Adolescent, parent, and provider attitudes toward a machine learning based clinical decision support system for selecting treatment for youth depression.

BACKGROUND: Machine learning based clinical decision support systems (CDSSs) have been proposed as a means of advancing personalized treatment planning for disorders, such as depression, that have a multifaceted etiology, course, and symptom profile. However, machine learning based models for treatment selection are rare in the field of psychiatry. They have also not yet been translated for use in clinical practice. Understanding key stakeholder attitudes toward machine learning based CDSSs is critical for developing plans for their implementation that promote uptake by both providers and families. METHODS: In Study 1, a prototype machine learning based Clinical Decision Support System for Youth Depression (CDSS-YD) was demonstrated to focus groups of adolescents with a diagnosis of depression (n = 9), parents (n = 11), and behavioral health providers (n = 8). Qualitative analysis was used to assess their attitudes towards the CDSS-YD. In Study 2, behavioral health providers were trained in the use of the CDSS-YD and they utilized the CDSS-YD in a clinical encounter with 6 adolescents and their parents as part of their treatment planning discussion. Following the appointment, providers, parents, and adolescents completed a survey about their attitudes regarding the use of the CDSS-YD. RESULTS: All stakeholder groups viewed the CDSS-YD as an easy to understand and useful tool for making personalized treatment decisions, and families and providers were able to successfully use the CDSS-YD in clinical encounters. Parents and adolescents viewed their providers as having a critical role in the use the CDSS-YD, and this had implications for the perceived trustworthiness of the CDSS-YD. Providers reported that clinic productivity metrics would be the primary barrier to CDSS-YD implementation, with the creation of protected time for training, preparation, and use as a key facilitator. CONCLUSIONS: Machine learning based CDSSs, if proven effective, have the potential to be widely accepted tools for personalized treatment planning. Successful implementation will require addressing the system-level barrier of having sufficient time and energy to integrate it into practice.

Protocol for a hybrid type 3 effectiveness-implementation trial of a pragmatic individual-level implementation strategy for supporting school-based prevention programming.

BACKGROUND: For approximately one in five children who have social, emotional, and behavioral (SEB) challenges, accessible evidence-based prevention practices (EBPPs) are critical. In the USA, schools are the primary setting for children's SEB service delivery. Still, EBPPs are rarely adopted and implemented by front-line educators (e.g., teachers) with sufficient fidelity to see effects. Given that individual behavior change is ultimately required for successful implementation, focusing on individual-level processes holds promise as a parsimonious approach to enhance impact. Beliefs and Attitudes for Successful Implementation in Schools for Teachers (BASIS-T) is a pragmatic, multifaceted pre-implementation strategy targeting volitional and motivational mechanisms of educators' behavior change to enhance implementation and student SEB outcomes. This study protocol describes a hybrid type 3 effectiveness-implementation trial designed to evaluate the main effects, mediators, and moderators of the BASIS-T implementation strategy as applied to Positive Greetings at the Door, a universal school-based EBPP previously demonstrated to reduce student disruptive behavior and increase academic engagement. METHODS: This project uses a blocked randomized cohort design with an active comparison control (ACC) condition. We will recruit and include approximately 276 teachers from 46 schools randomly assigned to BASIS-T or ACC conditions. Aim 1 will evaluate the main effects of BASIS-T on proximal implementation mechanisms (attitudes, subjective norms, self-efficacy, intentions to implement, and maintenance self-efficacy), implementation outcomes (adoption, reach, fidelity, and sustainment), and child outcomes (SEB, attendance, discipline, achievement). Aim 2 will examine how, for whom, under what conditions, and how efficiently BASIS-T works, specifically by testing whether the effects of BASIS-T on child outcomes are (a) mediated via its putative mechanisms of behavior change, (b) moderated by teacher factors or school contextual factors, and (c) cost-effective. DISCUSSION: This study will provide a rigorous test of BASIS-T-a pragmatic, theory-driven, and generalizable implementation strategy designed to target theoretically-derived motivational mechanisms-to increase the yield of standard EBPP training and support strategies. TRIAL REGISTRATION: ClinicalTrials.gov ID: NCT05989568. Registered on May 30, 2023.

Adolescent, parent, and provider attitudes toward a machine-learning based clinical decision support system for selecting treatment for youth depression.

