Twin and family studies using Western samples have established that child and adolescent anxiety and depression are under substantial genetic, modest shared environmental, and substantial non-shared environmental influences. Generalizability of these findings to non-Western societies remains largely unknown, particularly regarding the changes of genetic and environmental influences with age. The current study examined changes in genetic and environmental influences on self-reported anxiety and depression from late childhood to mid-adolescence among a Chinese twin sample. Sex differences were also examined.

Self-reported anxiety and depression were collected from 712 10- to 12-year-old Chinese twins (mean = 10.88 years, 49% males) and again 3 years later. Quantitative genetic modeling was used to examine developmental changes in genetic and environmental influences on anxiety and depression, and sex differences.

Heritability of anxiety and depression in late childhood (23 and 20%) decreased to negligible in mid-adolescence, while shared environmental influences increased (20 and 27% to 57 and 60%). Shared environmental factors explained most of the continuity of anxiety and depression (75 and 77%). Non-shared environmental factors were largely time-specific. No sex differences were observed.

Shared environmental influences might be more pronounced during the transition period of adolescence in non-Western societies such as China. Future research should examine similarities and differences in the genetic and environmental etiologies of child and adolescent internalizing and other psychopathology in development between Western and non-Western societies.

Social functioning (SF) difficulties are ubiquitous among individuals at clinical high risk for psychosis (CHR), but it is not yet clear why. One possibility is suggested by the observation that effective SF requires adaptive emotion awareness and regulation. Previous reports have documented deficits in emotion awareness and regulation in individuals with schizophrenia, and have shown that such deficits predicted SF. However, it is unknown whether these deficits are present prior to the onset of psychosis or whether they are linked to SF in CHR individuals.

We conducted a cross-sectional comparison of emotion awareness and regulation in 54 individuals at CHR, 87 with schizophrenia and 50 healthy controls (HC). Then, within the CHR group, we examined links between emotion awareness, emotion regulation and SF as indexed by the Global Functioning Scale: Social (Cornblatt et al. 2007).

Group comparisons indicated significant differences between HC and the two clinical groups in their ability to identify and describe feelings, as well as the use of suppression and reappraisal emotion-regulation strategies. Specifically, the CHR and schizophrenia groups displayed comparable deficits in all domains of emotion awareness and emotion regulation. A hierarchical multiple regression analysis indicated that difficulties describing feelings accounted for 23.2% of the SF variance.

The results indicate that CHR individuals display substantial emotion awareness and emotion-regulation deficits, at severity comparable with those observed in individuals with schizophrenia. Such deficits, in particular difficulties describing feelings, predate the onset of psychosis and contribute significantly to poor SF in this population.

Anorexia nervosa (AN) is a serious illness leading to substantial morbidity and mortality. The treatment of AN very often is protracted; repeated hospitalizations and lost productivity generate substantial economic costs in the health care system. Therefore, this study aimed to determine the differential cost-effectiveness of out-patient focal psychodynamic psychotherapy (FPT), enhanced cognitive–behavioural therapy (CBT-E), and optimized treatment as usual (TAU-O) in the treatment of adult women with AN.

The analysis was conducted alongside the randomized controlled Anorexia Nervosa Treatment of OutPatients (ANTOP) study. Cost-effectiveness was determined using direct costs per recovery at 22 months post-randomization (n = 156). Unadjusted incremental cost-effectiveness ratios (ICERs) were calculated. To derive cost-effectiveness acceptability curves (CEACs) adjusted net-benefit regressions were applied assuming different values for the maximum willingness to pay (WTP) per additional recovery. Cost–utility and assumptions underlying the base case were investigated in exploratory analyses.

Costs of in-patient treatment and the percentage of patients who required in-patient treatment were considerably lower in both intervention groups. The unadjusted ICERs indicated FPT and CBT-E to be dominant compared with TAU-O. Moreover, FPT was dominant compared with CBT-E. CEACs showed that the probability for cost-effectiveness of FTP compared with TAU-O and CBT-E was ⩾95% if the WTP per recovery was ⩾€9825 and ⩾€24 550, respectively. Comparing CBT-E with TAU-O, the probability of being cost-effective remained <90% for all WTPs. The exploratory analyses showed similar but less pronounced trends.

Depending on the WTP, FPT proved cost-effective in the treatment of adult AN.

A significant number of patients with schizophrenia fail to respond to antipsychotic medication. Although several studies have investigated associated patient characteristics, the emerging findings from genetic studies offer further scope for study.

