Dataset Information

Predicting self-harm within six months after initial presentation to youth mental health services: A machine learning study.

ABSTRACT:

Background

A priority for health services is to reduce self-harm in young people. Predicting self-harm is challenging due to their rarity and complexity, however this does not preclude the utility of prediction models to improve decision-making regarding a service response in terms of more detailed assessments and/or intervention. The aim of this study was to predict self-harm within six-months after initial presentation.

Method

The study included 1962 young people (12-30 years) presenting to youth mental health services in Australia. Six machine learning algorithms were trained and tested with ten repeats of ten-fold cross-validation. The net benefit of these models were evaluated using decision curve analysis.

Results

Out of 1962 young people, 320 (16%) engaged in self-harm in the six months after first assessment and 1642 (84%) did not. The top 25% of young people as ranked by mean predicted probability accounted for 51.6% - 56.2% of all who engaged in self-harm. By the top 50%, this increased to 82.1%-84.4%. Models demonstrated fair overall prediction (AUROCs; 0.744-0.755) and calibration which indicates that predicted probabilities were close to the true probabilities (brier scores; 0.185-0.196). The net benefit of these models were positive and superior to the 'treat everyone' strategy. The strongest predictors were (in ranked order); a history of self-harm, age, social and occupational functioning, sex, bipolar disorder, psychosis-like experiences, treatment with antipsychotics, and a history of suicide ideation.

Conclusion

Prediction models for self-harm may have utility to identify a large sub population who would benefit from further assessment and targeted (low intensity) interventions. Such models could enhance health service approaches to identify and reduce self-harm, a considerable source of distress, morbidity, ongoing health care utilisation and mortality.

SUBMITTER: Iorfino F

PROVIDER: S-EPMC7775066 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting self-harm within six months after initial presentation to youth mental health services: A machine learning study.

Iorfino Frank F Ho Nicholas N Carpenter Joanne S JS Cross Shane P SP Davenport Tracey A TA Hermens Daniel F DF Yee Hannah H Nichles Alissa A Zmicerevska Natalia N Guastella Adam A Scott Elizabeth E Hickie Ian B IB

PloS one 20201231 12

<h4>Background</h4>A priority for health services is to reduce self-harm in young people. Predicting self-harm is challenging due to their rarity and complexity, however this does not preclude the utility of prediction models to improve decision-making regarding a service response in terms of more detailed assessments and/or intervention. The aim of this study was to predict self-harm within six-months after initial presentation.<h4>Method</h4>The study included 1962 young people (12-30 years) p ...[more]

PMID: 33382713

Similar Datasets

Project description:BACKGROUND:Adolescence and young adulthood is a vulnerable time during which young people experience many development milestones, as well as an increased incidence of mental illness. During this time, youth also transition between Child and Adolescent Mental Health Services (CAMHS) to Adult Mental Health Services (AMHS). This transition puts many youth at risk of disengagement from service use; however, our understanding of this transition from the perspective of youth is limited. This systematic review aims to provide a more comprehensive understanding of youth experiences of transition from CAMHS to AMHS, through a qualitative thematic synthesis of the extant literature in this area. METHOD:Published and unpublished literature was searched using keywords targeting three subject areas: Transition, Age and Mental Health. Studies were included if they qualitatively explored the perceptions and experiences of youth who received mental health services in both CAMHS and AMHS. There were no limitations on diagnosis or age of youth. Studies examining youth with chronic physical health conditions were excluded. RESULTS:Eighteen studies, representing 14 datasets and the experiences of 253 unique service-users were included. Youth experiences of moving from CAMHS and AMHS are influenced by concurrent life transitions and their individual preferences regarding autonomy and independence. Youth identified preparation, flexible transition timing, individualized transition plans, and informational continuity as positive factors during transition. Youth also valued joint working and relational continuity between CAMHS and AMHS. CONCLUSIONS:Youth experience a dramatic culture shift between CAMHS and AMHS, which can be mitigated by individualized and flexible approaches to transition. Youth have valuable perspectives to guide the intelligent design of mental health services and their perspectives should be used to inform tools to evaluate and incorporate youth perspectives into transitional service improvement. TRIAL REGISTRATION:Clinical Trial or Systematic Review Registry: PROSPERO International Prospective Register of Systematic Reviews CRD42014013799 .

