C. Barr Taylor’s research while affiliated with Palo Alto University and other places

Importance: Given the rising prevalence of anxiety, depression, and eating disorders among college students, optimizing population treatment outcomes for the college population is needed. Objective: To validate a machine learning model using baseline sociodemographic and clinical information to predict which college students would show long-term benefits from transdiagnostic guided self-help (GSH).Design: Full populations of college students were screened, and those over 18 and with or at risk for anxiety, depression, or eating disorders were randomized to GSH or control. In a secondary analysis, we used baseline variables to predict those who did (n=1,380 students) or did not (n=1,723) remit from any anxiety, depression, and eating disorder at two-year follow-up following GSH. Setting: The trial was conducted across 26 U.S. colleges and universities between October 2019 and December 2023.Participants: Those with or at risk of a DSM-5 diagnosis of panic disorder (PD), generalized anxiety disorder (GAD), social anxiety disorder (SAD), depression, and/or eating disorders (EDs) (excluding anorexia nervosa) were enrolled. Intervention(s): Participants received transdiagnostic cognitive-behavioral GSH via SilverCloud Health. Main Outcome(s) and Measure(s): Discriminatory accuracy of a Super Learner algorithm trained for the binary classification of prevention and remission of all (GAD, SAD, PD, depression, and EDs) disorders at a 2-year follow-up based on baseline features.Results: Of 3,103 participants within GSH (Mage = 20.2; SD ± 4.0), 70.4% identified as female and 64.6% as white. A Super Learner algorithm achieved acceptable discriminatory accuracy with an AUC of .77 (95% CI = .74-.79) to distinguish the probability of 2-year prevention and remission from depression, anxiety, and eating disorders (threshold, .485; accuracy, .72, sensitivity, .71; specificity, .72; PPV, .68; NPV, .75). Higher baseline severity in depression, anxiety, and binge eating were associated with an increased probability of non-remission, whereas those who identified as male and with better mental health quality of life were associated with an increased probability of prevention and remission. Conclusion and relevance: Routinely collected pre-treatment data may identify 2-year outcome variability from GSH, thereby potentially facilitating population-level mental health care delivery on college campuses.Trial Registration: ClinicalTrials.gov: NCT04162847

The current pilot study examines engagement with and preliminary effectiveness of an mHealth intervention designed for teens with eating disorders (EDs) to delineate specific user characteristics associated with intervention engagement and the impact of this engagement on ED symptoms. Teens 14–17 years old with or at high-risk for an ED were recruited from social media (n=29) and provided access to an mHealth intervention for 2 months. At baseline, participants were surveyed on ED and other mental health symptoms and demographics. Bivariate analyses were used to examine associations between baseline characteristics and time spent in the app (<10 vs. ≥ 10 minutes). Qualitative feedback from participants on intervention content and usability was also collected and reported. Out of the 29 participants, 22 (76%) utilized the app at least once after gaining access. The median number of logins for these users was 6, with an interquartile range spanning from 3 to 15. Over half of teens spent 10 minutes or more engaging with the app during the study period (n=15, 52%). Compared to those who spent less than 10 minutes with the app, those who spent more than 10 minutes engaging with the app were slightly younger, more likely to endorse less chronic ED symptoms, and less likely to report social anxiety disorder (ps < 0.05). Teens’ distinct user characteristics impact rates of uptake and engagement with an ED-focused mHealth intervention and should be considered in the design and iteration of these tools. mHealth tools have the potential to improve ED recovery outcomes among teens, and future studies should further evaluate the effectiveness of these tools and integration of content to support severe ED symptoms and other comorbid mental health issues.

Background A better understanding of the relationships between insomnia and anxiety, mood, eating, and alcohol-use disorders is needed given its prevalence among young adults. Supervised machine learning provides the ability to evaluate which mental disorder is most associated with heightened insomnia among U.S. college students. Combined with Bayesian network analysis, probable directional relationships between insomnia and interacting symptoms may be illuminated. Methods The current exploratory analyses utilized a national sample of college students across 26 U.S. colleges and universities collected during population-level screening before entering a randomized controlled trial. We used a 4-step statistical approach: (1) at the disorder level, an elastic net regularization model examined the relative importance of the association between insomnia and 7 mental disorders (major depressive disorder, generalized anxiety disorder, social anxiety disorder, panic disorder, post-traumatic stress disorder, anorexia nervosa, and alcohol use disorder); (2) This model was evaluated within a hold-out sample. (3) at the symptom level, a completed partially directed acyclic graph (CPDAG) was computed via a Bayesian hill-climbing algorithm to estimate potential directionality among insomnia and its most associated disorder [based on SHAP (SHapley Additive exPlanations) values)]; (4) the CPDAG was then tested for generalizability by assessing (in)equality within a hold-out sample using structural hamming distance (SHD). Results Of 31,285 participants, 20,597 were women (65.8%); mean (standard deviation) age was 22.96 (4.52) years. The elastic net model demonstrated clinical significance in predicting insomnia severity in the training sample [R² = .44 (.01); RMSE = 5.00 (0.08)], with comparable performance in the hold-out sample (R² = .33; RMSE = 5.47). SHAP values indicated that the presence of any mental disorder was associated with higher insomnia scores, with major depressive disorder as the most important disorder associated with heightened insomnia (mean |SHAP|= 3.18). The training CPDAG and hold-out CPDAG (SHD = 7) suggested depression symptoms presupposed insomnia with depressed mood, fatigue, and self-esteem as key parent nodes. Conclusion These findings provide insights into the associations between insomnia and mental disorders among college students and warrant further investigation into the potential direction of causality between insomnia and depression. Trial registration Trial was registered on the National Institute of Health RePORTER website (R01MH115128 || 23/08/2018).

