Optimizing precision medicine for second-step depression treatment: a machine learning approach.

BACKGROUND: Less than a third of patients with depression achieve successful remission with standard first-step antidepressant monotherapy. The process for determining appropriate second-step care is often based on clinical intuition and involves a protracted course of trial and error, resulting in substantial patient burden and unnecessary delay in the provision of optimal treatment. To address this problem, we adopt an ensemble machine learning approach to improve prediction accuracy of remission in response to second-step treatments. METHOD: Data were derived from the Level 2 stage of the STAR*D dataset, which included 1439 patients who were randomized into one of seven different second-step treatment strategies after failing to achieve remission during first-step antidepressant treatment. Ensemble machine learning models, comprising several individual algorithms, were evaluated using nested cross-validation on 155 predictor variables including clinical and demographic measures. RESULTS: The ensemble machine learning algorithms exhibited differential classification performance in predicting remission status across the seven second-step treatments. For the full set of predictors, AUC values ranged from 0.51 to 0.82 depending on the second-step treatment type. Predicting remission was most successful for cognitive therapy (AUC = 0.82) and least successful for other medication and combined treatment options (AUCs = 0.51-0.66). CONCLUSION: Ensemble machine learning has potential to predict second-step treatment. In this study, predictive performance varied by type of treatment, with greater accuracy in predicting remission in response to behavioral treatments than to pharmacotherapy interventions. Future directions include considering more informative predictor modalities to enhance prediction of second-step treatment response.

Predictors of pharmacotherapy outcomes for body dysmorphic disorder: a machine learning approach.

BACKGROUND: Serotonin-reuptake inhibitors (SRIs) are first-line pharmacotherapy for the treatment of body dysmorphic disorder (BDD), a common and severe disorder. However, prior research has not focused on or identified definitive predictors of SRI treatment outcomes. Leveraging precision medicine techniques such as machine learning can facilitate the prediction of treatment outcomes. METHODS: The study used 10-fold cross-validation support vector machine (SVM) learning models to predict three treatment outcomes (i.e. response, partial remission, and full remission) for 97 patients with BDD receiving up to 14-weeks of open-label treatment with the SRI escitalopram. SVM models used baseline clinical and demographic variables as predictors. Feature importance analyses complemented traditional SVM modeling to identify which variables most successfully predicted treatment response. RESULTS: SVM models indicated acceptable classification performance for predicting treatment response with an area under the curve (AUC) of 0.77 (sensitivity = 0.77 and specificity = 0.63), partial remission with an AUC of 0.75 (sensitivity = 0.67 and specificity = 0.73), and full remission with an AUC of 0.79 (sensitivity = 0.70 and specificity = 0.79). Feature importance analyses supported constructs such as better quality of life and less severe depression, general psychopathology symptoms, and hopelessness as more predictive of better treatment outcome; demographic variables were least predictive. CONCLUSIONS: The current study is the first to demonstrate that machine learning algorithms can successfully predict treatment outcomes for pharmacotherapy for BDD. Consistent with precision medicine initiatives in psychiatry, the current study provides a foundation for personalized pharmacotherapy strategies for patients with BDD.

Mechanisms of cognitive-behavioral therapy effects on symptoms of body dysmorphic disorder: a network intervention analysis.

BACKGROUND: Body dysmorphic disorder (BDD) is a severe and undertreated condition. Although cognitive-behavioral therapy (CBT) is the first-line psychosocial treatment for this common disorder, how the intervention works is insufficiently understood. Specific pathways have been hypothesized, but only one small study has examined the precise nature of treatment effects of CBT, and no prior study has examined the effects of supportive psychotherapy (SPT). METHODS: This study re-examined a large trial (n = 120) comparing CBT to SPT for BDD. Network intervention analyses were used to explore symptom-level data across time. We computed mixed graphical models at multiple time points to examine relative differences in direct and indirect effects of the two interventions. RESULTS: In the resulting networks, CBT and SPT appeared to differentially target certain symptoms. The largest differences included CBT increasing efforts to disengage from and restructure unhelpful thoughts and resist BDD rituals, while SPT was directly related to improvement in BDD-related insight. Additionally, the time course of differences aligned with the intended targets of CBT; cognitive effects emerged first and behavioral effects second, paralleling cognitive restructuring in earlier sessions and the emphasis on exposure and ritual prevention in later sessions. Differences in favor of CBT were most consistent for behavioral targets. CONCLUSIONS: CBT and SPT primarily affected different symptoms. To improve patient care, the field needs a better understanding of how and when BDD treatments and treatment components succeed. Considering patient experiences at the symptom level and over time can aid in refining or reorganizing treatments to better fit patient needs.