Project description:AIMS:The mental health of youth is continually changing and requires reliable monitoring to ensure that adequate social and economic resources are allocated. This study assessed trends in mental health among Canadian youth, 12-24 years old. Specifically, we examined the prevalence of poor/fair perceived mental health, diagnosis of mood and anxiety disorders, suicidality, perceived stress and sleep problems, substance use, and mental health consultations. METHODS:Data were collected from eight cycles of the annual Canadian Community Health Survey (2011-2018). Prevalence of mental health outcomes was calculated from each survey, and meta-regression was used to assess trends over time. In the absence of a significant trend over time, the eight cycles were pooled together using meta-analysis techniques to gain precision. Trends in prevalence were assessed for the overall sample of youth (12-24 years) and separately for male and female adolescents (12-18 years) and young adults (19-24 years). RESULTS:The prevalence of poor/fair perceived mental health, diagnosed mood and anxiety disorders, and past-year mental health consultations increased from 2011 to 2018, most strongly among young adult females. Past-year suicidality increased among young adult females but did not change for other age and sex groups. Notably, the prevalence of binge drinking decreased by 2.4% per year for young adult males, 1.0% for young adult females and 0.7% per year for adolescent males, while staying relatively stable for adolescent females. Prevalence of cannabis use declined among adolescents before legalisation (2011-2017); however, this trend did not persist in 2018. Instead, the 2018 prevalence was 5.6% higher than the 2017 prevalence (16.3 v. 10.7%). The combined prevalence of other illicit drug use was stable at 4.6%; however, cocaine use and hallucinogens increased by approximately 0.2% per year. CONCLUSIONS:Our findings highlight a growing need for youth mental health services, as indicated by a rise in the prevalence of diagnosed mood and anxiety disorders and past-year mental health consultations. The reason for these observed increases is less apparent - it may represent a true rise in the prevalence of mental illness, or be an artefact of change in diagnostic practices, mental health literacy or diminishing stigma. Nonetheless, the findings indicate a need for the health care system to respond to the rising demand for mental health services among youth.

Project description:BackgroundAdolescence is the peak age of onset for mental illness, with half of all people who will ever have a mental illness experiencing their first episode prior to 18 years of age. Early onset of mental illness is a significant predictor for future episodes. However, adolescents and young adults are less likely than the population as a whole to either seek or receive treatment for a mental illness. The knowledge and attitudes of the adults in an adolescent's life may affect whether or not help is sought, and how quickly. In 2007, the Youth Mental Health First Aid Program was launched in Australia with the aim to teach adults, who work with or care for adolescents, the skills needed to recognise the early signs of mental illness, identify potential mental health-related crises, and assist adolescents to get the help they need as early as possible. This paper provides a description of the program, some initial evaluation and an outline of future directions.MethodsThe program was evaluated in two ways. The first was an uncontrolled trial with 246 adult members of the Australian public, who completed questionnaires immediately before attending the 14 hour course, one month later and six months later. Outcome measures were: recognition of schizophrenia or depression; intention to offer and confidence in offering assistance; stigmatising attitudes; knowledge about adolescent mental health problems and also about the Mental Health First Aid action plan. The second method of evaluation was to track the uptake of the program, including the number of instructors trained across Australia to deliver the course, the number of courses they delivered, and the uptake of the YMHFA Program in other countries.ResultsThe uncontrolled trial found improvements in: recognition of schizophrenia; confidence in offering help; stigmatising attitudes; knowledge about adolescent mental health problems and application of the Mental Health First Aid action plan. Most results were maintained at follow-up. Over the first 3 years of this program, a total of 318 instructors were trained to deliver the course and these instructors have delivered courses to 10,686 people across all states and territories in Australia. The program has also spread to Canada, Singapore and England, and will spread to Hong Kong, Sweden and China in the near future.ConclusionsInitial evaluation suggests that the Youth Mental Health First Aid course improves participants' knowledge, attitudes and helping behaviour. The program has spread successfully both nationally and internationally.Trial registrationACTRN12609000033246.

Dataset Information

Predicting self-harm within six months after initial presentation to youth mental health services: A machine learning study.

Background

Method

Results

Conclusion

Publications

Predicting self-harm within six months after initial presentation to youth mental health services: A machine learning study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets