A Prototype Software Framework for Transferable Computational Health Economic Models and Its Early Application in Youth Mental Health.

We are developing an economic model to explore multiple topics in Australian youth mental health policy. To help make that model more readily transferable to other jurisdictions, we developed a software framework for authoring modular computational health economic models (CHEMs) (the software files that implement health economic models). We specified framework user requirements for: a simple programming syntax; a template CHEM module; tools for authoring new CHEM modules; search tools for finding existing CHEM modules; tools for supplying CHEM modules with data; reproducible analysis and reporting tools; and tools to help maintain a CHEM project website. We implemented the framework as six development version code libraries in the programming language R that integrate with online services for software development and research data archiving. We used the framework to author five development version R libraries of CHEM modules focussed on utility mapping in youth mental health. These modules provide tools for variable validation, dataset description, multi-attribute instrument scoring, construction of mapping models, reporting of mapping studies and making out of sample predictions. We assessed these CHEM module libraries as mostly meeting transparency, reusability and updatability criteria that we have previously developed, but requiring more detailed documentation and unit testing of individual modules. Our software framework has potential value as a prototype for future tools to support the development of transferable CHEMs.Code: Visit https://www.ready4-dev.com for more information about how to find, install and apply the prototype software framework.

An overview of clustering methods with guidelines for application in mental health research.

Cluster analyzes have been widely used in mental health research to decompose inter-individual heterogeneity by identifying more homogeneous subgroups of individuals. However, despite advances in new algorithms and increasing popularity, there is little guidance on model choice, analytical framework and reporting requirements. In this paper, we aimed to address this gap by introducing the philosophy, design, advantages/disadvantages and implementation of major algorithms that are particularly relevant in mental health research. Extensions of basic models, such as kernel methods, deep learning, semi-supervised clustering, and clustering ensembles are subsequently introduced. How to choose algorithms to address common issues as well as methods for pre-clustering data processing, clustering evaluation and validation are then discussed. Importantly, we also provide general guidance on clustering workflow and reporting requirements. To facilitate the implementation of different algorithms, we provide information on R functions and libraries.

Accelerating youth mental health services research.

OBJECTIVE: We describe a research program to advance youth mental health service research in Australia, addressing two core knowledge gaps: the lack of available routine outcome measures and lack of understanding of how to assess and monitor complexity and heterogeneity in illness presentation and trajectory. CONCLUSIONS: Our research identifies better routine outcome measures (ROM) that are: designed specifically for the developmental nuances of the 12-25-year age range; multidimensional; and meaningful to young people, their carers, and service providers. Alongside much-needed new measures of complexity and heterogeneity, these tools will inform service providers to better meet the needs of young people presenting with mental health problems.

Understanding the complexity, patterns, and correlates of alcohol and other substance use among young people seeking help for mental ill-health.

PURPOSE: Use of alcohol and other substances is a multifaceted issue impacting young people across multiple life domains. This paper aims to elucidate patterns of substance use and associated demographic and clinical factors among young people seeking treatment for their mental health. METHODS: Young people (12-25 years old) were recruited from five youth-specific primary mental health ("headspace") services in Australia. Self-reported substance use and harms in the past 3 months were measured using WHO-ASSIST. Network analyses were conducted to evaluate interrelationships between use and harms associated with different substances. Subgroups were then identified based on whether participants reported using high centrality substances, and associated demographic and clinical factors were assessed with multinomial logistic regression. RESULTS: 1107 youth participated. 70% reported use of at least one substance in the past 3 months, with around 30% of those reporting related health, social, legal or financial problems. Network analysis highlighted substantial interconnections between use and harm indicators for all substances, with amphetamine-type stimulants (ATS) and cannabis being high central substances. Higher levels of substance use and harms were reported in subgroups with ATS or cannabis use and different risk factors were associated with these subgroups. CONCLUSIONS: Findings highlight the importance of screening for substance use in youth primary mental healthcare settings, offering a key opportunity for early intervention. Clinicians should be aware of the inner connections of use and harms of different drugs and the role of cannabis and amphetamine use as a marker for more substance use profiles.

