychoses remain unclear. This study examines the clinical trajectories of first clinically diagnosed FEP&CUD and cannabis‐induced psychosis (CIP) in comparison with FEP without substance use disorder (SUD). Methods: From the linkage of nationwide administrative and medical registers of Sweden (2006–2021), we identified 1772 individuals (84.1% men) with incident CIP, 1360 individuals (84.3% men) with FEP&CUD and matched individuals with FEP but without SUD and followed them up until 2023. We compared annual prevalence of psychiatric diagnoses before and after cohort entry, medication use, days in inpatient care, and the risks of hospitalization for psychosis and death. Results: Mean age at first diagnosis was 26.6 years (SD 8.3) for incident CIP and 26.9 years (SD 8.4) for FEP&CUD and FEP without SUD. Individuals in the FEP without SUD cohort were more likely to have a record of schizophrenia diagnosis compared to the FEP&CUD and CIP cohorts during the first (14.41%, 6.25%, 2.37%) and second year (10.88%, 7.13%, 3.27%) after cohort entry. However, FEP&CUD and CIP cohorts had a more pronounced burden of inpatient treatment as well as elevated risks of hospitalization for psychosis (aHR = 1.66; 95% CI 1.49–1.86 and aHR = 1.47; 1.32–1.64) and death (aHR = 2.02; 1.44–2.82 and aHR = 1.80; 1.30–2.50) compared to individuals with FEP without SUD. Conclusions: Cannabis‐related psychoses, that is, FEP&CUD and CIP, were associated with poor outcomes, including high risk of hospitalization for psychosis and increased mortality, underscoring the need for targeted interventions. Moreover, the overlapping clinical trajectories suggest that cannabis‐related psychoses may not represent clinically distinct entities but instead lie on a continuum.

Background and Hypothesis: Individuals with psychosis have an increased risk of committing and being victims of violence. There are frequent media reports of psychotic individuals assaulting strangers, which may cause fear and stigmatization among the general public. We hypothesize that homicides targeting strangers by psychotic individuals are rare. Study Design: Systematic review and meta-analysis to assess what percentage of homicide offenders suffering from psychosis target strangers. Medline database was searched with search term ‘psychosis OR schizophrenia AND homicide’ from inception to 10/2024. Articles published in peer-reviewed journals, written in English, and reporting the total number of homicide offenders with psychosis (ICD-10: F20, F22, F25, F30-F31, F32.3, F29) and stranger victims were included. PRISMA guidelines were followed. Studies with inadequate data were excluded. A random-effects meta-analysis using meta and metafor packages in R version 4.4.2 was conducted using the restricted maximum-likelihood (REML) method to account for variability across studies. The primary outcome was the pooled rate of stranger homicides among individuals with psychosis who committed a homicide, expressed as a proportion with 95% confidence intervals (CIs). Study Results: Thirteen studies were included, comprising a total of 1,438 perpetrators who had killed 177 strangers. Meta-analysis of these studies indicates that 12.7% (95% CI: 7.85–17.56 and heterogeneity I2 89.49%) of the homicides by psychotic individuals are targeted at strangers. Male gender explained 26.9% of between-study variance (P-value<0.05). Conclusions: Although still rare, the percentage of homicides committed by psychotic individuals and targeted at strangers is higher than previously reported.

There are no established prognostic tools for predicting relapse risk in first-episode bipolar disorder (FEBD), limiting the use of personalized treatment approaches. We aimed to develop and validate a machine learning (ML) model to predict relapse in FEBD patients and assess whether pharmacotherapy effectiveness varies across predicted risk strata. We used nationwide registry data from Sweden (n = 30,402; follow-up = 2006–2021) and Finland (n = 13,790; follow-up = 1996–2018). The developed ML model achieved an area under the receiver operating characteristic curve (AUROC) of 0.71 (95% CI = 0.69–0.72, highest for relapse due to psychotic mania [AUROC = 0.85, 95% CI = 0.80–0.89]) in the Swedish cohort (internal validation) and 0.68 (95% CI = 0.66–0.69, highest for nonpsychotic mania [AUROC = 0.74, 95% CI = 0.69–0.78]) in the Finnish cohort (external validation). The model incorporated only seven accessible predictors, including pharmacotherapies during the first 30 days post-FEBD (lithium, combination treatments, and antipsychotics), outpatient follow-up within 30 days post-FEBD, prolonged initial hospitalization for FEBD, history of prior psychiatric hospitalizations, and sickness absence days. Among high-risk patients (N = 10,119, 31.77%), long-acting injectable (LAI) antipsychotics were associated with the greatest reduction in psychiatric rehospitalization risk (HR = 0.44, 95% CI 0.29–0.67). In the low-risk group (n = 21,736, 68.23%), only combinations of quetiapine with valproate (HR = 0.79, 95% CI = 0.65–0.95) or lamotrigine (HR = 0.86, 95% CI = 0.74–0.99) were associated with reduced rehospitalization risk. Available online for research purposes, our internally and externally validated prognostic model may inform clinical decision-making by identifying high-risk individuals who could benefit from early initiation of LAI antipsychotics, promoting a shift toward proactive, risk-guided treatment in FEBD patients.