Rising early warning signals in affect associated with future changes in depression: a dynamical systems approach.

BACKGROUND: Predicting future states of psychopathology such as depressive episodes has been a hallmark initiative in mental health research. Dynamical systems theory has proposed that rises in certain 'early warning signals' (EWSs) in time-series data (e.g. auto-correlation, temporal variance, network connectivity) may precede impending changes in disorder severity. The current study investigates whether rises in these EWSs over time are associated with future changes in disorder severity among a group of patients with major depressive disorder (MDD). METHODS: Thirty-one patients with MDD completed the study, which consisted of daily smartphone-delivered surveys over 8 weeks. Daily positive and negative affect were collected for the time-series analyses. A rolling window approach was used to determine whether rises in auto-correlation of total affect, temporal standard deviation of total affect, and overall network connectivity in individual affect items were predictive of increases in depression symptoms. RESULTS: Results suggested that rises in auto-correlation were significantly associated with worsening in depression symptoms (r = 0.41, p = 0.02). Results indicated that neither rises in temporal standard deviation (r = -0.23, p = 0.23) nor in network connectivity (r = -0.12, p = 0.59) were associated with changes in depression symptoms. CONCLUSIONS: This study more rigorously examines whether rises in EWSs were associated with future depression symptoms in a larger group of patients with MDD. Results indicated that rises in auto-correlation were the only EWS that was associated with worsening future changes in depression.

Psychology Internship Training Amidst COVID-19: Balancing Training Opportunities, Patient Care, and Risk of Exposure.

The emergence of the 2019 novel coronavirus (COVID-19) has dramatically altered how psychologists deliver its training. At least for the time being, virtual care has become the primary method for delivering mental health services. This has allowed patients and clinicians to continue to access and provide services in a way that would have been impossible years ago. Not only has this shift impacted patients, but it has also impacted supervision and training. The impact has been especially profound on inpatient units where the psychiatric and medical acuity is high of patients and the therapeutic milieu is an important aspect of treatment. The purpose of this paper is to review the impact of COVID-19 on pre-doctoral psychology interns during their rotation on an inpatient psychiatry unit at the start of the pandemic (January to June of 2020) and use these experiences to onboard the next class of interns in the new academic year (July 2020 to June 2021) using a hybrid model of in-person and virtual training experiences. At the end of 2020/2021 rotation, we voluntarily asked interns to complete a questionnaire that was developed based on the qualitative experiences of the previous class to assess the effectiveness of this hybrid model. We also surveyed multi-disciplinary staff members who were essential personnel and required to work in person during this time about their experiences of safety and support. With this information, we explore and offer guidance to other inpatient training sites who are likely to encounter similar challenges during this time. In particular, we discuss the integration of virtual technology into this training experience, as well as the restructuring of clinical and supervisory experiences. We highlighted several short-term strategies that we have flexibly adapted to our inpatient unit. The lessons learned herein seek to guide supervisors and trainees alike in adapting their psychology training programs to meet the evolving demands of COVID-19.

Engagement with smartphone-delivered behavioural activation interventions: a study of the MoodMission smartphone application.

BACKGROUND: Despite increased research interest in smartphone mental health applications (MHapps), few studies have examined user engagement and its determinants. MoodMission is a MHapp that targets low mood and anxiety via evidence-based techniques including behavioural activation (BA). AIMS: The present study aimed to investigate (i) whether BA interventions delivered with visual psychoeducation had greater engagement than BA interventions delivered with solely written psychoeducation, (ii) whether BA interventions targeting mastery would have greater engagement than those targeting pleasure, and (iii) the relationship between level of engagement and MHapp benefit. METHOD: Participants downloaded MoodMission and completed activities and within-app evaluations over a 30-day period. Data from 238 MoodMission users were analysed via multi-level modelling and linear regression. RESULTS: The average number of app-based activities completed was 5.46 and the average self-reported engagement level was in the low to moderate range. As hypothesized, higher levels of engagement significantly predicted more positive activity appraisal. CONCLUSIONS: The results suggest that BA technique beliefs are involved in MHapp engagement and future research examining user appraisals of techniques is warranted.