In 612 schizophrenia patients with detailed clinical information, common genetic variants indexed by polygenic risk scores, and rare variants indexed by deletion and duplication burden genomewide, we explored potential genetic predictors alongside other established risk factors for treatment resistance. Clinical outcomes of treatment resistance were also calculated using lifetime measures of positive, negative/disorganized and mood symptoms as well as number of hospitalizations and suicide attempts.

Logistic regression models identified a significant relationship between treatment resistance and total duplication burden genomewide, years of formal schooling and age at onset. Clinically, treatment-resistant patients were characterized by greater negative/disorganized and positive symptoms and greater number of hospitalizations.

Taken together, these findings suggest genetic information, specifically the genomewide burden of rare copy number variants, may increase our understanding and clinical management of patients with treatment-resistant schizophrenia.

Impairment in social cognition is an established finding in autism spectrum disorders (ASD). Emerging evidence suggests that attention-deficit/hyperactivity disorder (ADHD) might be also associated with deficits in theory of mind (ToM) and emotion recognition. However, there are inconsistent findings, and it has been debatable whether such deficits persist beyond childhood and how similar social cognitive deficits are in ADHD v. ASD.

We conducted a meta-analysis of social cognition, including emotion recognition and ToM, studies in ADHD compared with healthy controls and ASD. The current meta-analysis involved 44 studies comparing ADHD (n = 1999) with healthy controls (n = 1725) and 17 studies comparing ADHD (n = 772) with ASD (n = 710).

Facial and vocal emotion recognition (d = 0.40–0.44) and ToM (d = 0.43) abilities were significantly impaired in ADHD. The most robust facial emotion recognition deficits were evident in anger and fear. Social cognitive deficits were either very subtle (emotion recognition) or non-significant (ToM) in adults with ADHD. Deficits in social cognition, especially ToM, were significantly more pronounced in ASD compared with ADHD. General cognitive impairment has contributed to social cognitive deficits in ADHD.

Performance of individuals with ADHD on social cognition lies intermediate between ASD and healthy controls. However, developmental trajectories of social cognition probably differ between ADHD and ASD as social cognitive deficits in ADHD might be improving with age in most individuals. There is a need for studies investigating a potential subtype of ADHD with persistent social cognitive deficits and exploring longitudinal changes in social cognition during development.

The concept of cognitive reserve (CR) hypothesizes that intellectually stimulating activities provide resilience against brain pathology/disease. Whereas brain abnormalities and cognitive impairment are frequently reported in bipolar disorder (BD), it is unknown whether the impact of brain alterations can be lessened by higher CR in BD.

We tested if higher CR would reduce the influence of total volumes of deep white matter hypointensities (WMH), ventricular cerebrospinal fluid (CSF), and prefrontal cortex on memory, executive, and attention/speed functions in patients with BD (n = 75). Linear regression models with interaction terms for CR and brain volumes were applied to directly test if CR reduces the influence of brain pathology on cognitive domains.

CR reduced the influence of total volumes of deep WMH (β = −0.38, Q = 0.003) and ventricular CSF (β = −41, Q = 006) on executive functions.

The interactions between CR and total volumes of deep WMH/ventricular CSF appear to account for executive functioning in BD. The results suggest that the concept of CR is applicable in BD. Higher reserve capacity in BD alters the relationship between brain pathology and clinical presentation.

Late-life depression (LLD) in the elderly was reported to present with emotion dysregulation accompanied by high perceived loneliness. Previous research has suggested that LLD is a disorder of connectivity and is associated with aberrant network properties. On the other hand, perceived loneliness is found to adversely affect the brain, but little is known about its neurobiological basis in LLD. The current study investigated the relationships between the structural connectivity, functional connectivity during affective processing, and perceived loneliness in LLD.

The current study included 54 participants aged >60 years of whom 31 were diagnosed with LLD. Diffusion tensor imaging (DTI) data and task-based functional magnetic resonance imaging (fMRI) data of an affective processing task were collected. Network-based statistics and graph theory techniques were applied, and the participants’ perceived loneliness and depression level were measured. The affective processing task included viewing affective stimuli.

Structurally, a loneliness-related sub-network was identified across all subjects. Functionally, perceived loneliness was related to connectivity differently in LLD than that in controls when they were processing negative stimuli, with aberrant networking in subcortical area.