Background: Machine-learning based clinical decision support systems (CDSSs) have been proposed as a means of advancing personalized treatment planning for disorders, such as depression, that have a multifaceted etiology, course, and symptom profile. However, machine-learning based models for treatment selection are rare in the field of psychiatry. They have also not yet been translated for use in clinical practice. Understanding key stakeholder attitudes toward machine learning-based CDSSs is critical for developing plans for their implementation that promote uptake by both providers and families. Methods: In Study 1, a machine-learning based Clinical Decision Support System for Youth Depression (CDSS-YD) was demonstrated to focus groups of adolescents with a diagnosis of depression (n = 9), parents (n = 11), and behavioral health providers (n = 8). Qualitative analysis was used to assess their attitudes towards the CDSS-YD. In Study 2, behavioral health providers were trained in the use of the CDSS-YD and they utilized the CDSS-YD in a clinical encounter with 6 adolescents and their parents as part of their treatment planning discussion. Following the appointment, providers, parents, and adolescents completed a survey about their attitudes regarding the use of the CDSS-YD. Results: All stakeholder groups viewed the CDSS-YD as an easy to understand and useful tool for making personalized treatment decisions, and families and providers were able to successfully use the CDSS-YD in clinical encounters. Parents and adolescents viewed their providers as having a critical role in the use the CDSS-YD, and this had implications for the perceived trustworthiness of the CDSS-YD. Providers reported that clinic productivity metrics would be the primary barrier to CDSS-YD implementation, with the creation of protected time for training, preparation, and use as a key facilitator. Conclusions: The CDSS-YD has the potential to be a widely accepted and useful tool for personalized treatment planning. Successful implementation will require addressing the system-level barrier of having sufficient time and energy to integrate it into practice.

Examining engagement effects in an adaptive preventive intervention for college student drinking.

OBJECTIVE: This study determined the characteristics of engagement and whether engagement in an adaptive preventive intervention (API) was associated with reduced binge drinking and alcohol-related consequences. METHOD: Incoming students were recruited for a sequential multiple assignment randomized trial (SMART; N = 891, 62.4% female, 76.8% non-Hispanic White) with an assessment-only control group. The API occurred during the first semester of college, with outcomes assessed at the end of the semester. The API involved two stages. Stage 1 included universal intervention components (personalized normative feedback [PNF] and self-monitoring). Stage 2 bridged heavy drinkers to access additional resources. We estimated the effect of engagement in Stage 1 only and in the whole API (Stages 1 and 2) among the intervention group, and the effect of the API versus control had all students assigned an API engaged, on alcohol-related outcomes. RESULTS: Precollege binge drinking, intention to pledge a fraternity/sorority, and higher conformity motives were most associated with lower odds of Stage 1 engagement. Action (readiness to change) and PNF engagement were associated with Stage 2 engagement. API engagement was associated with significant reductions in alcohol-related consequences among heavy drinkers. Compared to the control, we estimated the API would reduce the relative increase in alcohol-related consequences from baseline to follow-up by 25%, had all API students engaged. CONCLUSIONS: Even partial engagement in each component of the "light-touch" API rendered benefits. Analyses suggested that had all students in the intervention group engaged, the API would significantly reduce the change in alcohol-related consequences over the first semester in college. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

An example of implementing a safety protocol in remote intervention and survey research with college students.

INTRODUCTION: This article draws attention to the need for open evaluation and reporting on safety protocols in survey and intervention research. We describe a protocol for responding to those who indicate increased risk of self-harm (i.e. suicidality or potentially lethal alcohol use) as an example and report on the outcome of our procedures. METHODS: Participants were first-year college students (n = 891) participating in an intervention trial for binge drinking. We describe the protocol, provide descriptive outcomes, and examine whether participant sex, attrition, or study intervention condition were related to endorsing items that indicated risk for suicidality or potentially lethal alcohol use. RESULTS: Of the 891 participants, 167 (18.7%) were identified as being at risk in one or more study wave. Of those, we were able to successfully contact 100 (59.9%), 76 (45.5%) by phone, and 24 (14.4%) by email. Of those 100, 78 accepted mental health resources as a result of outreach. Participant sex, attrition, and intervention condition were not related to risk. DISCUSSION: This article may aid other research teams in developing similar protocols. Strategies to reach an even greater proportion of high-risk participants are needed. A body of literature documenting published safety protocols in research and the associated outcomes would help to identify opportunities for improvement.

Association between COVID-19-related loneliness or worry and symptoms of anxiety and depression among first-year college students.

