An abstract is not available for this content. As you have access to this content, full HTML content is provided on this page. A PDF of this content is also available in through the 'Save PDF' action button.
In establishing prognostic models, often aided by machine learning methods, much effort is concentrated in identifying good predictors. However, the same level of rigor is often absent in improving the outcome side of the models. In this study, we focus on this rather neglected aspect of model development. We are particularly interested in the use of longitudinal information as a way of improving the outcome side of prognostic models. This involves optimally characterizing individuals’ outcome status, classifying them, and validating the formulated prediction targets. None of these tasks are straightforward, which may explain why longitudinal prediction targets are not commonly used in practice despite their compelling benefits. As a way of improving this situation, we explore the joint use of empirical model fitting, clinical insights, and cross-validation based on how well formulated targets are predicted by clinically relevant baseline characteristics (antecedent validators). The idea here is that all these methods are imperfect but can be used together to triangulate valid prediction targets. The proposed approach is illustrated using data from the longitudinal assessment of manic symptoms study.
Abstract Background This study is a cost-effectiveness study of two implementation strategies designed to train therapists in college and university counseling centers to deliver interpersonal psychotherapy. Costs of implementing a train-the-trainer (TTT) strategy versus an expert consultation strategy were estimated, and their relative effects upon therapist outcomes were calculated and compared. Methods Twenty four counseling centers were recruited across the United States. These centers were randomized to either a TTT (experimental) condition, in which an in-house therapist trained other center therapists, or an expert consultation condition, in which center therapists participated in a workshop and received 12 months of ongoing supervision. The main outcome was therapist fidelity (adherence and competence) to interpersonal psychotherapy, assessed via audio recordings of therapy sessions, and analyzed using linear mixed models. Costs of each condition were quantified using time-driven activity-based costing methods, and involved a costing survey administered to center directors, follow up interviews and validation checks, and comparison of time tracking logs of trainers in the expert condition. Mean costs to produce one therapist were obtained for each condition. The costs to produce equivalent improvements in therapist-level outcomes were then compared between the two conditions. Results Mean cost incurred by counseling centers to train one therapist using the TTT strategy was $3,407 (median = $3,077); mean cost to produce one trained therapist in the control condition was $2,055 (median = $1,932). Therapists in the TTT condition, on average, demonstrated a 0.043 higher adherence score compared to therapists in the control condition; however, this difference was not statistically significant. For the competence outcome, effect size for therapists in the TTT condition was in the large range (1.16; 95% CI: 0.85–1.46; p < .001), and therapists in this condition, on average, demonstrated a 0.073 higher competence score compared to those in the expert consultation condition (95% CI, 0.008–0.14; p = .03). Counseling centers that used the TTT model incurred $353 less in training costs to produce equivalent improvements in therapist competence. Conclusions Despite its higher short run costs, the TTT implementation strategy produces greater increases in therapist competence when compared to expert consultation. Expanding resources to support this platform for service delivery can be an effective way to enhance the mental health care of young people seeking care in college and university counseling centers. Trial registration ClinicalTrials.gov Identifier: NCT02079142.
Exposure therapy is an effective treatment for posttraumatic stress disorder (PTSD), but a comprehensive, emotion-focused perspective on how psychotherapy affects brain function is lacking. The authors assessed changes in brain function after prolonged exposure therapy across three emotional reactivity and regulation paradigms.Individuals with PTSD underwent functional MRI (fMRI) at rest and while completing three tasks assessing emotional reactivity and regulation. Individuals were then randomly assigned to immediate prolonged exposure treatment (N=36) or a waiting list condition (N=30) and underwent a second scan approximately 4 weeks after the last treatment session or a comparable waiting period, respectively.Treatment-specific changes were observed only during cognitive reappraisal of negative images. Psychotherapy increased lateral frontopolar cortex activity and connectivity with the ventromedial prefrontal cortex/ventral striatum. Greater increases in frontopolar activation were associated with improvement in hyperarousal symptoms and psychological well-being. The frontopolar cortex also displayed a greater variety of temporal resting-state signal pattern changes after treatment. Concurrent transcranial magnetic stimulation and fMRI in healthy participants demonstrated that the lateral frontopolar cortex exerts downstream influence on the ventromedial prefrontal cortex/ventral striatum.Changes in frontopolar function during deliberate regulation of negative affect is one key mechanism of adaptive psychotherapeutic change in PTSD. Given that frontopolar connectivity with ventromedial regions during emotion regulation is enhanced by psychotherapy and that the frontopolar cortex exerts downstream influence on ventromedial regions in healthy individuals, these findings inform a novel conceptualization of how psychotherapy works, and they identify a promising target for stimulation-based therapeutics.
When identification of causal effects relies on untestable assumptions regarding nonidentified parameters, sensitivity of causal effect estimates is often questioned. For proper interpretation of causal effect estimates in this situation, deriving bounds on causal parameters or exploring the sensitivity of estimates to scientifically plausible alternative assumptions can be critical. In this paper, we propose a practical way of bounding and sensitivity analysis, where multiple identifying assumptions are combined to construct tighter common bounds. In particular, we focus on the use of competing identifying assumptions that impose different restrictions on the same non-identified parameter. Since these assumptions are connected through the same parameter, direct translation across them is possible. Based on this cross-translatability, various information in the data, carried by alternative assumptions, can be effectively combined to construct tighter bounds on causal effects. Flexibility of the suggested approach is demonstrated focusing on the estimation of the complier average causal effect (CACE) in a randomized job search intervention trial that suffers from noncompliance and subsequent missing outcomes.
