Background: Apart from individuals with clinical psychosis, community surveys have shown that many otherwise-well individuals endorse items designed to identify psychosis. The clinical relevance of psychotic-like experiences (PLE) in individuals who are not psychotic is not clear. This thesis aimed to examine the prevalence, demographic and psychosocial correlates and antecedents of PLE in the Australian community. Method: Data from three population-based studies was examined. The Australian National Survey of Mental Health and Wellbeing interviewed 10,641 individuals living in private dwellings in Australia. We examined the impact of selected demographic variables on endorsement of psychosis screen and probe items from the Composite International Diagnostic Interview (CIDI). We also examined the effect of exposure to traumatic events (with and without Post Traumatic Stress Disorder (PTSD)) on the endorsement of CIDI psychosis items. The youth component of the Australian National Survey of Mental Health and Wellbeing collected demographic and clinical data on a nationally representative sample of 1261 adolescents aged 13-17 years. The prevalence of hallucination endorsement on the Youth Self Report (YSR) in Australian adolescents was examined. The association between hallucination endorsement, demographic variables (sex, age, urbanicity and family composition) and clinical variables (self-reported depression, marijuana and alcohol use, DSM IV diagnoses, emotional/ behavioral problems as reported by the parent on the Child Behaviour Check List (CBCL)) were examined. The Mater- University of Queensland Study of Pregnancy (MUSP) is a prospective longitudinal birth cohort study of 7223 mothers and their offspring who received antenatal care between 1981 and 1984. Psychotic-like experiences were assessed in the offspring at the 21 year follow-up using the Peters Delusional Inventory (PDI) and the CIDI. We examined the prevalence of PLEs, and examined the effect of age (the age range at the 21 year follow-up was 18 to 23 years) and sex on PLE. A second study examined the emotional and behavioural antecedents of PLE at 21 years as measured by the CBCL at five and 14 years and the YSR at 14 years. Results: An estimated 11.7% of Australian adults endorsed at least one psychosisscreening item. Higher endorsement was associated with younger age, migrants from non- English speaking backgrounds, not being married, unemployment, living in an urban region and lowest socioeconomic levels. Exposure to any traumatic event was associated with increased endorsement of PLE (Rate Ratio =2.68; 95% CI 2.18, 3.30). A diagnosis of PTSD further increased endorsement of PLE (Rate Ratio =9.24; 95% CI 6.95, 12.27). Hallucinations were reported by 8.4% of Australian adolescents. Those living in blended or sole parent families were more likely to report hallucinations than those living with both biological parents (OR 3.27; 95% CI 1.93, 5.54; OR 2.60; 95% CI 1.63, 4.13 respectively). Hallucinations were more prevalent in adolescents who had high CBCL scores or elevated depression symptoms (OR 3.30; 95% CI 2.10, 5.20; OR 5.02; 95% CI 3.38, 7.45 respectively). Hallucinations were more prevalent in those adolescents who had smoked cannabis more than twice in the month prior to the survey (OR 3.27; 95% CI 1.76, 6.08). In the 21 year follow-up of the MUSP study, older age (18-20 compared to 21-23 years) was significantly associated with a reduction of CIDI delusions (OR 0.66, 95% CI 0.48, 0.92) and PDI total scores (OR=0.68, 95% CI 0.55, 0.83). Women were significantly more likely to endorse items related to hallucinations (OR=1.49, 95% CI 1.14, 1.95) but not delusions. High CBCL scores at 5 and 14 years predicted high total PDI scores; Those in the highest quartile of YSR scores at 14 years were nearly four times more likely to have high levels of psychotic-like experiences at age 21 (OR=3.71, 95% CI 2.92, 4.71). Adolescent onset psychopathology and continuous psychopathology through both childhood and adolescence strongly predict PLE at age 21. Conclusions: Psychotic-like experiences are relatively common in population surveys of Australian adults and adolescents. In adults, the demographic correlates of PLE are similar to those of schizophrenia. There is a strong association between PLE and exposure to trauma. The correlates of PLE in adolescents are different to those of adults. The reduction in prevalence of delusions between late adolescence and young adulthood is coincident with normal neurophysiological changes in the frontal lobes, suggesting hypotheses suitable for future research. The association between marijuana use and hallucinations in adolescents is further evidence of the potential harm caused by use of cannabis. The onset or persistence of emotional distress during adolescence is associated with an increased risk of PLE in adulthood. These findings are relevant to both clinical practice and future research. Psychotic symptoms create diagnostic uncertainty. There is potential for patients with anxiety (PTSD) or mood disorders to be incorrectly diagnosed with a psychotic illness and receive inappropriate management. Equally, there is a risk that subjects who have an emerging psychotic disorder and comorbid anxiety or depression or have had exposure to traumatic events may have treatment of their psychosis delayed as psychotic symptoms are incorrectly ascribed to a syndrome other than psychosis. Clinicians need to be aware of this diagnostic tension so that accurate assessments and appropriate treatments can be initiated at the earliest possible opportunity. The resolution of childhood emotional distress prior to adolescence appears to reduce the risk of PLE and possibly psychosis in adults. Strategies targeting emotionally distressed children may offer opportunities for psychosis prevention. Further research is required to explore the relationship between PLE and vulnerability to psychosis and other mental health problems. It remains unclear if PLE are reliable endophenotypes for schizophrenia. If this is the case, examining genetic and environmental risk factors for onset and persistence of PLE, and the neuroimaging changes that occur as individuals with PLE convert to psychosis will provide important new clues to our understanding of the aetiopathogenesis of psychosis.