Prevalence and Correlates of Tobacco Use in Young People Presenting to Australian Primary Mental Health Services.

INTRODUCTION: In Australian youth primary mental health settings it is unclear as to the rates and correlates of tobacco use at service entry. AIMS AND METHODS: We aimed to delineate the prevalence and correlates of recent tobacco use (eg, cigarettes, chewing tobacco, cigars, etc) in the past 3 months in young people at their first presentation to primary mental health services as a function of age. Cross-sectional self-report measures were collected using a tablet device from young people presenting to one of five Australian primary mental health (headspace) services. Logistic regression assessed correlates of past 3-month tobacco use in adolescents (12-17 years) and young adults (18-25 years). RESULTS: Regular (at least monthly) tobacco use in the past 3 months was found in 23.4% (n = 247, N = 1055) of the sample. Increasing age (odds ratio [OR] =1.47 per year; 95% confidence interval [CI]: 1.15 to 1.89), male sex (OR = 1.98; 95% CI: 1.02 to 3.83), being in a relationship (OR = 1.96; 95% CI: 1.01 to 3.82), and poorer functioning (OR = 0.95 per unit Social and Occupational Functioning Assessment Scale increase; 95% CI: 0.91 to 0.99) predicted regular tobacco use in adolescents, but not in young adults. Living in a regional location (OR = 2.10; 95% CI: 1.40 to 3.13) and not studying (OR = 0.47; 95% CI: 0.31 to 0.73) predicted tobacco use in young adults. Having a diagnosed mental illness other than depression and/or anxiety predicted tobacco use in both groups (adolescents OR = 2.49; 95% CI: 1.26 to 4.94; young adults OR = 1.80; 95% CI: 1.13 to 2.89). CONCLUSIONS: Nearly a quarter of young people with mental illness are using tobacco, supporting the need for early intervention approaches. Adapting treatment targets by age could improve the impact of interventions in adolescents versus young adults. Poor functioning and lack of engagement in education were associated with tobacco use in both age groups, respectively; however, more research is needed to determine the direction of these relationships. IMPLICATIONS: Young people with mental illness have a high prevalence of recent tobacco use and this is evident when they first present to youth primary mental health services. Youth-oriented mental health settings may provide a unique window for tobacco use prevention and early intervention to reduce smoking in people with mental illness, a priority population. Age-specific targeted approaches might be needed in adolescents and young adults.

The Neuropsychological Symptoms Self-Report: psychometric properties in an adolescent and young adult mental health cohort.

BACKGROUND: Subjective cognitive symptoms are common in young people receiving mental health treatment and are associated with poorer outcomes. The aim of this study was to determine the psychometric properties of the Neuropsychological Symptoms Self-Report (NSSR), an eight-item measure recently developed to provide a snapshot of young people's perceived change in cognitive functioning in relation to mental health treatment. METHOD: The sample included 633 youth aged 12-25 years (Mage = 18.2, 66.5% female, 88.6% Australian-born) who had sought mental health treatment in primary headspace services. At three-month follow-up, participants completed the NSSR and self-report measures of depression and anxiety. RESULTS: Excellent internal consistency was found: Cronbach's alpha = 0.93. The NSSR had negative correlations with self-reported anxiety (r = -.33, p < .001) and depression (r = -.48, p < .001) symptoms, suggesting a link with affective symptoms, but still independence of constructs. Exploratory and confirmatory factor analyses supported a single-factor model. Item response theory (IRT) analysis suggested good model fit (homogeneity, data integrity, scalability, local independence and monotonicity) for all items. There was some evidence of measurement noninvariance (for item thresholds) by sex and age, but not diagnosis. IRT models also supported briefer six- and three-item versions of the NSSR. CONCLUSION: In busy clinical practice, clinicians need a rapid and reliable method for determining whether cognitive symptoms are of concern and in need of further assessment and treatment. Study findings support the NSSR as a brief, psychometrically sound measure for assessing subjective cognitive functioning in adolescents and young adults receiving mental health treatment.

Not in education, employment and training status in the early stages of bipolar I disorder with psychotic features.