Perceived loneliness was identified to have a unique role in relation to the negative affective processing in LLD at the functional brain connectional and network levels. The findings increas our understanding of LLD and provide initial evidence of the neurobiological mechanisms of loneliness in LLD. Loneliness might be a potential intervention target in depressive patients.

Data on gender-specific profiles of cognitive functions in patients with Parkinson's disease (PD) are rare and inconsistent, and possible disease-confounding factors have been insufficiently considered.

The LANDSCAPE study on cognition in PD enrolled 656 PD patients (267 without cognitive impairment, 66% male; 292 with mild cognitive impairment, 69% male; 97 with PD dementia, 69% male). Raw values and age-, education-, and gender-corrected Z scores of a neuropsychological test battery (CERAD-Plus) were compared between genders. Motor symptoms, disease duration, l-dopa equivalent daily dose, depression - and additionally age and education for the raw value analysis - were taken as covariates.

Raw-score analysis replicated results of previous studies in that female PD patients were superior in verbal memory (word list learning, p = 0.02; recall, p = 0.03), while men outperformed women in visuoconstruction (p = 0.002) and figural memory (p = 0.005). In contrast, gender-corrected Z scores showed that men were superior in verbal memory (word list learning, p = 0.02; recall, p = 0.02; recognition, p = 0.04), while no difference was found for visuospatial tests. This picture could be observed both in the overall analysis of PD patients as well as in a differentiated group analysis.

Normative data corrected for gender and other sociodemographic variables are relevant, since they may elucidate a markedly different cognitive profile compared to raw scores. Our study also suggests that verbal memory decline is stronger in women than in men with PD. Future studies are needed to replicate these findings, examine the progression of gender-specific cognitive decline in PD and define different underlying mechanisms of this dysfunction.

Lysergic acid diethylamide (LSD) is a potent serotonergic hallucinogen or psychedelic that modulates consciousness in a marked and novel way. This study sought to examine the acute and mid-term psychological effects of LSD in a controlled study.

A total of 20 healthy volunteers participated in this within-subjects study. Participants received LSD (75 µg, intravenously) on one occasion and placebo (saline, intravenously) on another, in a balanced order, with at least 2 weeks separating sessions. Acute subjective effects were measured using the Altered States of Consciousness questionnaire and the Psychotomimetic States Inventory (PSI). A measure of optimism (the Revised Life Orientation Test), the Revised NEO Personality Inventory, and the Peter's Delusions Inventory were issued at baseline and 2 weeks after each session.

LSD produced robust psychological effects; including heightened mood but also high scores on the PSI, an index of psychosis-like symptoms. Increased optimism and trait openness were observed 2 weeks after LSD (and not placebo) and there were no changes in delusional thinking.

The present findings reinforce the view that psychedelics elicit psychosis-like symptoms acutely yet improve psychological wellbeing in the mid to long term. It is proposed that acute alterations in mood are secondary to a more fundamental modulation in the quality of cognition, and that increased cognitive flexibility subsequent to serotonin 2A receptor (5-HT2AR) stimulation promotes emotional lability during intoxication and leaves a residue of ‘loosened cognition’ in the mid to long term that is conducive to improved psychological wellbeing.

Heritability estimates from twin studies of the multi-faceted phenotype of nicotine dependence (ND) range from moderate to high (31–60%), but vary substantially based on the specific ND-related construct examined. The current study estimated the aggregate role of common genetic variants on key ND constructs.

Genomic-relationship-matrix restricted maximum likelihood (GREML) was used to decompose phenotypic variance across multiple ND indices using 796 125 polymorphisms from 2346 unrelated ‘lifetime ever smokers’ of European ancestry. Measures included DSM-IV ND and Fagerström Test for Nicotine Dependence (FTND) summary measures and constituent constructs (e.g. withdrawal severity, tolerance, heaviness of smoking and time spent smoking). Exploratory and confirmatory factor models were used to describe the covariance structure across ND measures; resulting factor(s) were the subject(s) of GREML analyses.

Factor models indicated highly correlated DSM-IV and FTND factors for ND (0.545, 95% confidence interval 0.50–0.60) that could be represented as a higher-order factor (NIC DEP). Additive genetic influence on NIC DEP was 33% (s.e. = 0.14, p = 0.009). Post-hoc analyses indicated moderate genetic effects on the DSM-IV (34%, s.e. = 0.14, p = 0.008) and FTND (26%, s.e. = 0.14, p = 0.032) factors, both of which were influenced by the same genetic effects (rG-SNP = 1.00, s.e. = 0.09, p < 0.00001).