A complex network approach to clinical science.

Contemporary classification systems assume that psychiatric disorders are expressions of latent disease entities. However, some critics point to the comorbidity problem and other issues that question the validity of the latent disease model. An alternative to this traditional view is the complex network approach. This approach assumes that disorders exist as systems of inter-connected elements, without requiring that the elements are expressions of latent disease entities. Depending on the structure of the network, change can occur abruptly once the network reaches a tipping point. A dynamic complex network approach could be used to develop a functional analytic case conceptualization that may predict treatment change, relapse and recovery, thereby linking nosology and treatment. In conclusion, the complex network perspective offers an alternative and less restrictive approach to the latent disease model, while offering exciting new directions for future research in psychiatry.

Effect of treatments for depression on quality of life: a meta-analysis.

Cognitive-behavioral therapy (CBT) and selective serotonin reuptake inhibitors (SSRIs) are the two first-line treatments for depression, but little is known about their effects on quality of life (QOL). A meta-analysis was conducted to examine changes in QOL in adults with major depressive disorder who received CBT (24 studies examining 1969 patients) or SSRI treatment (13 studies examining 4286 patients) for their depression. Moderate improvements in QOL from pre to post-treatment were observed in both CBT (Hedges' g = .63) and SSRI (Hedges' g = .79) treatments. The effect size remained stable over the course of the follow-up period for CBT. No data were available to examine follow-ups in the SSRI group. QOL effect sizes decreased linearly with publication year, and greater improvements in depression were significantly associated with greater improvements in QOL for CBT, but not for SSRIs. CBT and SSRIs for depression were both associated with moderate improvements in QOL, but are possibly caused by different mechanisms.

Effect of regulating anger and sadness on decision-making.

The aim of the current study was to investigate the effects of reappraisal, acceptance, and rumination for regulating anger and sadness on decision-making. Participants (N = 165) were asked to recall two autobiographical events in which they felt intense anger and sadness, respectively. Participants were then instructed to reappraise, accept, ruminate, or not use any strategies to regulate their feelings of anger and sadness. Following this manipulation, risk aversion, and decision-making strategies were measured using a computer-based measure of risk-taking and a simulated real-life decision-making task. Participants who were instructed to reappraise their emotions showed the least anger and sadness, the most adaptive decision-making strategies, but the least risk aversion as compared to the participants in the other conditions. These findings suggest that emotion regulation strategies of negative affective states have an immediate effect on decision-making and risk-taking behaviors.

The development of a novel scale to assess intra- and interpersonal emotion regulation strategies: The Emotion Regulation Strategy Scale (ERSS).

Existing emotion regulation research focuses on how individuals use different strategies to manage their own emotions-also called intra-personal emotion regulation. However, people often leverage connections with others to regulate their own emotions-interpersonal emotion regulation. The goal of the present studies was to develop a comprehensive and efficient scale-the Emotion Regulation Strategies Scale (ERSS)-to assess nine specific emotion regulation strategies that individuals use both intra-personally and interpersonally. These emotion regulation strategies were cognitive reappraisal, distraction, situation selection, problem solving, acceptance, calming, savoring, rumination, and expressive suppression. Data were collected between 2020 and 2022. Study 1 adopted a qualitative approach to establish original scale items. Results of the confirmatory factor analysis in Study 2 confirmed a nine-factor solution for both the intra- and the interpersonal scales and finalized scale items. A second confirmatory factor analysis in Study 3 found the ERSS for both the intra-personal and interpersonal scale models to possess good model fit. Correlations from Study 3 showed the ERSS subscales to be related in expected ways to existing emotion regulation scales, yet not redundant with these scales. The degree to which individuals used the range of intra- and interpersonal emotion regulation strategies assessed on the ERSS also related to the levels of clinical symptoms. The ERSS represents a comprehensive novel scale that can flexibly assess a range of specific emotion regulation strategies used both intra- and interpersonally. Future work should be conducted using the ERSS cross culturally and in clinical samples. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

The use of coaching in smartphone app-based cognitive behavioral therapy for body dysmorphic disorder.