OBJECTIVE: There is a lack of existing research regarding young people with bipolar I disorder (BD-I) and psychotic features, who are not in education, employment, and training (NEET). Thus, the aims of the study were to: (a) establish rates of NEET at service entry to a specialist early intervention service; (b) delineate premorbid and current variables associated with NEET status at service entry and (c) examine correlates of NEET status at discharge. METHOD: Medical file audit methodology was utilized to collect information on 118 patients with first episode psychotic mania treated at the Early Psychosis Prevention and Intervention Centre (EPPIC), Melbourne, Australia. NEET status was determined using the modified vocation status index (MVCI). Bivariate and multivariable logistic variables were used to examine relationships between premorbid, service entry and treatment variables, and NEET status at service entry and discharge. RESULTS: The NEET rate was 33.9% at service entry, and 39.2% at discharge. Variables associated with NEET status at service entry were premorbid functioning and polysubstance use. NEET status at service entry was the only significant correlate of NEET status at discharge. When service entry NEET was taken out of the model, substance use during treatment was predictive of NEET status at discharge. CONCLUSIONS: NEET status at service entry was related to a history of premorbid decline, and risk factors such as substance use and forensic issues. NEET status can decline during treatment, and utility of vocational intervention programs specifically for BD, in addition to specialist early intervention, needs to be examined.

Validation of a brief tool to assess and monitor suicidal ideation: The Youth Suicide Ideation Screen (YSIS-3).

BACKGROUND: . Suicidal ideation is common in young people. Assessment and monitoring can help to understand its dynamic nature and inform treatment planning. AIMS: . The aims were to evaluate psychometric properties of a brief screening tool - the Youth Suicide Ideation Screen-3 item (YSIS-3) and determine its correlation with the Suicidal Ideation Question-Junior (SIQ-JR) and measures of depressive and anxiety symptoms a. METHOD: . Cronbach's alpha (α) was used to evaluate internal consistency of the YSIS-3 in 1107 young people. Convergent validity was evaluated using network analysis based on individual item polychoric correlations. Structural equation modelling was used to understand latent constructs; pearson product moment correlations to validate the integrity of the underlying construct. and two-way mixed Intraclass Correlation Coefficient to examine longitudinal stability and Cohen's d to assess sensitivity to change. RESULTS: . The YSIS-3 was found to have internal consistency, convergent validity and criteria validity almost identical to the 15-item SIQ-JR. There was a high correlation (0.91) between the latent factor underpinning YSIS-3 and SIQ-JR. The YSIS-3 was more sensitive to change among those presenting with suicidal ideation at baseline. CONCLUSIONS: . The YSIS-3 is a valid measure for the assessment and monitoring of suicidal ideation, and is sensitive to change. This brief measure can contribute to regular monitoring of the presence and nature of suicidal ideation, which when included in comprehensive assessment, will inform the clinical management of those at risk of suicide.

Subjective cognitive functioning in relation to changes in levels of depression and anxiety in youth over 3 months of treatment.

BACKGROUND: Subjective cognitive difficulties are common in mental illness and have a negative impact on role functioning. Little is understood about subjective cognition and the longitudinal relationship with depression and anxiety symptoms in young people. AIMS: To examine the relationship between changes in levels of depression and anxiety and changes in subjective cognitive functioning over 3 months in help-seeking youth. METHOD: This was a cohort study of 656 youth aged 12-25 years attending Australian headspace primary mental health services. Subjective changes in cognitive functioning (rated as better, same, worse) reported after 3 months of treatment was assessed using the Neuropsychological Symptom Self-Report. Multivariate multinomial logistic regression analysis was conducted to evaluate the impact of baseline levels of and changes in depression (nine-item Patient Health Questionnaire; PHQ9) and anxiety symptoms (seven-item Generalised Anxiety Disorder scale; GAD7) on changes in subjective cognitive function at follow-up while controlling for covariates. RESULTS: With a one-point reduction in PHQ9 at follow-up, there was an estimated 11-18% increase in ratings of better subjective cognitive functioning at follow-up, relative to stable cognitive functioning. A one-point increase in PHQ9 from baseline to follow-up was associated with 7-14% increase in ratings of worse subjective cognitive functioning over 3 months, relative to stable cognitive functioning. A similar attenuated pattern of findings was observed for the GAD7. CONCLUSIONS: A clear association exists between subjective cognitive functioning outcomes and changes in self-reported severity of affective symptoms in young people over the first 3 months of treatment. Understanding the timing and mechanisms of these associations is needed to tailor treatment.

