Background: The use of artificial intelligence for psychological advice shows promise for enhancing accessibility and reducing costs, but it remains unclear whether AI-generated advice can match the quality and empathy of experts.

Method: In a blinded, comparative cross-sectional design, licensed psychologists and psychotherapists assessed the quality, empathy, and authorship of psychological advice, which was either AI-generated or authored by experts.

Results: AI-generated responses were rated significantly more favorable for emotional (OR = 1.79, 95 % CI [1.1, 2.93], p = .02) and motivational empathy (OR = 1.84, 95 % CI [1.12, 3.04], p = .02). Ratings for scientific quality (p = .10) and cognitive empathy (p = .08) were comparable to expert advice. Participants could not distinguish between AI- and expert-authored advice (p = .27), but perceived expert authorship was associated with more favorable ratings across these measures (ORs for perceived AI vs. perceived expert ranging from 0.03 to 0.15, all p < .001). For overall preference, AI-authored advice was favored when assessed blindly based on its actual source (β = 6.96, p = .002). Nevertheless, advice perceived as expert-authored was also strongly preferred (β = 6.26, p = .001), with 93.55 % of participants preferring the advice they believed came from an expert, irrespective of its true origin.

Conclusions: AI demonstrates potential to match expert performance in asynchronous written psychological advice, but biases favoring perceived expert authorship may hinder its broader acceptance. Mitigating these biases and evaluating AI's trustworthiness and empathy are important next steps for safe and effective integration of AI in clinical practice.

Introduction: Stress-related disorders such as adjustment disorder (AD) and exhaustion disorder (ED) are associated with substantial suffering and high societal costs. Cognitive behavioral therapy (CBT) is a promising treatment for symptom reduction but has not been rigorously compared with other active treatments. This study aimed to investigate the efficacy of CBT compared to an active control treatment (General Health Promotion; GHP) for individuals diagnosed with AD or ED. Methods: A total of 300 adults diagnosed with AD or ED were randomly assigned to a 12-week CBT (n = 151) or GHP (n = 149), both delivered as therapist-supported online treatments. The primary outcome was post-treatment scores on the 10-item Perceived Stress Scale. Secondary outcomes included several mental health symptom domains and functional impairment. All outcomes were assessed at baseline, post-treatment, and at 1-year follow-up. The trial was pre-registered on Clinicaltrials.gov (NCT04797273). Results: The CBT intervention was not superior to GHP in reducing symptoms of perceived stress and secondary symptoms. Both treatments generated large within-group effect sizes pre- to post-treatment (Cohen's d = 1.19 and 1.06, respectively) and results were maintained to the 1-year follow-up. The diagnostic group moderated the treatment effect pre- to post-treatment and indicated that CBT was superior to GHP for individuals diagnosed with AD but not for those diagnosed with ED. Conclusions: Common factors likely play an important role in symptom reduction for individuals with stress-related disorders. A refined understanding of the conceptualization of stress-related disorders and which specific mechanisms to target in respective patient groups is needed to improve treatment outcomes.

Introduction: Stress-related disorders, including adjustment disorder and exhaustion disorder, are associated with cognitive complaints and prolonged sickness absence (SA). This study evaluated the effects of cognitive behavioral therapy (CBT) compared with an active control treatment (entitled General Health Promotion: GHP) on these outcomes and examined the association between cognitive impairment and SA. Methods: In this randomized clinical trial, 300 participants were randomized to CBT or GHP. SA data from the Swedish Social Insurance register were collected from 1 year before to 1 year after treatment. Cognitive functioning, assessed via a validated test battery, was measured at baseline, posttreatment, and 1-year follow-up (1YFU). Hurdle negative binomial models evaluated SA and linear mixed-effects models analyzed cognitive outcomes. Results: CBT and GHP had comparable effects on SA and cognitive functioning with no significant between-group effects (all p > 0.05). Analyzing effects across the entire sample, the number of SA days among those with any SA did not significantly change (p = 0.946) but the likelihood of having any SA decreased significantly by the 1YFU (OR = 0.92, 95% CI [0.87–0.96], p < 0.001). Independent of treatment group, overall cognitive functioning improved both posttreatment (d = −0.42, 95% CI [−0.6 to −0.25], p < 0.001) and at 1YFU (d = −0.72, 95% CI [−0.91 to −0.54], p < 0.001). Improvements in cognitive functioning pre- to posttreatment were associated with fewer SA days at the 1YFU (IRR = 0.54, 95% CI [0.33–0.87], p = 0.01). Conclusion: This study found no specific effect of CBT on SA or cognitive functioning for individuals with stress-related disorders. Cognitive functioning improved regardless of treatment. Early support before SA initiation is likely important to prevent long-term SA.