Background: Body dysmorphic disorder (BDD) is severe and undertreated. Digital mental health could be key to expanding access to evidence-based treatments, such as cognitive behavioral therapy for BDD (CBT-BDD). Coach guidance is posited to be essential for effective uptake of digital interventions. However, little is known about how different patients may use coaching, what patterns correspond to meaningful outcomes, and how to match coaching to patient needs. Methods: Participants were 77 adults who received a 12-week guided smartphone CBT-BDD. Bachelor's-level coaches were available via asynchronous messaging. We analyzed the 400 messages sent by users to coaches during treatment. Message content was coded using the efficiency model of support (i.e., usability, engagement, fit, knowledge, and implementation). We aimed to clarify when and for what purposes patients with BDD used coaching, and if we can meaningfully classify patients by these patterns. We then assessed potential baseline predictors of coach usage, and whether distinct patterns relate to clinical outcomes. Results: Users on average sent 5.88 messages (SD = 4.51, range 1-20) and received 9.84 (SD = 5.74, range 2-30). Regarding frequency of sending messages, latent profile analysis revealed three profiles, characterized by: (1) peak mid-treatment (16.88 %), (2) bimodal/more communication early and late in treatment (10.39 %), and (3) consistent low/no communication (72.73 %). Regarding content, four profiles emerged, characterized by mostly (1) engagement (51.95 %), (2) fit (15.58 %), (3) knowledge (15.58 %), and (4) miscellaneous/no messages (16.88 %). There was a significant relationship between frequency profile and age, such that the early/late peak group was older than the low communication group, and frequency profile and adherence, driven by the mid-treatment peak group completing more modules than the low contact group. Regarding content, the engagement and knowledge groups began treatment with more severe baseline symptoms than the fit group. Content profile was associated with dropout, suggesting higher dropout rates in the miscellaneous/no contact group and reduced rates in the engagement group. There was no relationship between profile membership and other outcomes. Discussion: The majority of participants initiated little contact with their coach and the most common function of communications was to increase engagement. Results suggest that older individuals may prefer or require more support than younger counterparts early in treatment. Additionally, whereas individuals using coaching primarily for engagement may be at lower risk of dropping out, those who do not engage at all may be at elevated risk. Findings can support more personalized, data-driven coaching protocols and more efficient allocation of coaching resources.

Using Machine Learning to Examine Suicidal Ideation After Traumatic Brain Injury: A Traumatic Brain Injury Model Systems National Database Study.

OBJECTIVE: The aim of the study was to predict suicidal ideation 1 yr after moderate to severe traumatic brain injury. DESIGN: This study used a cross-sectional design with data collected through the prospective, longitudinal Traumatic Brain Injury Model Systems network at hospitalization and 1 yr after injury. Participants who completed the Patient Health Questionnaire-9 suicide item at year 1 follow-up ( N = 4328) were included. RESULTS: A gradient boosting machine algorithm demonstrated the best performance in predicting suicidal ideation 1 yr after traumatic brain injury. Predictors were Patient Health Questionnaire-9 items (except suicidality), Generalized Anxiety Disorder-7 items, and a measure of heavy drinking. Results of the 10-fold cross-validation gradient boosting machine analysis indicated excellent classification performance with an area under the curve of 0.882. Sensitivity was 0.85 and specificity was 0.77. Accuracy was 0.78 (95% confidence interval, 0.77-0.79). Feature importance analyses revealed that depressed mood and guilt were the most important predictors of suicidal ideation, followed by anhedonia, concentration difficulties, and psychomotor disturbance. CONCLUSIONS: Overall, depression symptoms were most predictive of suicidal ideation. Despite the limited clinical impact of the present findings, machine learning has potential to improve prediction of suicidal behavior, leveraging electronic health record data, to identify individuals at greatest risk, thereby facilitating intervention and optimization of long-term outcomes after traumatic brain injury.

Temporal and contemporaneous network structures of affect and physical activity in emotional disorders.