Overall, common single nucleotide polymorphisms accounted for a large proportion of the genetic influences on ND-related phenotypes that have been observed in twin studies. Genetic contributions across distinct ND scales were largely influenced by shared genetic factors.

Post-traumatic stress disorder (PTSD) has been linked to hypertension, but most research on PTSD and hypertension is cross-sectional, and potential mediators have not been clearly identified. Moreover, PTSD is twice as common in women as in men, but understanding of the PTSD-hypertension relationship in women is limited. We examined trauma exposure and PTSD symptoms in relation to incident hypertension over 22 years in 47 514 civilian women in the Nurses’ Health Study II.

We used proportional hazards models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for new-onset hypertension (N = 15 837).

PTSD symptoms assessed with a screen were modestly associated with incident hypertension in a dose-response fashion after adjusting for potential confounders. Compared to women with no trauma exposure, women with 6–7 PTSD symptoms had the highest risk of developing hypertension (HR 1.20, 95% CI 1.12–1.30), followed by women with 4–5 symptoms (HR 1.17, 95% CI 1.10–1.25), women with 1–3 symptoms (HR 1.12, 95% CI 1.06–1.18), and trauma-exposed women with no symptoms (HR 1.04, 95% CI 1.00–1.09). Findings were maintained, although attenuated, adjusting for hypertension-relevant medications, medical risk factors, and health behaviors. Higher body mass index and antidepressant use accounted for 30% and 21% of the PTSD symptom-hypertension association, respectively.

Screening for hypertension and reducing unhealthy lifestyle factors, particularly obesity, in women with PTSD may hold promise for offsetting cardiovascular risk.

Major depressive disorder (MDD) is moderately heritable, however genome-wide association studies (GWAS) for MDD, as well as for related continuous outcomes, have not shown consistent results. Attempts to elucidate the genetic basis of MDD may be hindered by heterogeneity in diagnosis. The Center for Epidemiological Studies Depression (CES-D) scale provides a widely used tool for measuring depressive symptoms clustered in four different domains which can be combined together into a total score but also can be analysed as separate symptom domains.

We performed a meta-analysis of GWAS of the CES-D symptom clusters. We recruited 12 cohorts with the 20- or 10-item CES-D scale (32 528 persons).

One single nucleotide polymorphism (SNP), rs713224, located near the brain-expressed melatonin receptor (MTNR1A) gene, was associated with the somatic complaints domain of depression symptoms, with borderline genome-wide significance (pdiscovery = 3.82 × 10−8). The SNP was analysed in an additional five cohorts comprising the replication sample (6813 persons). However, the association was not consistent among the replication sample (pdiscovery+replication = 1.10 × 10−6) with evidence of heterogeneity.

Despite the effort to harmonize the phenotypes across cohorts and participants, our study is still underpowered to detect consistent association for depression, even by means of symptom classification. On the contrary, the SNP-based heritability and co-heritability estimation results suggest that a very minor part of the variation could be captured by GWAS, explaining the reason of sparse findings.

The study aimed to subtype patients with schizophrenia on the basis of social cognition (SC), and to identify cut-offs that best discriminate among subtypes in 809 out-patients recruited in the context of the Italian Network for Research on Psychoses.

A two-step cluster analysis of The Awareness of Social Inference Test (TASIT), the Facial Emotion Identification Test and Mayer–Salovey–Caruso Emotional Intelligence Test scores was performed. Classification and regression tree analysis was used to identify the cut-offs of variables that best discriminated among clusters.

We identified three clusters, characterized by unimpaired (42%), impaired (50.4%) and very impaired (7.5%) SC. Three theory-of-mind domains were more important for the cluster definition as compared with emotion perception and emotional intelligence. Patients more able to understand simple sarcasm (⩾14 for TASIT-SS) were very likely to belong to the unimpaired SC cluster. Compared with patients in the impaired SC cluster, those in the very impaired SC cluster performed significantly worse in lie scenes (TASIT-LI <10), but not in simple sarcasm. Moreover, functioning, neurocognition, disorganization and SC had a linear relationship across the three clusters, while positive symptoms were significantly lower in patients with unimpaired SC as compared with patients with impaired and very impaired SC. On the other hand, negative symptoms were highest in patients with impaired levels of SC.

If replicated, the identification of such subtypes in clinical practice may help in tailoring rehabilitation efforts to the person's strengths to gain more benefit to the person.