Developing and testing the F-SIM, a measure of social inclusion for people with mental illness.

Social inclusion is an important contributor to good mental health and greater mental health outcomes for people with psychiatric disorders. A psychometrically-sound measure of social inclusion is required to facilitate progress in this area. The aim here was to report on preliminary findings from a novel, user-friendly measure of social inclusion that comprehensively assesses the construct. Preliminary testing of the Filia Social Inclusion Measure (F-SIM) was conducted with ninety participants (30 consumers; 30 family members/carers; 30 community members). Participants completed the self-report measure and a usability questionnaire. Preliminary findings demonstrated poorer social inclusion for people with mental illness compared to those without, with differences seen in each of five domains (housing and services, social functioning, occupational functioning, finances and health). Differences were also seen regarding family members or carers, with consistently poorer social inclusion than general community members observed. Participants reported the F-SIM as easy to use, and considered it to measure social inclusion well, indicating good face validity. The F-SIM demonstrates an ability to differentiate between groups. Implications for use and suggestions for future research are detailed. Following further psychometric assessment, the F-SIM will have wide applicability in clinical and research settings.

Social inclusion and its interrelationships with social cognition and social functioning in first-episode psychosis.

AIM: People with psychosis are at risk of social exclusion. Research is needed in this area due to the lack of direct measurement of social inclusion, which becomes salient in adolescence and is relevant to first-episode psychosis (FEP; the onset of which typically occurs during or shortly after adolescence). Social inclusion may be impacted by impaired social cognition and social functioning, which are related features observed in psychosis. The aim of this study was to explore interrelationship(s) between social cognition, social functioning and social inclusion in FEP while controlling for symptomatology (positive, negative and depressive symptoms) and demographic characteristics. METHODS: A series of cross-sectional hierarchical multiple regressions were conducted to examine whether: social cognition (theory of mind, emotion recognition) predicted social functioning; social functioning predicted social inclusion, and whether social functioning mediated the relationship between social cognition and social inclusion in people aged 15 to 25 (M = 20.49, SD = 2.41) with FEP (N = 146). Age, sex, premorbid IQ, positive and negative psychotic symptoms and depression were control variables. RESULTS: Poor facial emotion recognition (ß = -.22, P < .05) and negative symptoms (ß = -.45, P < .001) predicted lower social functioning. Role-specific social functioning (ie, current employment) predicted greater social inclusion (ß = .17, P < .05). Higher depression symptomatology predicted lower social inclusion (ß = -.43, P < .001). Social functioning did not mediate the relationship between social cognition and inclusion. Psychotic symptoms were unrelated to social inclusion. CONCLUSIONS: Employment and depression may influence social inclusion somewhat independently of psychotic symptomatology in FEP. Inferences should be viewed with caution given this study did not involve longitudinal data.

What is social inclusion? A thematic analysis of professional opinion.

OBJECTIVE: Social inclusion is increasingly recognized as an important contributor to positive mental health outcomes, particularly for people with mental illness. There is a lack of consensus regarding what it means to be socially included and what the key contributors to social inclusion may be. The aim of this investigation was to determine such key contributors, as identified by those with professional experience. METHOD: A thematic analysis of literature regarding social inclusion was conducted to obtain the opinions of professionals regarding key contributors of social inclusion. Seventy-one pieces of literature were reviewed: peer-reviewed literature (academic literature regarding social inclusion in general [n = 25] and social inclusion and mental illness [n = 26]), and gray literature (organizational reports [n = 20]). Within- and between-groups analyses were performed to determine group differences and increase understanding of which contributors were deemed important consistently across groups. RESULTS: A comprehensive list of 90 contributors to social inclusion and exclusion was compiled, categorized into 13 domains based on commonalities. Contributors related to employment and education, housing and neighborhood, and social activities and support were highly cited. Differences were observed between-groups regarding specificity of contributors, with organizational reports reporting more detailed contributors. CONCLUSIONS AND IMPLICATIONS FOR PRACTICE: A comprehensive and specific understanding of social inclusion was obtained. This improved understanding will allow for better measurement of social inclusion which will assist in evaluating programs and interventions, identifying areas of greatest need, and in planning services, policy and strategies to target specific contributors proven to improve social inclusion and subsequent mental health outcomes. (PsycINFO Database Record

Men's perceived barriers to help seeking for depression: Longitudinal findings relative to symptom onset and duration.

