Search

This manuscript presents the protocol and participatory planning process for implementing the Community Resilience Learning Collaborative and Research Network (C-LEARN) study. C-LEARN is designed to determine how to build a service program and individual client capacity to improve mental health-related quality of life among individuals at risk for depression, with exposure to social risk factors or concerns about environmental hazards in areas of Southern Louisiana at risk for events such as hurricanes and storms. The study uses a Community Partnered Participatory Research (CPPR) framework to incorporate community priorities into study design and implementation. The first phase of C-LEARN is assessment of community priorities, assets, and opportunities for building resilience through key informant interviews and community agency outreach. Findings from this phase will inform the implementation of a two-level (program-level and individual client level) randomized study in up to four South Louisiana communities. Within communities, health and social-community service programs will be randomized to Community Engagement and Planning (CEP) for multi-sector coalition support or Technical Assistance (TA) for individual program support to implement evidence-based and community-prioritized intervention toolkits, including an expanded version of depression collaborative care and resources (referrals, manuals) to address social risk factors such as financial or housing instability and for a community resilience approach to disaster preparedness and response. Within each arm, the study will randomize individual adult clients to one of two mobile applications that provide informational resources on services for depression, social risk factors, and disaster response or also provide psychoeducation on Cognitive Behavioral Therapy to enhance coping with stress and mood. Planned data collection includes baseline, six-month and brief monthly surveys for clients, and baseline and 12-month surveys for administrators and staff.

BackgroundWe examined whether digital phenotyping of spontaneous speech, such as the use of specific word categories during speech samples, was associated with depressive symptoms in youth who were at familial and clinical risk for mood disorders.MethodsParticipants (ages 13-19) had active mood symptoms, mood instability, and at least one parent with bipolar or major depressive disorder. During a randomized trial of family-focused therapy, participants were instructed to make weekly calls to a central voice server and leave speech samples in response to automated prompts. We coded youths' speech samples with the Linguistic Inquiry and Word Count system and used machine learning to identify the combination of speech features that were most closely associated with the course of depressive symptoms over 18 weeks.ResultsA total of 253 speech samples were collected from 44 adolescents (mean age = 15.8 years; SD = 1.6) over 18 weeks. Speech containing affective processes, social processes, drives toward risk or reward, nonfluencies, and time orientation words were correlated with depressive symptoms at concurrent time periods (ps

Expressed emotion (EE), a measure of attitudes among caregivers towards a patient with a psychiatric disorder, is a robust predictor of relapse across mood and psychotic disorders. Because the measurement of EE is time-intensive and costly, its use in clinical settings has been limited. In an effort to automate EE classification, we evaluated whether machine learning (ML) applied to lexical features of speech samples can accurately categorize parents as high or low in EE and in its subtypes (criticism, overinvolvement, and warmth). The sample was 123 parents of youth who had active mood symptoms and a family history of bipolar disorder. Using ML algorithms, we achieved 75.2-81.8% accuracy (sensitivities of ~0.7 and specificities of ~0.8) in classifying parents as high or low in EE and EE subtypes. Additionally, machine-derived EE classifications and observer-rated EE classifications had simiar relationships with youth mood symptoms, parental distress, and family conflict. Of note, criticism related to greater manic severity, parental distress, and family conflict. Study findings indicate that EE classification can be automated through lexical analysis and suggest potential for facilitating larger-scale applications in clinical settings. The results also provide initial indications of the digital phenotypes that underlie EE and its subtypes.

Individuals with serious mental illness experience changes in their clinical states over time that are difficult to assess and that result in increased disease burden and care utilization. It is not known if features derived from speech can serve as a transdiagnostic marker of these clinical states. This study evaluates the feasibility of collecting speech samples from people with serious mental illness and explores the potential utility for tracking changes in clinical state over time. Patients (n = 47) were recruited from a community-based mental health clinic with diagnoses of bipolar disorder, major depressive disorder, schizophrenia or schizoaffective disorder. Patients used an interactive voice response system for at least 4 months to provide speech samples. Clinic providers (n = 13) reviewed responses and provided global assessment ratings. We computed features of speech and used machine learning to create models of outcome measures trained using either population data or an individual's own data over time. The system was feasible to use, recording 1101 phone calls and 117 hours of speech. Most (92%) of the patients agreed that it was easy to use. The individually-trained models demonstrated the highest correlation with provider ratings (rho = 0.78, p