Objectif. L’enquête Santé mentale en Population Générale a été réalisée, en 2003, par le centre collaborateur de l’OMS (CCOMS-Lille) en partenariat avec l’hôpital psychiatrique Mahfoud Boucebci (EHS-Alger). Les données portant sur le risque suicidaire sont exploitées dans le présent travail. Son objectif principal est d’étudier la relation entre le trouble de stress post-traumatique et le risque suicidaire. Méthode. Huit cent quatre-vingt-dix-neuf sujets vivant en Algérie (n=899) ont participé à cette enquête. Le Mini International Neuropsychiatric Interview (MINI) a été utilisé afin de recueillir des données relatives aux troubles mentaux. Résultats. 61 % de la population (n=548) a été exposée à des événements potentiellement « traumatisants ». La prévalence du PTSD est estimée à 13,5 % (n=121) dans l’échantillon global. 13,6 % des sujets (n=122) présentent un risque suicidaire. Aucune relation directe entre le PTSD et le risque suicidaire n’a été observée. Toutefois, le PTSD augmente significativement le risque suicidaire chez les non-pratiquants (OR=5.81 ; IC [1.948-17.328] ; p=0.001) en comparaison avec les sujets pratiquants. Le risque suicidaire est 10 fois plus élevé chez les sujets pratiquants souffrant de problème d’alcool (OR =10.26 ; IC [3.133-33.609] ; p<=0.0001). Conclusion. Cette première étude en population générale souligne la prévalence relativement élevée du risque suicidaire et de PTSD. La pratique religieuse est un élément protecteur contre les conduites suicidaires. Les résultats qui en découlent pourraient être exploités dans le but de fonder une démarche de prévention du risque suicidaire. / Objective. The Mental Health Survey on General Population was conducted in 2003 by the WHO collaborating centres in Lille (WHOCC-Lille) in partnership with the Mahfoud Boucebci psychiatric hospital (EHS-Alger). Data regarding suicidal risk are exploited in the present study. The main objective aims at studying the relationship between post traumatic stress disorder and suicidal risk.Method. Eight hundred ninety-nine subjects living in Algeria (n=899) took part in this survey. Mini International Neuropsychiatric Interview (MINI) was used to collect data related to mental disorders and suicidal behaviors. Results. 61% of the population (n=548) were exposed to « traumatic » events. PTSD prevalence is estimated to be 13.5% (n = 121) in the overall sample. 13.6% of the subjects (n = 122) present a suicidal risk. There is no direct relationship between PTSD and suicidal risk.However, PTSD significantly increases the suicidal risk in non-practising subjects (OR = 5.81, CI [1.948-17.328], p = 0.001) compared to practising subjects. Suicidal risk is 10 times higher in practising subjects suffering from alcohol problems (OR = 10.26, IC [3.133-33.609], p <= 0.0001). Conclusion. This first study in the general population highlights the relatively high prevalence of suicidal risk and PTSD. Religious practice is a protective element against suicidal behavior. These results could be exploited so that a preventive approach to suicidal risk can be put in place.

