Predicting Mental Health Treatment Access Among Adolescents With Elevated Depressive Symptoms: Machine Learning Approaches

Author(s):  
Mallory L. Dobias ◽  
Michael B. Sugarman ◽  
Michael C. Mullarkey ◽  
Jessica L. Schleider
2019 ◽  
Author(s):  
Mallory Dobias ◽  
Michael Brandon Sugarman ◽  
Michael C Mullarkey ◽  
Jessica L. Schleider

Objective: A majority of adolescents experiencing depression never access treatment. To increase access to effective mental health care, it is critical to understand factors associated with increased versus decreased odds of adolescent treatment access. This study investigated the relative importance of individual depression symptoms and sociodemographic variables in predicting whether and where adolescents with depression accessed mental health treatments. Method: We performed a pre-registered, secondary analysis of data from the 2017 National Survey of Drug Use and Health (NSDUH), a nationally representative sample of non-institutionalized civilians in the United States. Using four cross-validated random forest models, we predicted whether high-symptom adolescents (N = 1,671; ages 12-17 years) accessed specific mental health treatments in the previous 12 months (“yes/no” for inpatient, outpatient, school, any). Results: 53.38% of adolescents with elevated depressive symptoms accessed treatment of any kind. Even with depressive symptoms and sociodemographic factors included as predictors, pre-registered random forests explained < 0.00% of pseudo out-of-sample deviance in adolescent access to inpatient, outpatient, school, or overall treatments. Exploratory elastic net models explained 0.80 - 2.50% of pseudo out-of-sample deviance in adolescent treatment access across all four treatment types. Conclusions: Neither individual depressive symptoms nor any socioeconomic variables meaningfully predicted specific or overall mental health treatment access in high-symptom adolescents. This study highlights substantial limitations in our capacity to predict whether and where high-symptom adolescents access mental health treatment and underscores the broader need for more accessible, scalable adolescent depression treatments.


Author(s):  
Frances Shaw

This paper situates a discussion of Her within contemporary developments in empathic machine learning for mental health treatment and therapy. Her simultaneously hooks into and critiques a particular imaginary about what artificial intelligence can do when combined with big data. Shaw threads the representation of empathy and artificial intelligence in the film into discussions of contemporary mental health research, in particular possibilities for the automation of treatment, whether through machine learning or guided interventions. Her provides some useful ways to think through utopian, dystopian, and ambivalent readings of such applications of technology in a broader sense, raising questions about sincerity and loss of human connectivity, relational ethics and automated empathy.


2020 ◽  
Vol 45 (6) ◽  
pp. 633-642
Author(s):  
Elizabeth R Wolock ◽  
Alexander H Queen ◽  
Gabriela M Rodríguez ◽  
John R Weisz

Abstract Objective In research with community samples, children with chronic physical illnesses have shown elevated anxiety and depressive symptoms, compared to healthy peers. Less is known about whether physical illnesses are associated with elevated internalizing symptoms even among children referred for mental health treatment—a pattern that would indicate distinctive treatment needs among physically ill children receiving mental health care. We investigated the relationship between chronic physical illness and internalizing symptomatology among children enrolling in outpatient mental health treatment. Method A total of 262 treatment-seeking children ages 7–15 and their caregivers completed a demographic questionnaire, Child Behavior Checklist, and Youth Self-Report during a pre-treatment assessment. Physical illnesses were identified through caregiver report. Results There was no overall association between the presence/absence of chronic physical illness and parent- or child-reported symptoms. However, number of chronic physical illnesses was related to parent- and child-reported affective symptoms. Children with two or more chronic physical illnesses had more severe depressive symptoms than those with fewer physical illnesses. Conclusion Having multiple chronic illnesses may elevate children’s risk of depression symptomatology, even in comparison to other children seeking mental health care. This suggests a need to identify factors that may exacerbate depression symptoms in physically ill children who are initiating therapy and to determine whether different or more intensive services may be helpful for this group. The findings suggest the potential utility of screening for depression in youth with chronic physical illnesses, as well as addressing mental and physical health concerns during treatment.


2011 ◽  
Vol 62 (11) ◽  
pp. 1353-1360 ◽  
Author(s):  
Amber M. Gum ◽  
Lindsay Iser ◽  
Bellinda L. King-Kallimanis ◽  
Andrew Petkus ◽  
Anne DeMuth ◽  
...  

2021 ◽  
Author(s):  
Chantelle A Roulston ◽  
Sarah McKetta ◽  
Maggi Price ◽  
Kathryn Fox ◽  
Jessica L. Schleider

Objective: Many youth with mental health needs cannot access treatment, with multiply-marginalized youth, such as sexual minority youth of Color (SMYoC), experiencing both structural and identity-related barriers to care. The COVID-19 pandemic threatens to exacerbate multi-level treatment access barriers facing SMYoC youth nationwide. However, little large-scale research has examined access to mental health care among SMYoC across the United States, either during or prior to the pandemic. Such work is critical to understanding and ameliorating barriers in this domain. Methods: Using data from adolescents who self-identified as SMYoC and who endorsed a desire for mental health support during the COVID-19 pandemic (N=470, ages 13-16, from 43 U.S. states), we examined associations between state-level, structural factors (income inequality; mental healthcare provider shortage; anti-Black racism; homophobia; and the interaction between anti-Black racism and homophobia) and SMYoC mental health treatment access. Results: Multinomial logistic regressions revealed state-level mental healthcare provider shortage as the only significant predictor of SMYoC reporting they never (versus always) accessed mental health support during the COVID-19 pandemic. SMYoC living in areas with both lower homophobia and lower anti-Black racism were more likely to report always (versus sometimes) accessing mental health treatment. Conclusions: Results highlight the critical importance of considering diverse structural factors and applying an intersectional lens when exploring barriers to mental health treatment among multiply-marginalized youth. In locations where provider shortages are less severe, cultural stigma—including anti-Black racism and homophobia—may still pose challenges for SMYoC in need of mental health care.


Sign in / Sign up

Export Citation Format

Share Document