Introduction: Despite well‐known diagnostic and neurobiological overlaps between psychopathic traits and schizophrenia, it has remained unclear whether psychopathic traits increase the risk for later schizophrenia. Former studies have proven only a weak correlation between psychopathy and DSM axis I diagnoses. Methods: We combined data from individuals who underwent forensic psychiatric evaluations (FPEs) at Niuvanniemi Hospital between 1984 and 1993 with the records from the Care Register for Health Care to examine the relationship between psychopathic traits, measured by the Psychopathy Checklist‐Revised (PCL‐R), and the development of schizophrenia following the evaluation. We conducted survival analyses using Kaplan–Meier estimates and Cox proportional hazards models, with a follow‐up period of up to 40 years. Mortality data were obtained from the National Death Registry. Statistical analyses were adjusted for age, sex, criminal responsibility, and substance abuse disorder at the time of the FPE. Results: The study included 341 individuals (278 males [81.51%] and 63 females [18.49%], mean [SD] age 33.52 [11.49]) who were adults, criminally responsible, and did not have a psychotic illness, severe mental disability, or brain damage at FPE. Compared to individuals with total PCL‐R scores less than or equal to 10, those with scores of 11–24 (adjusted hazard ratio [aHR] = 5.30, 95% CI = 1.21–23.25) and 25 or higher (aHR = 9.33, 95% CI = 2.04–42.76) had a significantly higher risk of later hospitalization due to schizophrenia. Also, individuals classified as psychopathic (PCL‐ R ≥ 25) had a significantly higher risk of developing schizophrenia compared with those classified as non‐psychopathic (PCL‐ R < 25): aHR = 2.37, 95% CI =1.17–4.80. A total of 20% of psychopaths developed schizophrenia over the follow‐up. Conclusions: The novel results suggest that there is a link between higher PCL‐R scores and a higher risk of later‐life schizophrenia outbreak among non‐psychotic individuals undergoing FPE. Multiple factors can explain the finding, including substance use and mutual risk factors.

Key Points: Question: Is variation in global functioning among people with schizophrenia associated with genetic differences? Findings: In this genetic association study of 5991 adults with schizophrenia and 59,795 hospitalizations, polygenic scores for schizophrenia and educational attainment predicted global functioning at psychiatric admission and discharge, as well as the magnitude of functional improvement during hospitalization. Higher polygenic burden for schizophrenia was consistently associated with worse functional outcomes, including less functional improvement during hospitalization, while educational attainment polygenic score showed a more complex pattern of associations. Meaning: Polygenic scores may help disentangle heterogeneity in schizophrenia functioning and course. Importance: Schizophrenia is characterized by heterogeneity in disease outcomes and course. The extent to which this heterogeneity is associated with genetic variation is unclear. Objective: To investigate whether polygenic scores (PGS) for schizophrenia and educational attainment are associated with global functioning during inpatient psychiatric hospitalization among patients with schizophrenia. Design, Setting, and Participants: Adults with schizophrenia were recruited nationwide for the SUPER-Finland Study between 2015 and 2018. Global functioning scores recorded during hospitalizations between 1994 and 2019 were extracted from a complete, longitudinal register. Data for the current genetic association study were analyzed between May 2024 and April 2025. We used linear mixed-effects models to examine associations among PGS and global functioning at admission and discharge, as well as functional change during hospitalization. Exposures: Psychiatric hospitalization and PGS for schizophrenia and educational attainment. Main Outcomes and Measures: Admission global functioning, discharge global functioning, and functional change during hospitalization. Results: We analyzed 117,810 global functioning scores from 59,795 hospitalizations and 5991 participants (2733 [45.62%] female, median [IQR] age = 47 [20] years). Higher schizophrenia PGS predicted lower admission global functioning ( β = −0.20; 95% CI, −0.40 to 0.00; P = .05) and discharge global functioning ( β = −0.36; 95% CI,−0.56 to −0.17; P < .001), and less functional improvement ( β = −0.31; 95% CI, −0.49 to −0.14; P < .001). Higher educational attainment PGS predicted greater functional improvement ( β = 0.20; 95% CI, 0.03 to 0.37; P = .02) but worse admission global functioning ( β = −0.23; 95% CI, −0.43 to −0.04; P = .02). Conclusions and Relevance: Higher genetic liability for schizophrenia is associated with worse global functioning across psychiatric hospitalization, including less functional improvement. Integrating PGS and clinically relevant, longitudinal disease outcomes may help parse heterogeneity in schizophrenia prognosis and course.

Schizophrenia (SZ), schizoaffective disorder (SZA), bipolar disorder (BD), and psychotic depression (PD) are associated with premature death due to preventable general medical comorbidities (GMCs). The interaction between psychosis, risk factors, and GMCs is complex and should be elucidated. More research particularly among those with SZA or PD is warranted. We evaluated the association between registry-based psychotic disorders and GMC diagnoses in a large national sample of participants with different psychotic disorders. In addition, we examined whether body mass index (BMI) and smoking as risk factors for GMCs explain differences between diagnostic groups. This was a cross-sectional study of a clinical population of participants (n = 10,417) in the Finnish SUPER study. Registry-based diagnoses of psychotic disorders and hypertension, diabetes, chronic obstructive pulmonary disease (COPD), cancers, ischemic heart disease, and liver disorders were obtained. Participants’ BMI and self-reported smoking were recorded. Total effect of diagnostic category adjusted for age and sex as well as direct effect including known risk factors was calculated using logistic regression. Regardless of diagnostic category, participants had high BMI (average 30.3 kg/m²), and current smoking was common (42.4%). Diabetes and COPD were more common in SZ than in other diagnostic categories. The differences between psychotic disorders were not explained by obesity or smoking status only. Obesity and smoking were prevalent in all diagnostic categories of psychotic disorders, and continued efforts at prevention are warranted. Additional differences in GMC prevalence exist between psychotic disorders that are not explained by obesity and smoking.