OBJECTIVE: To examine the relationship between COVID-19-related distress and mental health among first-year college students. PARTICIPANTS: Data for this longitudinal study (n = 727) were collected before the school year (August 2019), end of fall semester (December 2019), and soon after the university suspended in-person instruction (April 2020). METHODS: We used multivariable log-linear and logistic regressions to examine continuous and dichotomous outcomes on the 9-item Patient Health Questionnaire and the 7-item Generalized Anxiety Disorder scale. RESULTS: The most consistent predictor of during-pandemic mental health was feeling extremely isolated (versus not at all), which was associated with increased symptom severity of depression (proportional change[95% CI] = 2.43[1.87, 3.15]) and anxiety (2.02[1.50, 2.73]) and greater odds of new moderate depression (OR[95% CI] = 14.83[3.00, 73.41]) and anxiety (24.74[2.91, 210.00]). Greater COVID-19-related concern was also related to increased mental health symptoms. CONCLUSIONS: Results highlight the need for mental health services during crises that lead to social isolation.

Self-relevant appeals to engage in self-monitoring of alcohol use: A microrandomized trial.

OBJECTIVE: While self-monitoring can help mitigate alcohol misuse in young adults, engagement with digital self-monitoring is suboptimal. The present study investigates the utility of two types of digital prompts (reminders) to encourage young adults to self-monitor their alcohol use. These prompts leverage information that is self-relevant (i.e., represents and is valuable) to the person. METHOD: Five hundred ninety-one college students (Mage = 18; 61% = female, 76% = White) were enrolled in an 8-week intervention study involving biweekly digital self-monitoring of their alcohol use. At baseline, participants selected an item they would like to purchase for themselves and their preferred charitable organization. Then, biweekly, participants were microrandomized to a prompt highlighting the opportunity to either (a) win their preferred item (self-interest prompt); or (b) donate to their preferred charity (prosocial prompt). Following self-monitoring completion, participants allocated reward points toward lottery drawings for their preferred item or charity. RESULTS: The self-interest (vs. prosocial) prompt was significantly more effective in promoting proximal self-monitoring at the beginning of the study, Est = exp(.14) = 1.15; 95% confidence interval (CI) [1.01, 1.29], whereas the prosocial (vs. self-interest) prompt was significantly more effective at the end, Est = exp(-.17) = 0.84; 95% CI [0.70, 0.98]. Further, the prosocial (vs. self-interest) prompt was significantly more effective among participants who previously allocated all their reward points to drawings for their preferred item, Est = exp(-.15) = 0.86; 95% CI [.75, .97]. CONCLUSIONS: These results suggest that the advantage of prompts that appeal to a person's self-interest (vs. prosocial) motives varies over time and based on what reward options participants prioritized in previous decisions. Theoretical and practical implications for intervention design are discussed. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

The Work and Social Adjustment Scale for Youth: Psychometric properties of the teacher version and evidence of contextual variability in psychosocial impairments.

Effective mental health services require accurate assessment of psychosocial impairments linked to mental health concerns. Youth who experience these impairments do so within and across various contexts (e.g., school, home). Youth may display symptoms of mental health concerns without co-occurring impairments, and vice versa. Yet, nearly all impairment measures presume that those assessed display mental health concerns. Consequently, we recently developed youth and parent versions of a five-item measure of youth psychosocial impairments (i.e., Work and Social Adjustment Scale for Youth [WSASY]), structured to assess any youth, regardless of mental health status. Across two studies, we developed and tested a WSASY teacher version, in a large sample of 382 student teacher reports (Study 1), and a subsample of 66 youth who, along with their parents and teachers, completed the WSASY and a series of school- and home-based behavioral tasks (Study 2). In Study 1, WSASY teacher reports demonstrated excellent internal consistency and unique relations with teacher reports on well-established measures of psychosocial strengths and difficulties. In Study 2, teacher, youth, and parent WSASY reports demonstrated low correspondence with each other and context-specific relations with criterion variables. This low correspondence allowed us to capitalize on an integrative approach designed to optimize informant-specific variance. Integrative scores demonstrated robust, large-magnitude relations with criterion variables across multiple information sources. These findings provide important psychometric support for use of WSASY teacher reports, and pave the way toward integrating WSASY reports from multiple informants who observe youth psychosocial impairments within different contexts and from different perspectives. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Deeply Tailored Adaptive Interventions to Reduce College Student Drinking: a Real-World Application of Q-Learning for SMART Studies.

