Search

Background: This paper aims to synthesise the literature on machine learning (ML) and big data applications for mental health, highlighting current research and applications in practice., Methods: We employed a scoping review methodology to rapidly map the field of ML in mental health. Eight health and information technology research databases were searched for papers covering this domain. Articles were assessed by two reviewers, and data were extracted on the article's mental health application, ML technique, data type, and study results. Articles were then synthesised via narrative review., Results: Three hundred papers focusing on the application of ML to mental health were identified. Four main application domains emerged in the literature, including: (i) detection and diagnosis; (ii) prognosis, treatment and support; (iii) public health, and; (iv) research and clinical administration. The most common mental health conditions addressed included depression, schizophrenia, and Alzheimer's disease. ML techniques used included support vector machines, decision trees, neural networks, latent Dirichlet allocation, and clustering., Conclusions: Overall, the application of ML to mental health has demonstrated a range of benefits across the areas of diagnosis, treatment and support, research, and clinical administration. With the majority of studies identified focusing on the detection and diagnosis of mental health conditions, it is evident that there is significant room for the application of ML to other areas of psychology and mental health. The challenges of using ML techniques are discussed, as well as opportunities to improve and advance the field.

Objective: There has been interest in the antecedents and mental health impacts of eating and body image disturbances in adolescence. Less is known about longer‐term mental health impacts into young adulthood, as longitudinal studies with data spanning this developmental period are rare. We capitalize on mental health data collected across adolescence and young adulthood from a population‐based cohort study that has been following >2000 Australian children and their families from infancy to young adulthood. Method: This sample comprised 1,568 participants who completed the Eating Disorder Inventory drive for thinness and bulimic behavior (the severity of binge‐purge patterns) subscales, and a modified version of the body dissatisfaction subscale in mid‐adolescence (15–16 years), or the Depression Anxiety Stress Scales in young adulthood (19–20, 23–24, and 27–28 years). Results: After adjusting for baseline demographic and prior mental health factors (<13 years of age), all three indices of eating and body image disturbances in adolescence predicted each mental health outcome in young adulthood. Mental health risks associated with adolescent body dissatisfaction and bulimic behavior scores remained stable across young adulthood, with men having more pronounced problems associated with bulimic behavior scores than women. In contrast, mental health risks associated with adolescent drive for thinness scores diminished across this period similarly for men and women. Discussion: Findings suggest that adolescent eating and body image disturbances may have long‐term mental health impacts that extend into young adulthood. This underscores the need for early preventative intervention, and longer‐term monitoring and support for body image and eating disturbances. [ABSTRACT FROM AUTHOR]

Postpartum depression (PPD) is a significant mental health issue in mothers and fathers alike; yet at-risk fathers often come to the attention of health care professionals late due to low awareness of symptoms and reluctance to seek help. This study aimed to examine whether passive social media markers are effective for identifying fathers at risk of PPD. We collected 67,796 Reddit posts from 365 fathers, spanning a 6-month period around the birth of their child. A list of "at-risk" words was developed in collaboration with a perinatal mental health expert. PPD was assessed by evaluating the change in fathers' use of words indicating depressive symptomatology after childbirth. Predictive models were developed as a series of support vector machine classifiers using behavior, emotion, linguistic style, and discussion topics as features. The performance of these classifiers indicates that fathers at risk of PPD can be predicted from their prepartum data alone. Overall, the best performing model used discussion topic features only with a recall score of 0.82. These findings could assist in the development of support and intervention tools for fathers during the prepartum period, with specific applicability to personalized and preventative support tools for at-risk fathers. [ABSTRACT FROM AUTHOR]

Abstract: Background and Aims: Studies have linked adolescent alcohol use with adverse consequences in adulthood, yet it is unclear how strong the associations are and to what extent they may be due to confounding. Our aim was to estimate the strength of association between different patterns of adolescent drinking and longer‐term psychosocial harms taking into account individual, family and peer factors. Design: Participant‐level data were integrated from four long‐running longitudinal studies: Australian Temperament Project, Christchurch Health and Development Study, Mater Hospital and University of Queensland Study of Pregnancy and Victorian Adolescent Health Cohort Study. Setting: Australia and New Zealand. Participants: Participants were assessed on multiple occasions between ages 13 and 30 years (from 1991 to 2012). Number of participants varied (up to n = 9453) by analysis. Measurements: Three patterns of alcohol use (frequent, heavy episodic and problem drinking) were assessed prior to age 17. Thirty outcomes were assessed to age 30 spanning substance use and related problems, antisocial behaviour, sexual risk‐taking, accidents, socio‐economic functioning, mental health and partner relationships. Findings: After covariate adjustment, weekly drinking prior to age 17 was associated with a two‐ to threefold increase in the odds of binge drinking [odds ratio (OR) = 2.14; 95% confidence interval (CI) = 1.57–2.90], drink driving (OR = 2.78; 95% CI = 1.84–4.19), alcohol‐related problems (OR = 3.04; 95% CI = 1.90–4.84) and alcohol dependence (OR = 3.30; 95% CI = 1.69–6.47) in adulthood. Frequency of drinking accounted for a greater proportion of the rate of most adverse outcomes than the other measures of alcohol use. Associations between frequent, heavy episodic and problem drinking in adolescence and most non‐alcohol outcomes were largely explained by shared risk factors for adolescent alcohol use and poor psychosocial functioning. Conclusions: Frequency of adolescent drinking predicts substance use problems in adulthood as much as, and possibly more than, heavy episodic and problem drinking independent of individual, family and peer predictors of those outcomes. [ABSTRACT FROM AUTHOR]

Introduction and Aims: The extent to which young adult former cannabis users fare better than infrequent users is unclear. We investigated the association between cannabis use status at age 23 and substance use and mental health outcomes at age 27.Design and Methods: Data were from the 20+ year cohort of the PATH Through Life Study. Lifetime cannabis users (n = 1410) at age 23 were classified as former/occasional/regular users. Multivariable logistic regression was used to estimate the association between cannabis use status at age 23 and six outcomes assessed at age 27.Results: Compared with occasional cannabis users: (i) former users had odds of subsequent tobacco use [odds ratio (OR) = 0.67, 95% confidence interval (CI) 0.52-0.85], illicit drug use (cannabis, OR = 0.22, 95% CI 0.17-0.28; other illicit drugs, OR = 0.29, 95% CI 0.22-0.39) and mental health impairment (OR = 0.71, 95% CI 0.55-0.92) that were 29-78% lower; and (ii) regular users had odds of subsequent frequent alcohol use (OR = 2.34, 95% CI 0.67-1.34), tobacco use (OR = 3.67, 95% CI 2.54-5.30), cannabis use (OR = 11.73, 95% CI 6.81-20.21) and dependence symptoms (OR = 12.60, 95% CI 8.38-18.94), and other illicit drug use (OR = 2.95, 95% CI 2.07-4.21) that were 2-13 times greater. Associations attenuated after covariate adjustment, and most remained significant.Discussion and Conclusions: Clear associations exist between cannabis use status in young adulthood and subsequent mental health and substance use. While early intervention remains important to prevent regular cannabis use and the associated harms, experimentation with cannabis use in the years leading into young adulthood may not necessarily determine an immutable pathway to mental health problems and illicit substance use. [Silins E, Swift W, Slade T, Toson B, Rodgers B, Hutchinson DM. A prospective study of the substance use and mental health outcomes of young adult former and current cannabis users. Drug Alcohol Rev 2017;00:000-000]. [ABSTRACT FROM AUTHOR]