Mental health disorders commonly co-occur, even between conceptually distinct syndromes, such as internalizing and externalizing disorders. The current study investigated whether phenotypic, genetic, and environmental variance in negative emotionality and behavioral control account for the covariation between major depressive disorder (MDD) and alcohol use disorder (AUD).

A total of 3623 members of a national twin registry were administered structured diagnostic telephone interviews that included assessments of lifetime histories of MDD and AUD, and were mailed self-report personality questionnaires that assessed stress reactivity (SR) and behavioral control (CON). A series of biometric models were fitted to partition the proportion of covariance between MDD and AUD into SR and CON.

A statistically significant proportion of the correlation between MDD and AUD was due to variance specific to SR (men = 0.31, women = 0.27) and CON (men = 0.20, women = 0.19). Further, genetic factors explained a large proportion of this correlation (0.63), with unique environmental factors explaining the rest. SR explained a significant proportion of the genetic (0.33) and environmental (0.23) overlap between MDD and AUD. In contrast, variance specific to CON accounted for genetic overlap (0.32), but not environmental overlap (0.004). In total, SR and CON accounted for approximately 70% of the genetic and 20% of the environmental covariation between MDD and AUD.

This is the first study to demonstrate that negative emotionality and behavioral control confer risk for the co-occurrence of MDD and AUD via genetic factors. These findings are consistent with the aims of NIMH's RDoC proposal to elucidate how transdiagnostic risk factors drive psychopathology.

Metacognitive training (MCT) for schizophrenia spectrum is widely implemented. It is timely to systematically review the literature and to conduct a meta-analysis.

Eligible studies were selected from several sources (databases and expert suggestions). Criteria included comparative studies with a MCT condition measuring positive symptoms and/or delusions and/or data-gathering bias. Three meta-analyses were conducted on data gathering (three studies; 219 participants), delusions (seven studies; 500 participants) and positive symptoms (nine studies; 436 participants). Hedges’ g is reported as the effect size of interest. Statistical power was sufficient to detect small to moderate effects.

All analyses yielded small non-significant effect sizes (0.26 for positive symptoms; 0.22 for delusions; 0.31 for data-gathering bias). Corrections for publication bias further reduced the effect sizes to 0.21 for positive symptoms and to 0.03 for delusions. In blinded studies, the corrected effect sizes were 0.22 for positive symptoms and 0.03 for delusions. In studies using proper intention-to-treat statistics the effect sizes were 0.10 for positive symptoms and −0.02 for delusions. The moderate to high heterogeneity in most analyses suggests that processes other than MCT alone have an impact on the results.

The studies so far do not support a positive effect for MCT on positive symptoms, delusions and data gathering. The methodology of most studies was poor and sensitivity analyses to control for methodological flaws reduced the effect sizes considerably. More rigorous research would be helpful in order to create enough statistical power to detect small effect sizes and to reduce heterogeneity. Limitations and strengths are discussed.

Current ultra-high-risk (UHR) criteria appear insufficient to predict imminent onset of first-episode psychosis, as a meta-analysis showed that about 20% of patients have a psychotic outcome after 2 years. Therefore, we aimed to develop a stage-dependent predictive model in UHR individuals who were seeking help for co-morbid disorders.

Baseline data on symptomatology, and environmental and psychological factors of 185 UHR patients (aged 14–35 years) participating in the Dutch Early Detection and Intervention Evaluation study were analysed with Cox proportional hazard analyses.

At 18 months, the overall transition rate was 17.3%. The final predictor model included five variables: observed blunted affect [hazard ratio (HR) 3.39, 95% confidence interval (CI) 1.56–7.35, p < 0.001], subjective complaints of impaired motor function (HR 5.88, 95% CI 1.21–6.10, p = 0.02), beliefs about social marginalization (HR 2.76, 95% CI 1.14–6.72, p = 0.03), decline in social functioning (HR 1.10, 95% CI 1.01–1.17, p = 0.03), and distress associated with suspiciousness (HR 1.02, 95% CI 1.00–1.03, p = 0.01). The positive predictive value of the model was 80.0%. The resulting prognostic index stratified the general risk into three risk classes with significantly different survival curves. In the highest risk class, transition to psychosis emerged on average ⩾8 months earlier than in the lowest risk class.

Predicting a first-episode psychosis in help-seeking UHR patients was improved using a stage-dependent prognostic model including negative psychotic symptoms (observed flattened affect, subjective impaired motor functioning), impaired social functioning and distress associated with suspiciousness. Treatment intensity may be stratified and personalized using the risk stratification.

Maternal stress during pregnancy may increase the risk of preterm delivery (PD), but the associations between stress and subtypes of PD are not clear. We investigated maternal loss of a close relative and risks of very and moderately PD (<32 and 32–36 weeks, respectively) and spontaneous and medically indicated PD.

We studied 4 940 764 live singleton births in Denmark (1978–2008) and Sweden (1973–2006). We retrieved information on death of women's family members (children, partner, siblings, parents), birth outcomes and maternal characteristics from nationwide registries.

Overall, the death of a close family member the year before pregnancy or in the first 36 weeks of pregnancy was associated with a 7% increased risk of PD [95% confidence interval (CI) 1.04–1.10]. The highest hazard ratios (HR) for PD were found for death of an older child [HR (95% CI) 1.20 (1.10–1.31)] and for death of a partner [HR (95% CI) 1.31 (1.03–1.66)]. These losses were associated with higher risks of very preterm [HR (95% CI) 1.61 (1.29–2.01) and 2.07 (1.15–3.74), respectively] than of moderately preterm [HR (95% CI) 1.14 (1.03–1.26) and 1.22 (0.94–1.58), respectively] delivery. There were no substantial differences in the association between death of a child or partner and the risk of spontaneous v. medically indicated PD.

Death of a close family member the year before or during pregnancy was associated with an increased risk of PD, especially very PD. Possible mechanisms include both spontaneous and medically indicated preterm birth.

Deficits in gamma aminobutyric acid (GABA) neuron-related markers, including the GABA-synthesizing enzyme GAD67, the calcium-binding protein parvalbumin, the neuropeptide somatostatin, and the transcription factor Lhx6, are most pronounced in a subset of schizophrenia subjects identified as having a ‘low GABA marker’ (LGM) molecular phenotype. Furthermore, schizophrenia shares degrees of genetic liability, clinical features and cortical circuitry abnormalities with schizoaffective disorder and bipolar disorder. Therefore, we determined the extent to which a similar LGM molecular phenotype may also exist in subjects with these disorders.

Transcript levels for GAD67, parvalbumin, somatostatin, and Lhx6 were quantified using quantitative PCR in prefrontal cortex area 9 of 184 subjects with a diagnosis of schizophrenia (n = 39), schizoaffective disorder (n = 23) or bipolar disorder (n = 35), or with a confirmed absence of any psychiatric diagnoses (n = 87). A blinded clustering approach was employed to determine the presence of a LGM molecular phenotype across all subjects.

Approximately 49% of the subjects with schizophrenia, 48% of the subjects with schizoaffective disorder, and 29% of the subjects with bipolar disorder, but only 5% of unaffected subjects, clustered in the cortical LGM molecular phenotype.

These findings support the characterization of psychotic and bipolar disorders by cortical molecular phenotype which may help elucidate more pathophysiologically informed and personalized medications.

It is well-established that offspring of depressed mothers are at increased risk for suicidal ideation. However, pathways involved in the transmission of risk for suicidal ideation from depressed mothers to offspring are poorly understood. The aim of this study was to examine the contribution of potential mediators of this association, including maternal suicide attempt, offspring psychiatric disorder and the parent–child relationship.

Data were utilized from a population-based birth cohort (ALSPAC). Three distinct classes of maternal depression symptoms across the first 11 years of the child's life had already been identified (minimal, moderate, chronic-severe). Offspring suicidal ideation was assessed at age 16 years. Data were analysed using structural equation modelling.

There was evidence for increased risk of suicidal ideation in offspring of mothers with chronic-severe depression symptoms compared to offspring of mothers with minimal symptoms (odds ratio 3.04, 95% confidence interval 2.19–4.21). The majority of this association was explained through maternal suicide attempt and offspring psychiatric disorder. There was also evidence for an independent indirect effect via the parent–child relationship in middle childhood. There was no longer evidence of a direct effect of maternal depression on offspring suicidal ideation after accounting for all three mediators. The pattern of results was similar when examining mechanisms for maternal moderate depression symptoms.

Findings highlight that suicide prevention efforts in offspring of depressed mothers should be particularly targeted at both offspring with a psychiatric disorder and offspring whose mothers have made a suicide attempt. Interventions aimed at improving the parent–child relationship may also be beneficial.