M-bridge was a sequential multiple assignment randomized trial (SMART) that aimed to develop a resource-efficient adaptive preventive intervention (API) to reduce binge drinking in first-year college students. The main results of M-bridge suggested no difference, on average, in binge drinking between students randomized to APIs versus assessment-only control, but certain elements of the API were beneficial for at-risk subgroups. This paper extends the main results of M-bridge through an exploratory analysis using Q-learning, a novel algorithm from the computer science literature. Specifically, we sought to further tailor the two aspects of the M-bridge APIs to an individual and test whether deep tailoring offers a benefit over assessment-only control. Q-learning is a method to estimate decision rules that assign optimal treatment (i.e., to minimize binge drinking) based on student characteristics. For the first aspect of the M-bridge API (when to offer), we identified the optimal tailoring characteristic post hoc from a set of 20 candidate variables. For the second (how to bridge), we used a known effect modifier from the trial. The results of our analysis are two rules that optimize (1) the timing of universal intervention for each student based on their motives for drinking and (2) the bridging strategy to indicated interventions (i.e., among those who continue to drink heavily mid-semester) based on mid-semester binge drinking frequency. We estimate that this newly tailored API, if offered to all first-year students, would reduce binge drinking by 1 occasion per 2.5 months (95% CI: decrease of 1.45 to 0.28 occasions, p < 0.01) on average. Our analyses demonstrate a real-world implementation of Q-learning for a substantive purpose, and, if replicable in future trials, our results have practical implications for college campuses aiming to reduce student binge drinking.

Main outcomes of M-bridge: A sequential multiple assignment randomized trial (SMART) for developing an adaptive preventive intervention for college drinking.

Objective: The goal was to develop a universal and resource-efficient adaptive preventive intervention (API) for incoming first-year students as a bridge to indicated interventions to address alcohol-related risks. The aims were to examine: (a) API versus assessment-only control, (b) the different APIs (i.e., 4 intervention sequences) embedded in the study design, and (c) moderators of intervention effects on binge drinking. Method: A sequential multiple assignment randomized trial (SMART) included two randomizations: timing (summer before vs. first semester) of universal personalized normative feedback and biweekly self-monitoring and, for heavy drinkers, bridging strategy (resource email vs. health coaching invitation). Participants (N = 891, 62.4% female, 76.8% White) were surveyed at the end of first and second semesters. The primary outcome was binge drinking frequency (4+/5+ drinks for females/males); secondary outcomes were alcohol consequences and health services utilization. Results: API (vs. control) was not significantly associated with outcomes. There were no differences between embedded APIs. Among heavy drinkers, the resource email (vs. health coach invitation) led to greater health services utilization. Moderator analyses suggested students intending to pledge into Greek life benefited more from any API (vs. control; 42% smaller increase from precollege in binge drinking frequency). Conclusions: Although overall effects were not significant, students at high risk (i.e., entering fraternities/sororities) did benefit more from the intervention. Furthermore, the resource email was effective for heavier drinkers. A technology-based strategy to deliver targeted resource-light interventions for heavy drinkers may be effective for reducing binge drinking during the transition to college. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Binge drinking before and after a COVID-19 campus closure among first-year college students.

PURPOSE: The COVID-19 pandemic is associated with reports of increased substance use. College students are a population of concern for high risk binge drinking and their behavior may be particularly impacted by COVID-19 campus closures. Therefore, we examine first-year college students' binge drinking soon after their university's pandemic-related suspension of in-person operations. METHODS: Students from a single campus (N = 741; age: M = 18.05, SD = 0.22) completed one assessment in April-May 2020 post-campus closure (March 2020) including theoretically-informed measures (e.g., drinking motives, norms) and two items of self-reported pre- and post-closure binge drinking frequency, the focus of these analyses. RESULTS: About half of students consistently reported not binge drinking pre- and post-closure; 6.75% reported a consistent frequency of binge drinking pre- and post-closure. Many (39.41%) reported lower 30-day binge drinking post-campus closure compared to their pre-closure reports; few (4.18%) reported higher 30-day binge drinking frequency post-campus closure. Students reporting lower binge drinking post-closure showed differences in coping, social, and enhancement drinking motives and isolation. Students reporting greater post-closure binge drinking reported higher perceived drinking norms and were more likely to be in Greek life. CONCLUSION: This study demonstrates self-reported patterns in binge drinking among first-year college students at the point of COVID-19 campus closures. Pandemic-related college closures may have been a temporary environmental intervention on this high-risk behavior for some students. Although many students were not binge drinking, some continued binge drinking after closure and may benefit from preventive interventions.