Men's help seeking for depression continues to gain focussed research and clinical attention. In this study, 125 men ( M = 39.02 years) provided data on perceived barriers to mental health help seeking, and self-reported depression at baseline, and 15 weeks. Longitudinal depression caseness was used to investigate group differences in perceived barriers to help seeking. Those experiencing unremitting depression reported the highest perceived help-seeking barriers. This finding was consistent over all domains of help-seeking barriers, and it was independent of previous mental health help-seeking efforts. Findings are discussed in the broader context of men's mental health and health psychology.

Australian Schizophrenia Care and Assessment Programme: real-world schizophrenia: outcomes.

OBJECTIVE: It has been increasingly recognized that there is need to assess patient outcomes in schizophrenia across a broad range of dimensions. But few studies have attempted to do this in clinical populations and no systematic study has broadly assessed outcomes in schizophrenia in Australia using a longitudinal design. Thus, a real-world study, the Schizophrenia Care and Assessment Programme (SCAP), was structured to collect comprehensive information over time to inform policy debate and extend current knowledge about the course of schizophrenia in an Australian context. METHODS: A cohort of 347 patients with schizophrenia was followed up over 3 years. Clinical outcomes, occupational and psychosocial functioning and quality of life were assessed at 6 monthly intervals, and resource utilization and costing data were collected continuously from internal and external databases as well as from participants directly. RESULTS: The participants as a group experienced an overall decline in positive and negative symptoms of schizophrenia, a reduction in general psychopathology and a reduction in severity of depression. There was an improvement in functioning, a reduction in mental health-related disability and an improvement in patient- and observer-rated quality of life. Change of severity within the variously assessed domains over time appeared to be relatively independent. CONCLUSIONS: In the present sample of schizophrenia patients treatment was associated with positive health outcomes; but outcomes across assessment domains did not closely correlate across time. The scrutiny of a broad range of patient outcomes will assist with the assessment of new treatment modalities and with service planning.

Australian Schizophrenia Care and Assessment Programme: real-world schizophrenia: economics.

OBJECTIVE: The treatment of patients with schizophrenia consumes a considerable proportion of health service budgets, yet there have been few attempts to prospectively analyse the costs associated with this condition. Amid the current debate about where to invest scarce treatment resources to achieve optimal outcomes, real-world studies, such as the Schizophrenia Care and Assessment Programme (SCAP) contrast with hypothetically based models and provide comprehensive and broad-ranging data. METHOD: Direct health-care costs were prospectively studied in a cohort of 347 patients with schizophrenia in Dandenong, Australia over 3 years. Indirect costs were estimated from patient self-reported information. RESULTS: The average annual societal cost was AU $32,160 per participant in the first year of the study, AU $27,190 in the second year and AU $29,181 in the third year. Indirect costs accounted for 46% of the total costs in the first year, 52% of the total costs in the second year and 50% of the total costs in the third year. The most expensive component of treatment was inpatient hospital care, which accounted for 42%, 34% and 36% of the total costs in the first, second and third year, respectively. CONCLUSIONS: Considerable resources are required for the provision of treatment for patients with schizophrenia. But for the majority of people in this cohort, funding assertive treatment programmes and measures to reduce hospitalization was accompanied with enhanced functioning and quality of life, as well as a reduction in long-term societal and government costs. The distribution of health-care costs is highly skewed, with a relatively small proportion of patients (39%) consuming the majority of resources (80%). Improving rates of employment for this patient group could hold substantial benefits in reducing the overall economic and personal impact of this disorder.