Abstract Purpose The primary goal of this study was to test the feasibility of an educational online self-assessment of burnout, resilience, trauma, depression, anxiety, and common workplace stressors among nurses working in a pediatric intensive care unit or neonatal intensive care unit setting. The secondary, exploratory objectives were to estimate the prevalence of psychiatric symptoms in this sample and to identify those variables that most strongly predict burnout. Design and Methods Data from optional and anonymous online measures were analyzed for 115 nurses (67.9% aged 25–44; 61.7% Caucasian) working in an urban children's hospital pediatric or neonatal ICU. Multiple linear regressions identified demographic variables and workplace stressors that significantly predicted each of three components of burnout. Results Most respondents found the educational assessment and feedback to be helpful. Choosing nursing as a second career was associated with better resilience. Having worked in ICU settings longer and being older were both linked to lower levels of anxiety. Predictors of burnout varied across the three burnout subscales. Conclusions Implementation of an online self-assessment with immediate educational feedback is feasible in critical care settings. The variability of predictors across the three burnout subscales indicates the need for tailored interventions for those at risk. Future research may include follow-up of nurses to examine changes in scores over time and expansion of the tool for other medical personnel. Practice Implications An educational online self-assessment can be a helpful tool for pediatric critical care nurses experiencing varying degrees of burnout and distress. Highlights • Few tools exist to facilitate self-assessment of burnout and psychological distress among pediatric critical care nurses. • An online self-assessment tool with evidence-informed feedback is feasible and acceptable to pediatric critical care nurses. • Stressors that predict burnout vary across the three main subscales of the burnout measure. [ABSTRACT FROM AUTHOR]

Expressed emotion (EE), a measure of attitudes among caregivers towards a patient with a psychiatric disorder, is a robust predictor of relapse across mood and psychotic disorders. Because the measurement of EE is time-intensive and costly, its use in clinical settings has been limited. In an effort to automate EE classification, we evaluated whether machine learning (ML) applied to lexical features of speech samples can accurately categorize parents as high or low in EE and in its subtypes (criticism, overinvolvement, and warmth). The sample was 123 parents of youth who had active mood symptoms and a family history of bipolar disorder. Using ML algorithms, we achieved 75.2–81.8% accuracy (sensitivities of ~0.7 and specificities of ~0.8) in classifying parents as high or low in EE and EE subtypes. Additionally, machine-derived EE classifications and observer-rated EE classifications had simiar relationships with youth mood symptoms, parental distress, and family conflict. Of note, criticism related to greater manic severity, parental distress, and family conflict. Study findings indicate that EE classification can be automated through lexical analysis and suggest potential for facilitating larger-scale applications in clinical settings. The results also provide initial indications of the digital phenotypes that underlie EE and its subtypes. [ABSTRACT FROM AUTHOR]

We examined whether digital phenotyping of spontaneous speech, such as the use of specific word categories during speech samples, was associated with depressive symptoms in youth who were at familial and clinical risk for mood disorders. Participants (ages 13–19) had active mood symptoms, mood instability, and at least one parent with bipolar or major depressive disorder. During a randomized trial of family-focused therapy, participants were instructed to make weekly calls to a central voice server and leave speech samples in response to automated prompts. We coded youths' speech samples with the Linguistic Inquiry and Word Count system and used machine learning to identify the combination of speech features that were most closely associated with the course of depressive symptoms over 18 weeks. A total of 253 speech samples were collected from 44 adolescents (mean age = 15.8 years; SD = 1.6) over 18 weeks. Speech containing affective processes, social processes, drives toward risk or reward, nonfluencies, and time orientation words were correlated with depressive symptoms at concurrent time periods (p s < 0.01). Machine learning analyses revealed that affective processes, nonfluencies, drives and risk words combined to most strongly predict changes in depressive symptoms over 18 weeks of treatment. Study results were limited by the small sample and the exclusion of paralinguistic or contextual variables in analyzing speech samples. In youth at high risk for mood disorders, knowledge of speech patterns may inform prognoses during outpatient psychosocial treatment. • We identified speech features that associated with depressive symptoms • Speech was collected from youth at risk for mood disorders over an 18-week period • Affective, social, drive (risk/reward), nonfluencies, and time orientation words correlated with depressive symptoms • Machine learning found affective, nonfluencies, drive, and risk words combined to be most predictive depressive symptoms [ABSTRACT FROM AUTHOR]