New developments in machine learning-based analysis of speech can be hypothesized to facilitate the long-term monitoring of major depressive disorder (MDD) during and after treatment. To test this hypothesis, we collected 550 speech samples from telephone-based clinical interviews with 267 individuals in routine care. With this data, we trained and evaluated a machine learning system to identify the absence/presence of a MDD diagnosis (as assessed with the Structured Clinical Interview for DSM-IV) from paralinguistic speech characteristics. Our system classified diagnostic status of MDD with an accuracy of 66% (sensitivity: 70%, specificity: 62%). Permutation tests indicated that the machine learning system classified MDD significantly better than chance. However, deriving diagnoses from cut-off scores of common depression scales was superior to the machine learning system with an accuracy of 73% for the Hamilton Rating Scale for Depression (HRSD), 74% for the Quick Inventory of Depressive Symptomatology–Clinician version (QIDS-C), and 73% for the depression module of the Patient Health Questionnaire (PHQ-9). Moreover, training a machine learning system that incorporated both speech analysis and depression scales resulted in accuracies between 73 and 76%. Thus, while findings of the present study demonstrate that automated speech analysis shows the potential of identifying patterns of depressed speech, it does not substantially improve the validity of classifications from common depression scales. In conclusion, speech analysis may not yet be able to replace common depression scales in clinical practice, since it cannot yet provide the necessary accuracy in depression detection. This trial is registered with DRKS00023670.

Due to its anti-glucocorticoid properties, the steroid hormone dehydroepiandrosterone (DHEA) might play a role for coping with traumatic stress and posttraumatic stress disorder (PTSD). The majority of studies report elevated DHEA secretion and decreased cortisol/DHEA ratio associated with traumatic stress, however, contrasting results have also been published. One reason for this heterogeneity might be that in past studies, DHEA has been measured in plasma or saliva samples reflecting acute hormone levels. In comparison, the current study assessed the hair levels of DHEA and cortisol as long-term markers along with self-reported data on psychopathology and coping in 92 female adolescents aged 11–16 from the West Bank affected by the Israeli–Palestinian conflict. Results showed that trauma-exposed individuals had significantly higher DHEA levels (p = 0.013) and lower cortisol/DHEA ratios (p = 0.036) than participants from the non-trauma group. Furthermore, DHEA and cortisol/DHEA ratio emerged as associated with trauma load and timing, but not with coping. By applying the novel method of DHEA analysis from hair samples, this study adds to the growing literature on the interplay of DHEA, cortisol, traumatic stress and coping, and provides valuable starting points for further research.

This study was aimed to evaluate whether the efficacy of invoking anti-depressive self-statements to cope with depressed mood can be enhanced for depressed individuals by systematically guiding them to amplify the expression of conviction in their voice. Accordingly, we recruited N = 144 participants (48 clinically depressed individuals, 48 sub-clinically depressed individuals, and 48 non-depressed individuals). Participants were randomly assigned to an experimental or control condition. Across study conditions, participants completed a mood induction procedure, then read aloud scripted anti-depressive self-statements designed to reduce depressed mood. Participants in the experimental condition received instructions to heighten the prosodic expression of conviction in their voice; participants in the control condition received no prosodic expression instructions. Results showed that depressed participants achieved a more pronounced decrease of depressed mood in the experimental condition than in the control condition. Further, the results indicated no effects in sub-clinically depressed and non-depressed individuals. Finally, heightened conviction expressed by participants in the experimental condition was associated with lower depressed mood and diminished depressive symptom severity. Overall, our findings suggest that fostering the prosodic expression of conviction in depressed persons' voices, while they vocalize anti-depressive self-statements, represents a promising method for augmenting the efficacy of cognitive interventions for depression.

Background: It is well known from cross-sectional studies that individual coping strategies significantly influence the pathogenesis of posttraumatic stress disorder (PTSD). Equally, undisputed is the role of biological processes, e.g. of the so-called ‘stress hormone’ cortisol for the trajectory of PTSD. Ecological momentary assessment (EMA), the repeated collection of self-reported momentary states via smartphones, is ideal for shedding light upon symptom fluctuations and coping strategies. EMA may also constitute a promising approach to provide closer associations to biomarkers than retrospective self-report. The mobile application ‘CoachPTBS’, created to facilitate transition into health-care systems, bridges waiting periods for trauma-specific psychotherapy. CoachPTBS offers tools akin to EMA that could elucidate coping with stress symptoms. Moreover, the app’s self-management tools may improve coping strategies. However, these processes have never been examined in a combined, longitudinal fashion. Objective: The aim of the current study is to assess symptom fluctuations, coping strategies and long-term endocrine correlates of PTSD by a longitudinal, multimodal approach, combining traditional, online and EMA self-report with hair cortisol data and CoachPTBS as a possible novel mHealth tool. Method: 120 participants waiting for PTSD psychotherapy will be randomly grouped. After in-situ assessment and hair sample collection, 40 will receive CoachPTBS, using it daily throughout 4 weeks. A parallel group of 40 will participate in EMA, completing daily questionnaires on symptoms and coping. In between, online surveys will be conducted. After 6 weeks, a final interview and another hair sample collection will follow. Comparisons between these groups and waitlist-control, also consisting of 40 PTSD participants, and 40 non-traumatized participants assessed via EMA regarding aversive emotions and coping are planned. Discussion: Novel insights into the interplay of biological and coping strategies in PTSD are expected due to the innovative multimodal study design. Results will further explore benefits of eHealth tools on coping with PTSD. • Study Protocol presents a multimodal approach, combining traditional measures, those of Ecological Momentary Assessment (EMA) including CoachPTBS, and hair cortisol concentrations (HCC) as a biomarker of traumatic stress.• Effects on Posttraumatic stress disorder (PTSD) symptomology by participants’ differences in intra-individual coping strategies to be explored.• Effects of coping behaviour taught through mHealth tools, such as CoachPTBS, to be explored.• Contrasting retrospective recall and instant assessment through EMA in regarding PTSD symptoms and coping behaviour.