BACKGROUND: High negative affect, low positive affect, and limited physical activity figure prominently in psychopathology, but little is known about the interrelatedness of affect and physical activity in emotional disorders. METHODS: We combined ecological momentary assessment data with a network approach to examine the dynamic relations among positive affect, negative affect, and smartphone-based estimates of physical activity in 34 participants with anxiety and depressive disorders over a 2-week period. RESULTS: In the contemporaneous networks, the positive affect nodes exhibited greater overall strength centrality than negative affect nodes. The temporal networks indicated that the negative affect node 'sadness' exhibited the greatest out-strength centrality. Furthermore, physical activity was unconnected to the affect nodes in either the temporal or contemporaneous networks. CONCLUSIONS: Whereas positive affect plays a greater role in the contemporaneous experience of emotions, negative affect contributes more so to future affective states.

Cognitive-Behavioral Treatments for Anxiety and Stress-Related Disorders.

Cognitive-behavioral therapy (CBT) is a first-line, empirically supported intervention for anxiety disorders. CBT refers to a family of techniques that are designed to target maladaptive thoughts and behaviors that maintain anxiety over time. Several individual CBT protocols have been developed for individual presentations of anxiety. The article describes common and unique components of CBT interventions for the treatment of patients with anxiety and related disorders (i.e., panic disorder, social anxiety disorder, generalized anxiety disorder, obsessive-compulsive disorder, posttraumatic stress disorder, prolonged grief). Recent strategies for enhancing the efficacy of CBT protocols are highlighted as well.

Change processes in cognitive behavioral therapy and motivational interviewing for depression and heavy alcohol use: A network approach.

Objective: Both cognitive behavioral therapy for depression (CBT-D) combined with brief motivational interviewing (CBT-D + BMI) and alone are associated with symptom improvement among college students with co-occurring depression and heavy episodic drinking (HED). However, little is known about change processes underlying these different treatments.The current study uses a network approach to examine change process that may differentially underlie CBT-D + BMI relative to CBT alone. Methods: Participants included 94 college students with depression and HED who were randomized to either eight weeks of CBT-D + BMI or CBT alone. A network approach was adopted to examine how treatment condition influenced changes in the network structure of depression symptoms, heavy drinking, drinking motives, and consequences of alcohol. Network analyses were conducted using change scores representing the eight-week difference from pre-treatment to post-treatment assessments. Results: Relative to CBT-D alone, the combined CBT-D + BMI treatment influenced the symptom network structure by preferentially targeting reductions in drinking to cope motives and in the depression symptom 'loss of interest'. Conclusion: The current study revealed that combined CBT-D + BMI may confer therapeutic benefit through different network structure pathways than CBT-D alone. Specifically, augmenting CBT-D with BMI may influence change processes related to drinking motives, such as drinking to cope.

Beyond linear mediation: Toward a dynamic network approach to study treatment processes.

Few clinical scientists would disagree that more research is needed on the underlying mechanisms and processes of change in psychological therapies. In the dominant current approach, processes of change are studied through mediation. The study of mediation has been largely structured around a distinction between moderation and mediation first popularized by Baron and Kenny's (1986) seminal article, which is based on a nomothetic and cross-sectional framework. In this article, we argue that this approach is unable to adequately address change processes in psychological therapies, because it falsely assumes that treatment change is a linear, unidirectional, pauci-variate process and that the statistical assumptions are met to study processes of change in an individual using a nomothetic approach. In contrast, we propose that treatment is a dynamic process involving numerous variables that may form bi-directional and complex relationships that differ between individuals. Such relationships can best be studied using an individual dynamic network approach connected to nomothetic generalization methods that are based on a firm idiographic foundation. We argue that our proposal is available, viable, and can readily be integrated into existing research strategies. We further argue that adopting an individual dynamic network approach combined with experimental analyses will accelerate the study of treatment change processes, which is necessary as the field of evidence-based care moves toward a process-based model. We encourage future research to gather empirical evidence to examine this approach.

A randomized clinical trial of group and individual Cognitive-Behavioral Therapy approaches for Social Anxiety Disorder.