Background: While cognitive behavioral therapy has shown efficacy in treating stress-related disorders, such as adjustment disorder and exhaustion disorder, knowledge about factors contributing to treatment response is limited. Improved identification of such factors could enhance assessment procedures and treatment strategies. In addition, evaluating how traditional prediction methods and machine learning can complement each other may help bridge gaps in understanding and predicting treatment response. Objective: This study aims to (1) evaluate putative predictors of treatment response in patients with stress-related disorders using traditional prediction methods and (2) model treatment outcomes using a machine learning approach. This design combines the interpretability of traditional methods with the ability of machine learning to identify complex patterns. Methods: We will analyze data from a randomized controlled trial comparing 2 internet-delivered treatments, cognitive behavioral therapy versus an active control treatment, for patients diagnosed with adjustment disorder or exhaustion disorder (N=300). Prediction models will be based on pooled data from both treatment arms due to the limited sample size and lack of knowledge on predictors of treatment effects. Putative predictors include sociodemographic and clinical information, clinician-assessed data, self-rated symptoms, and cognitive test scores. The primary outcome of interest is responder status on the Perceived Stress Scale-10, evaluated based on the reliable change index posttreatment. For the traditional approach, univariate logistic regressions will be conducted for each predictor, followed by an ablation study for significant predictors. For the machine learning approach, 4 classifiers (logistic regression with elastic net, random forest, support vector machine, and AdaBoost) will be trained and evaluated. The dataset will be split into training (70%) and testing (30%) sets. Hyperparameter tuning will be conducted using 5-fold cross-validation with randomized search. Model performance will be assessed using balanced accuracy, precision, recall, and area under the curve. Results: All data were collected between April 2021 and September 2022. We hypothesize that key predictors will include younger age, education level, baseline symptom severity, treatment credibility, and history of sickness absence. We anticipate that the machine learning models will outperform a dummy model predicting the majority class and achieve a balanced accuracy of ≥67%, thus indicating clinical usefulness. Conclusions: This study will contribute to the limited research on predictors of treatment outcome in stress-related disorders. The findings could support the development of more personalized and effective treatments for individuals diagnosed with adjustment disorder or exhaustion disorder, potentially improving clinical practice and patient outcomes. If successful, this dual approach may encourage future studies with larger datasets and the implementation of machine learning models in clinical settings, ultimately enhancing precision in mental health care.

The present study aimed to establish test–retest reliability and investigate practice effects of the Mindmore cognitive assessment tool, a digital adaptation of traditional pencil and paper tests designed for self-administration. Additionally, normative change scores for the most frequently used tests were derived. A total of 149 healthy Swedish adults (aged 20–79) completed the test battery twice, 1 month apart. The battery assessed attention and processing speed, memory, language, visuospatial functions, and executive functions. Test–retest reliability, measured by ICC and Spearman coefficients, and practice effects were estimated for 22 main-scores and 33 sub-scores. Regression models were used to assess change in performance while controlling for demographics, computer equipment, testing location (online or in-laboratory) and baseline performance for 12 main-scores and nine sub-scores. Test–retest reliability was good for 11 main-scores (≥0.70), satisfactory for five (0.60–0.69), and minimal for six (<0.60) albeit three having satisfactory sub-scores. Practice effects were observed for tests with a major speed component, but not for reaction time, sustained attention, verbal memory and naming (alternate forms), nor visuospatial functions. Trackpad negatively influenced change for one test. Demographics and testing location did not significantly affect the change scores. Our study provides support for test–retest reliability and practice effects of the Mindmore cognitive assessment tool which were comparable to those of traditional tests. These findings, together with the normative change scores, can aid researchers and clinicians in interpreting test results and distinguishing between normal variations in performance and changes indicative of clinical impairment.

The Perceived Stress Scale (PSS) is a widely used instrument in both clinical and research settings. A two-factor structure has been identified in most psychometric evaluations of the scale, but the implications of this factor structure for clinical trial outcomes are ill-understood. This study used data from a randomized controlled trial investigating the effect of an internet-delivered cognitive behavioral therapy (ICBT) targeting chronic stress disorders versus a waitlist control condition (N = 100). We evaluated the internal consistency of the PSS-14 (the main outcome in the randomized controlled trial) and examined the fit of a two-factor model within an exploratory structural equation modeling framework. The factor analysis supported a two-factor solution, and parallel analysis confirmed that the retention of this solution was adequate. Both the “distress” and the “coping” factors, which were identified through factor analysis, demonstrated acceptable reliability. Mixed-effects modeling was employed to quantify the impact of treatment on its respective factor, treated as separate subscales in this study. Significant interaction effects between time and treatment group were found on both distress and coping, with the ICBT group showing marked improvements in these areas. In the ICBT group, the improvement in distress was slightly more pronounced than in coping at the 6-month follow-up. We conclude that PSS-14 sum scores may be valid for assessing treatment outcomes in clinical settings. However, our findings indicate that there may be differential effects on the distress and coping factors across time, which motivates further investigation of the factors in larger clinical trials.

Background: Patients with stress-related mental disorders often report cognitive impairment, but studies investigating objective cognitive impairment in patients with stress-related disorders have produced inconsistent findings. Aim: The primary aim of this study was to investigate objective cognitive functioning in patients diagnosed with the stress-related disorders adjustment disorder or exhaustion disorder, compared to a healthy normative group. Secondary aims were to conduct subgroup analyses of cognitive functioning between the diagnostic groups and explore associations between self-reported symptoms and cognitive functioning. Methods: Cognitive test results on a digitally self-administered cognitive test battery from 266 patients (adjustment disorder, n = 131; exhaustion disorder, n = 135) were cross-sectionally compared with results from a healthy normative group (N = 184 to 692) using one-tailed t-tests. ANOVAs were conducted for subgroup analyses, and regression analyses for associations between self-reported symptoms and cognitive functioning. Effect sizes were calculated. Results: Patients performed significantly worse than the normative group on all measures with small to moderate effect sizes ranging from d =-.13 to-.57. Those diagnosed with exhaustion disorder performed worse than norms on more measures than did patients with adjustment disorder, but no significant differences between diagnostic groups were found on any measure. Self-reported memory impairment was weakly associated with one of two memory measures. No clear associations between self-reported burnout symptoms and objective cognitive functioning were found. Conclusions: This study adds to the literature indicative of small to moderate objective cognitive impairments in patients diagnosed with stress-related mental disorders. Further exploration into mechanisms of cognitive functioning in different populations is needed for development of theoretical models that may explain the weak correlation between self-reported symptoms and objective measures.