Examining the Effects of a Brief, Group-Based Motivational Implementation Strategy on Mechanisms of Teacher Behavior Change.

Training and consultation are core implementation strategies used to support the adoption and delivery of evidence-based prevention programs (EBPPs), but are often insufficient alone to effect teacher behavior change. Motivational interviewing (MI) and related behavior change techniques (e.g., strategic education, social influence, implementation planning) delivered in a group format offer promising supplements to training and consultation to improve EBPP implementation. Beliefs and Attitudes for Successful Implementation in Schools for Teachers (BASIS-T) is a theoretically informed, motivational implementation strategy delivered in a group format prior to and immediately after EBPP training. The purpose of this study was to examine the proximal effects of BASIS-T on hypothesized mechanisms of behavior change (e.g., attitudes, subjective norms, intentions to implement) in the context of teachers receiving training and consultation to implement the Good Behavior Game. As part of a pilot trial, 83 elementary school teachers from 9 public elementary schools were randomly assigned (at the school-level to reduce contamination across participants) to a BASIS-T (n = 44) or active comparison control (n = 39) condition, with both conditions receiving Good Behavior Game (GBG) training and consultation. A series of mixed effects models revealed meaningful effects favoring BASIS-T on a number of hypothesized mechanisms of behavior change leading to increased motivation to implement GBG. The implications, limitations, and directions for future research on the use of MI with groups of individuals and other behavior change techniques to increase the yield of training and consultation are discussed.

A sequential multiple assignment randomized trial (SMART) protocol for empirically developing an adaptive preventive intervention for college student drinking reduction.

College student alcohol use and associated negative consequences are clear public health problems with consequences including damage to self, others, and institutions. This paper describes the protocol of a research study designed to answer a number of important questions in the development of an adaptive preventive intervention (API) to reduce high-risk drinking among first-year college students. The API is designed to educate students and to motivate heavy-drinking college students to engage in existing resources to support reducing high-risk alcohol use, by leveraging technology-based intervention modalities. The primary outcome is a reduction in binge drinking, with secondary outcomes of reducing negative alcohol-related consequences and increasing health services utilization. Adaptive preventive interventions have the potential to reduce the acute and long-term negative health consequences of young adult alcohol use.

A randomized preference trial to inform personalization of a parent training program implemented in community mental health clinics.

Incorporating participant preferences into intervention decision-making may optimize health outcomes by improving participant engagement. We describe the rationale for a preference-based approach to the personalization of community-based interventions. Compensating for the limitations of traditional randomized controlled trials (RCTs) and partially randomized preference trials (PRPTs), we employed a doubly randomized preference trial in the present study. Families (N = 129) presenting to community mental health clinics for child conduct problems were randomized to choice or no-choice conditions. Within each condition, parents were again randomized, or offered choices between home- and clinic-based, individual and group versions of a parent training program or services-as-usual. Participants were assessed at baseline, and treatment retention data were gathered. Families assigned to the choice condition were significantly less likely to drop out of treatment than those in the no-choice condition. In the choice condition, in-home treatment was the preferred modality, and across conditions, families were less likely to be retained in group and clinic modalities. Research on preferences may boost participant engagement and inform shared decision-making.

Two-year outcomes of the Early Risers prevention trial with formerly homeless families residing in supportive housing.

This article reports 2-year outcomes from a cluster randomized, controlled trial of the Early Risers (ER) program implemented as a selective preventive intervention in supportive housing settings for homeless families. Based on the goals of this comprehensive prevention program, we predicted that intervention participants receiving ER services would show improvement in parenting and child outcomes relative to families in treatment-as-usual sites. The sample included 270 children in 161 families, residing in 15 supportive housing sites; multimethod, multi-informant assessments conducted at baseline and yearly thereafter included parent and teacher report of child adjustment, parent report of parenting self-efficacy, and parent-child observations that yielded scores of effective parenting practices. Data were modeled in HLM7 (4-level model accounting for nesting of children within families and families within housing sites). Two years' postbaseline, intent-to-treat (ITT) analyses indicated that parents in the ER group showed significantly improved parenting self-efficacy, and parent report indicated significant reductions in ER group children's depression. No main effects of ITT were shown for observed parenting effectiveness. However, over time, average levels of parenting self-efficacy predicted observed effective parenting practices, and observed effective parenting practices predicted improvements in both teacher- and parent-report of child adjustment. This is the first study to our knowledge to demonstrate prevention effects of a program for homeless families residing in family supportive housing.