To compare the effectiveness of two Cognitive-Behavioral Therapy (CBT) interventions-an individual and a group intervention-in Social Anxiety Disorder therapy. We compared the two treatment groups against a waitlist condition in a randomized clinical trial with 86 young adults. The individual CBT intervention was Trial-Based Cognitive Therapy (TBCT) developed by De-Oliveira, a novel technique in which the therapist engages the patient in a simulated judicial trial with the goal of identifying and changing core dysfunctional beliefs. The group intervention consisted of exposition therapy based on the Hofmann and Otto protocol (Group CBT) to restructure negative and dysfunctional cognitions regarding social situations. Both interventions reduced psychiatric symptoms from pre- to post-test and primary social anxiety and depression symptoms relative to waitlist controls. The interventions were recently introduced in Brazil, and this is the first randomized control trial to compare TBCT and this Group CBT, which were effective in assessing changes in social anxiety symptoms as well as co-occurring psychiatric symptoms.

Comparar la efectividad de dos intervenciones de Terapia Cognitivo-Conductual (TCC)-intervención individual y grupal- en tratamiento del Trastorno de ansiedad social. Comparamos los dos grupos de tratamiento con una condición de lista de espera en un ensayo clínico aleatorizado con 86 adultos jóvenes. La intervención individual de TCC fue la Terapia Cognitiva Basada en Ensayos (TCBE) desarrollada por De-Oliveira, una técnica novedosa en la cual el terapeuta involucra al paciente en un juicio judicial simulado con el objetivo de identificar y cambiar las creencias disfuncionales centrales. La intervención grupal consistió en terapia de exposición basada en el protocolo Hofmann y Otto (TCC grupal) para reestructurar cogniciones negativas y disfuncionales con respecto a situaciones sociales. Ambas intervenciones redujeron los síntomas psiquiátricos antes y después de la prueba y los síntomas de ansiedad y depresión social primarios en relación con los controles de la lista de espera. Las intervenciones se introdujeron recientemente en Brasil, y este es el primer ensayo de control aleatorizado para comparar TCBE y TCC grupal, que fueron efectivos para los cambios en los síntomas de ansiedad social y los síntomas psiquiátricos concurrentes.

Emotion Regulation Flexibility in Generalized Anxiety Disorder.

BACKGROUND: People differ in their affective styles, which refers to habitual use of emotion regulation (ER) strategies. Previous research has shown that mental health is associated with an individual's adaptive flexibility of emotion regulation strategies rather than any one particular ER strategy. METHODS: The present study employed a person-centered approach using latent profile analyses to distinguish patients with generalized anxiety disorder based on their responses on an affective styles measure. RESULTS: Results of the latent profile analysis supported a three-class solution. Class 1 (26% of participants) identified individuals with the lowest scores of each affective style; class 2 (10%) included individuals with the highest scores of each style; and class 3 (64%) consisted of individuals who scored in the mid-range of each affective style. Greater ER flexibility was associated with better emotional functioning and quality of life. CONCLUSIONS: Patients with GAD differ in ER flexibility. The vast majority of patients appear to have only moderate or low ER flexibility. Those individuals with high ER flexibility show a greater quality of life and less emotional distress.

Anhedonia is central for the association between quality of life, metacognition, sleep, and affective symptoms in generalized anxiety disorder: A complex network analysis.

BACKGROUND: Poor quality of life, sleep problems, anhedonia, and negative metacognitions are common in anxiety and depression. To examine the nature of the relationship between these features and the role of metacognitions, anhedonia, and quality of life in anxiety and depression, we conducted a complex network analysis with items of self-report measures assessing quality of life, sleep, negative thinking styles, anxiety, and depression. METHODS: Participants were 226 treatment seeking individuals with a primary DSM-5 diagnosis of generalized anxiety disorder. Node centrality, strength, expected influence, community, and bridge estimation were calculated using partial correlation coefficients and glasso regularization. RESULTS: Results revealed that anhedonia was the most central node followed by quality of life nodes. Moreover, anhedonia exhibited the highest strength and expected influence, which were both stable, reliable metrics within the network. Metacognitions were not central nodes in the network, but were strong bridge symptoms between communities. LIMITATIONS: The results are limited by the cross-sectional nature of the data and the administration of self-report scales at one time-point, despite different rating anchors. CONCLUSION: These findings suggest that anhedonia is a crucial element for the association between quality of life, sleep problems, and negative cognitions.