Background: Over the past decades the public health relevance of mental health conditions in children and adolescents has been of growing concern. However, so far no detailed epidemiological data has been available for a representative national sample in Germany. Objectives: The present paper reports prevalence rates of general and specific mental health problems among children and adolescents in Germany and describes the link between symptoms and impairment as well as the treatment situation. Methods: The mental health module (BELLA study) examines mental health problems in a representative sub-sample of 2,863 families with children aged 7–17 from the National Health Interview and Examination Survey among Children and Adolescents (KiGGS). Mental health problems were determined using the extended version of the strengths and difficulties questionnaire (SDQ). Further standardised screening measures were employed to screen for anxiety disorders (SCARED), conduct disorder (CBCL), attention deficit-/ hyperactivity disorder (FBBHKS, Conners’ Scale) and depressive disorders (CES-DC). Furthermore, substance abuse and suicidal tendencies were assessed. Health-related quality of life (HRQoL) and health care use were determined. Results Overall, 14.5% of the children and adolescents aged 7–17 fulfilled the criteria for at least one specific mental health problem associated with impairment, or had an overall mental health problem indicated by an abnormal SDQ score and present impairment. However, high comorbidity was found in the children concerned. Symptoms of overall mental health problems were present in 8.6% of the children and 6.6% of the adolescents. This number was reduced to prevalence rates of 6.3 and 4.9% when additional impairment was taken as a criterion. Irrespective of the type of disorder, fewer than half of the children affected were reported as receiving treatment. However, for those suffering from mental health problems, large impairments in HRQoL were observed. Conclusions: The observed prevalence of mental health problems as well as their large impact on well-being and functioning calls for early prevention. This is especially important with regard to the large decrease in HRQoL in the children and adolescents affected.

Background: Suicidal behaviour is a major public health problem that has not abated over the past decade. Adopting machine learning algorithms that allow for combining risk factors that may increase the predictive accuracy of models of suicide behaviour is one promising avenue toward effective prevention and treatment. Methods: We used Canadian Community Health Survey – Mental Health and Canadian Forces Mental Health Survey to build conditional inference random forests models of suicidal behaviour in Canadian general population and Canadian Armed Forces. We generated risk algorithms for suicidal behaviour in each sample. We performed within- and between-sample validation and reported the corresponding performance metrics. Results: Only a handful of variables were important in predicting suicidal behaviour in Canadian general population and Canadian Armed Forces. Each model’s performance on within-sample validation was satisfactory, with moderate to high sensitivity and high specificity, while the performance on between-sample validation was conditional on the size and heterogeneity of the training sample. Conclusion: Using conditional inference random forest methodology on large nationally representative mental health surveys has the potential of generating models of suicidal behaviour that not only reflect its complex nature, but indicate that the true positive cases are likely to be captured by this approach.

Background: Over the past decades the public health relevance of mental health conditions in children and adolescents has been of growing concern. However, so far no detailed epidemiological data has been available for a representative national sample in Germany. Objectives: The present paper reports prevalence rates of general and specific mental health problems among children and adolescents in Germany and describes the link between symptoms and impairment as well as the treatment situation. Methods: The mental health module (BELLA study) examines mental health problems in a representative sub-sample of 2,863 families with children aged 7–17 from the National Health Interview and Examination Survey among Children and Adolescents (KiGGS). Mental health problems were determined using the extended version of the strengths and difficulties questionnaire (SDQ). Further standardised screening measures were employed to screen for anxiety disorders (SCARED), conduct disorder (CBCL), attention deficit-/ hyperactivity disorder (FBBHKS, Conners’ Scale) and depressive disorders (CES-DC). Furthermore, substance abuse and suicidal tendencies were assessed. Health-related quality of life (HRQoL) and health care use were determined. Results Overall, 14.5% of the children and adolescents aged 7–17 fulfilled the criteria for at least one specific mental health problem associated with impairment, or had an overall mental health problem indicated by an abnormal SDQ score and present impairment. However, high comorbidity was found in the children concerned. Symptoms of overall mental health problems were present in 8.6% of the children and 6.6% of the adolescents. This number was reduced to prevalence rates of 6.3 and 4.9% when additional impairment was taken as a criterion. Irrespective of the type of disorder, fewer than half of the children affected were reported as receiving treatment. However, for those suffering from mental health problems, large impairments in HRQoL were observed. Conclusions: The observed prevalence of mental health problems as well as their large impact on well-being and functioning calls for early prevention. This is especially important with regard to the large decrease in HRQoL in the children and adolescents affected.