Screening for bipolar disorder in a tertiary mental health centre using EarlyDetect: A machine learning-based pilot study

Practical pathway for the management of depression in the workplace: a Canadian perspective

Major depressive disorder (MDD) and other mental health issues pose a substantial burden on the workforce. Approximately half a million Canadians will not be at work in any week because of a mental health disorder, and more than twice that number will work at a reduced level of productivity (presenteeism). Although it is important to determine whether work plays a role in a mental health condition, at initial presentation, patients should be diagnosed and treated per appropriate clinical guidelines. However, it is also important for patient care to determine the various causes or triggers including work-related factors. Clearly identifying the stressors associated with the mental health disorder can help clinicians to assess functional limitations, develop an appropriate care plan, and interact more effectively with worker's compensation and disability programs, as well as employers. There is currently no widely accepted tool to definitively identify MDD as work-related, but the presence of certain patient and work characteristics may help. This paper seeks to review the evidence specific to depression in the workplace, and provide practical tips to help clinicians to identify and treat work-related MDD, as well as navigate disability issues.

Identifying neurodevelopmental disabilities from nationalised preschool health check

OBJECTIVE

Models of psychometric screening to identify individuals with neurodevelopmental disabilities (NDDs) have had limited success. In Aotearoa/New Zealand, routine developmental surveillance of preschool children is undertaken using the Before School Check (B4SC), which includes psychometric and physical health screening instruments. This study aimed to determine whether combining multiple screening measures could improve the prediction of NDDs.

METHODS

Linked administrative health data were used to identify NDDs, including attention deficit hyperactivity disorder, autism spectrum disorder and intellectual disability, within a multi-year national cohort of children who undertook the B4SC. Cox proportional hazards models, with different combinations of potential predictors, were used to predict onset of a NDD. Harrell's c-statistic for composite models were compared with a model representing recommended cutoff psychometric scores for referral in New Zealand.

RESULTS

Data were examined for 287,754 children, and NDDs were identified in 10,953 (3.8%). The best-performing composite model combining the Strengths and Difficulties Questionnaire, the Parental Evaluation of Developmental Status, vision screening and biological sex had 'excellent' predictive power (C-statistic: 0.83) compared with existing referral pathways which had 'poor' predictive power (C-statistic: 0.68). In addition, the composite model was able to improve the sensitivity of NDD diagnosis detection by 13% without any reduction in specificity.

CONCLUSIONS

Combination of B4SC screening measures using composite modelling could lead to significantly improved identification of preschool children with NDDs when compared with surveillance that rely on individual psychometric test results alone. This may optimise access to academic, personal and family support for children with NDDs.

Screening for Adulthood ADHD and Comorbidities in a Tertiary Mental Health Center Using EarlyDetect: A Machine Learning-Based Pilot Study

Screening for adult Attention-Deficit/Hyperactivity Disorder (ADHD) and differentiating ADHD from comorbid mental health disorders remains to be clinically challenging. A screening tool for ADHD and comorbid mental health disorders is essential, as most adult ADHD is comorbid with several mental health disorders. The current pilot study enrolled 955 consecutive patients attending a tertiary mental health center in Canada and who completed EarlyDetect assessment, with 45.2% of patients diagnosed with ADHD. The best ADHD classification model using composite scoring achieved a balanced accuracy of 0.788, showing a 2.1% increase compared to standalone ADHD screening, detecting four more patients with ADHD per 100 patients. The classification model including ADHD with comorbidity was also successful (balanced accuracy = 0.712). The results suggest the novel screening method can improve ADHD detection accuracy and inform the risk of ADHD with comorbidity, and may further inform specific comorbidity including MDD and BD.

Single classifier vs. ensemble machine learning approaches for mental health prediction

Early prediction of mental health issues among individuals is paramount for early diagnosis and treatment by mental health professionals. One of the promising approaches to achieving fully automated computer-based approaches for predicting mental health problems is via machine learning. As such, this study aims to empirically evaluate several popular machine learning algorithms in classifying and predicting mental health problems based on a given data set, both from a single classifier approach as well as an ensemble machine learning approach. The data set contains responses to a survey questionnaire that was conducted by Open Sourcing Mental Illness (OSMI). Machine learning algorithms investigated in this study include Logistic Regression, Gradient Boosting, Neural Networks, K-Nearest Neighbours, and Support Vector Machine, as well as an ensemble approach using these algorithms. Comparisons were also made against more recent machine learning approaches, namely Extreme Gradient Boosting and Deep Neural Networks. Overall, Gradient Boosting achieved the highest overall accuracy of 88.80% followed by Neural Networks with 88.00%. This was followed by Extreme Gradient Boosting and Deep Neural Networks at 87.20% and 86.40%, respectively. The ensemble classifier achieved 85.60% while the remaining classifiers achieved between 82.40 and 84.00%. The findings indicate that Gradient Boosting provided the highest classification accuracy for this particular mental health bi-classification prediction task. In general, it was also demonstrated that the prediction results produced by all of the machine learning approaches studied here were able to achieve more than 80% accuracy, thereby indicating a highly promising approach for mental health professionals toward automated clinical diagnosis.

Resting-state functional connectivity does not predict individual differences in the effects of emotion on memory

Emotion-laden events and objects are typically better remembered than neutral ones. This is usually explained by stronger functional coupling in the brain evoked by emotional content. However, most research on this issue has focused on functional connectivity evoked during or after learning. The effect of an individual’s functional connectivity at rest is unknown. Our pre-registered study addresses this issue by analysing a large database, the Cambridge Centre for Ageing and Neuroscience, which includes resting-state data and emotional memory scores from 303 participants aged 18–87 years. We applied regularised regression to select the relevant connections and replicated previous findings that whole-brain resting-state functional connectivity can predict age and intelligence in younger adults. However, whole-brain functional connectivity predicted neither an emotional enhancement effect (i.e., the degree to which emotionally positive or negative events are remembered better than neutral events) nor a positivity bias effect (i.e., the degree to which emotionally positive events are remembered better than negative events), failing to support our pre-registered hypotheses. These results imply a small or no association between individual differences in functional connectivity at rest and emotional memory, and support recent notions that resting-state functional connectivity is not always useful in predicting individual differences in behavioural measures.

Individualized identification of sexual dysfunction of psychiatric patients with machine-learning

Sexual dysfunction (SD) is prevalent in patients with mental health disorders and can significantly impair their quality of life. Early recognition of SD in a clinical setting may help patients and clinicians to optimize treatment options of SD and/or other primary diagnoses taking SD risk into account and may facilitate treatment compliance. SD identification is often overlooked in clinical practice; we seek to explore whether patients with a high risk of SD can be identified at the individual level by assessing known risk factors via a machine learning (ML) model. We assessed 135 subjects referred to a tertiary mental health clinic in a Western Canadian city using health records data, including age, sex, physician’s diagnoses, drug treatment, and the Arizona Sexual Experiences Scale (ASEX). A ML model was fitted to the data, with SD status derived from the ASEX as target outcomes and all other variables as predicting variables. Our ML model was able to identify individual SD cases—achieving a balanced accuracy of 0.736, with a sensitivity of 0.750 and a specificity of 0.721—and identified major depressive disorder and female sex as risk factors, and attention deficit hyperactivity disorder as a potential protective factor. This study highlights the utility of SD screening in a psychiatric clinical setting, demonstrating a proof-of-concept ML approach for SD screening in psychiatric patients, which has marked potential to improve their quality of life.