BACKGROUND: Previous studies showed that comorbidity and demographic factors added to burden on health-related quality of life (HRQoL). Only one study explored the relationship between HRQoL and comorbidity in college students with mental disorders, leaving generalizability of findings uncertain. Less is known about the association of demographics on HRQoL. This study investigated HRQoL based on demographics and comorbidity among college students with mental disorders. METHODS: Participants were students (N=5,535) across 26 U.S. colleges and universities who met criteria for depression, generalized anxiety, panic, social anxiety, post-traumatic stress, or eating disorders based on self-report measures. ANOVA and linear regressions were conducted. RESULTS: Overall, female, minoritized (gender, sexual orientation, race, or ethnicity), and lower socioeconomic status students reported lower HRQoL than male, heterosexual, White, non-Hispanic, and higher socioeconomic status peers. After accounting for comorbidity, differences in physical HRQoL based on sex assigned at birth and gender were no longer significant. For mental HRQoL, only gender and sexual orientation remained significant. A greater number of comorbidities was associated with lower HRQoL regardless of demographic group. LIMITATIONS: The non-experimental design limits causal inference. The study focused on univariable associations without examining potential interactions between demographic factors. Future research should explore structural factors like discrimination. CONCLUSION: Results suggested that increased comorbidities placed an additional burden on HRQoL and that certain demographic groups were more vulnerable to HRQoL impairment among students with mental disorders. Findings suggest the need for prevention of disorders and their comorbidity and implementing tailored interventions for specific student subgroups with increased vulnerability.

Objective Few individuals with eating disorders (EDs) receive treatment. Innovations are needed to identify individuals with EDs and address care barriers. We developed a chatbot for promoting services uptake that could be paired with online screening. However, it is not yet known which components drive effects. This study estimated individual and combined contributions of four chatbot components on mental health services use (primary), chatbot helpfulness, and attitudes toward changing eating/shape/weight concerns (“change attitudes,” with higher scores indicating greater importance/readiness). Method s Two hundred five individuals screening with an ED but not in treatment were randomized in an optimization randomized controlled trial to receive up to four chatbot components: psychoeducation, motivational interviewing, personalized service recommendations, and repeated administration (follow‐up check‐ins/reminders). Assessments were at baseline and 2, 6, and 14 weeks. Results Participants who received repeated administration were more likely to report mental health services use, with no significant effects of other components on services use. Repeated administration slowed the decline in change attitudes participants experienced over time. Participants who received motivational interviewing found the chatbot more helpful, but this component was also associated with larger declines in change attitudes. Participants who received personalized recommendations found the chatbot more helpful, and receiving this component on its own was associated with the most favorable change attitude time trend. Psychoeducation showed no effects. Discussion Results indicated important effects of components on outcomes; findings will be used to finalize decision making about the optimized intervention package. The chatbot shows high potential for addressing the treatment gap for EDs.

Highlights In response to a comment on our recent article, this paper highlights that the goal of our chatbot was to reach individuals with eating disorders who are not engaged with traditional mental health services using a scalable digital screening and intervention approach. We describe our rationale for the lack of clinical assessments in our study, namely, the importance of conducting digital mental health research under real‐world conditions and the need to reduce barriers to research participation for individuals in need of treatment for eating disorders. We emphasise the substantial public health impact that a chatbot, like the one we developed and evaluated, could have for increasing treatment‐seeking and uptake among adults with eating disorders and encourage researchers to consider study designs that are conducted under real‐world conditions.

The pandemic forced virtually all mental health professionals to use some form of digital technology, yet few receive any training in digital mental health tools (DMHT). Therapists and students in all mental health treatment settings should be trained to routinely include DMHT in their practice. In this article, we describe why DMHT will play an increasingly important part in clinical mental health practice and discuss how we train psychologists and counselors in our eClinic to learn the basics of incorporating digital technologies into the care they provide. More specifically, we describe the three basic components of the training: (a) an asynchronous onboarding program; (b) a didactic curriculum, implemented via monthly core skill development seminars led by experts in digital mental health practice; and (c) ongoing weekly supervision by licensed supervisors. The eClinic training approach described is a work in progress, and we expect to adjust it to follow the evolution of digital tools for mental health assessment and treatment.

Objective We developed a chatbot aimed to facilitate mental health services use for eating disorders (EDs) and offered the opportunity to enrol in a research study and use the chatbot to all adult respondents to a publicly available online ED screen who screened positive for clinical/subclinical EDs and reported not currently being in treatment. We examined the rates and correlates of enrolment in the study and uptake of the chatbot. Method Following screening, eligible respondents (≥18 years, screened positive for a clinical/subclinical ED, not in treatment for an ED) were shown the study opportunity. Chi‐square tests and logistic regressions explored differences in demographics, ED symptoms, suicidality, weight, and probable ED diagnoses between those who enroled and engaged with the chatbot versus those who did not. Results 6747 respondents were shown the opportunity (80.0% of all adult screens). 3.0% enroled, of whom 90.2% subsequently used the chatbot. Enrolment and chatbot uptake were more common among respondents aged ≥25 years old versus those aged 18–24 and less common among respondents who reported engaging in regular dietary restriction. Conclusions Overall enrolment was low, yet uptake was high among those that enroled and did not differ across most demographics and symptom presentations. Future directions include evaluating respondents' attitudes towards treatment‐promoting tools and removing barriers to uptake.