Objective: Early identification of failing psychological treatments could be of high clinical value, but therapists themselves have been found to be bad at predicting who will benefit or not. Previous research has some methodological limitations, and therapists' predictive accuracy has never been examined in internet-delivered treatments. Method: Therapists providing internet-delivered cognitive behavior therapy for depression, social anxiety disorder, and panic disorder in routine psychiatric care made outcome predictions for 897 patients during the fourth week of treatment. Therapists' accuracies were also compared to the accuracy of a simple statistical model and benchmarks for clinically acceptable/useful levels of accuracy from previous research. Results: Therapists were more accurate than chance, but their balanced accuracy was on average nine percentage points lower than the balanced accuracy of the statistical model (though confidence intervals often overlapped) and only in one case did the predictions reach the clinical acceptance and utility benchmarks. Therapists could predict on average 16% of the variance in outcome. Therapists were overly optimistic, predicting positive outcomes on average twice as often as they occurred. They differed in confidence in their predictions, though this did not affect how correct they were. Conclusions: Internet-delivered cognitive behavior therapy-therapists can often predict treatment outcomes better than chance, but generally not as well as the statistical model, and probably not accurately enough that they would be willing to act on their predictions, or that they could be used in an adaptive treatment strategy. Our previous findings suggest that patients would benefit from statistical monitoring and prediction tools in clinical settings.

Background: Symptom reduction occurring early in depression treatment is associated with favourable posttreatment outcome, but it is not known how early reduction in specific depression symptoms affect treatment outcome. We aimed to determine the impact of symptom-specific change from pre-treatment to week four during internet-delivered CBT (ICBT) on overall and symptom-specific depression severity at post-treatment. We hypothesized that change in mood and emotional involvement would be most strongly associated with later overall depression severity.

Methods: 1300 participants with Major Depressive Disorder were followed over 12 weeks of ICBT using the selfreport Montgomery-& Aring;sberg Depression Rating Scale gauging nine symptoms. Linear models, informed by causal inference and cross-lagged network analysis methods, were used to estimate associations between early symptom-specific change and post-treatment depression severity, controlling for register-based and self-reported pre-treatment confounders.

Results: Early reduction in all symptoms was associated with lower overall and symptom-specific depression severity post-ICBT. Seven symptoms showed similar associations between early change and overall depression severity post-treatment: mood (standardized beta [(3] = 0.44), feelings of unease ((3 = 0.39), ability to concentrate ((3 = 0.46), initiative ((3 = 0.43), emotional involvement ((3 = 0.42), pessimism ((3 = 0.44), and zest for life ((3 = 0.42). Change in sleep ((3 = 0.27) and appetite ((3 = 0.27) had weaker associations with overall depression severity at post-treatment and were the only symptoms showing the hypothesized difference compared with mood and emotional involvement.

Conclusions: The impact of early symptom-specific reduction on post-treatment depression severity in ICBT for MDD may be similar across most symptoms, but less for the sleep and appetite symptoms, although causal interpretations rests on several assumptions.

Digital self-help can improve access to mental health care, but poor engagement limits effectiveness. This single-blind 2 x 2 x 2 factorial randomized controlled trial examined whether an optimized graphical user interface (GUI), automated reminders (AR), and an adaptive treatment strategy (ATS) improved engagement and outcomes in a digital self-help insomnia intervention. Adults (N = 447) with moderate to severe insomnia were randomized to combinations of the factors. The GUI improved self-rated engagement, sleep log activity, login frequency, and usability. AR increased sleep log activity and logins, while ATS improved satisfaction. All three combined significantly improved insomnia symptoms (d = 0.50). No severe adverse effects were reported. Clinicians spent 13.74 min on average on the ATS. Statistical analyses included linear and multilevel regression. Factors were effect coded. Intervention design can enhance engagement and outcomes, requiring minimal clinician time. Pre-registered 2023-04-11 (ClinicalTrials.gov, NCT05826002). Funded by the Swedish Ministry of Health and Social Affairs, grant number: S2018/03855/FS.

Background: There has recently been an increased emphasis on patient empowerment and collaboration within their healthcare. However, there is widely a lack of clarity to the concept of empowerment and existing measurement tools lack uniformity, covering diverse domains and related concepts.

Objectives: This study aims to conduct a psychometric evaluation of the Swedish version of the Empowerment Scale- Making Decisions, focusing on its structural validity and reliability in assessing patient empowerment. This includes a detailed examination of the factor structure across two different contexts, psychiatric care (n = 211) and primary care (n = 210). We will compare several confirmatory factor analysis (CFA) models proposed in previous research to identify the best fit. If no models provide a good fit, we intend to suggest a new scale for further evaluation.

Method: The dimensionality of the scale was tested by comparing four CFA models, together with a one-factor solution, to identify the best fit for the two samples. Reliability measures were determined by coefficient Omega (omega) as well as Cronbach's alpha (alpha).

Results: There was limited support for the one-factor solution in both samples, challenging the scale's assumed unidimensionality (primary care sample: x2(350) = 1074, p <.001, CFI = 0.58, TLI = 0.54, RMSEA = 0.10 (90% CI: 0.09 - 0.11), SRMR = 0.11; psychiatric care sample: (x²(350) = 1307, p = < 0.001, CFI = 0.66, TLI = 0.63, RMSEA = 0.11 (90% CI:0.11;0.12), SRMR = 0.10). None of the previously suggested factor solutions demonstrated satisfactory fit. However, a three factor-solution entailed the less complexity and best model fit (primary care sample: (x²(270) = 503, p = < 0.001),CFI = 0.85, TLI = 0.84, RMSEA = 0.06 (90% CI 0.06;0.07), SRMR = 0.07; psychiatric care sample: (x²(270) = 622, p <.001), CFI = 0.87, TLI = 0.86, RMSEA = 0.08 (90% CI 0.07;0.09), SRMR = 0.07). Based on this, we continued with exploratory refinements of this solution and arrived at two adjusted three-factor models based on each sample. These two adjusted models displayed only slight differences, and in a last step we removed the items that differed between the samples to arrive at one solution appropriate for use in health care settings in general. As a result, an improved and shortened adaptation of the scale was put forward that included 18 items targeting the subscales Self-Esteem, Powerlessness and Activism. This solution remained relatively clear to the previously proposed solutions (primary care sample:(x²(131) = 240, p <.001), CFI = 0.91, TLI = 0.90, RMSEA = 0.06 (90% CI 0.05;0.08), SRMR = 0.07; psychiatric care sample: (x²(131) = 379, p <.001), CFI = 0.88, TLI = 0.86, RMSEA = 0.09 (90% CI 0.08;0.10), SRMR = 0.07; combined sample: (x²(131) = 432, p <.001), CFI = 0.91, TLI = 0.90, RMSEA = 0.07 (90% CI 0.07;0.08), SRMR = 0.06).

Conclusion: The results reinforce the difficulties in measuring empowerment given the complexity of this concept. The improved and shortened adaptation of the scale could potentially be used within health care settings to measure empowerment, but further research is needed to conceptualize and measure empowerment in patients with mental health problems. Given scarce support for the scale's unidimensionallity, future research should explore using multiple instruments targeting different constructs to measure patient empowerment more comprehensively.

Background: While protocol-based psychological treatments have significantly advanced mental health care, real-world accessibility remains a challenge. Primary care, the main provider of mental health services, faces barriers such as limited resources and a diverse patient population with varying needs, making it difficult to rely solely on time-intensive, protocolized treatments. The Primary Care Behavioral Health (PCBH) model promotes brief, flexible interventions that may better accommodate these needs. However, limited research on these interventions raises concerns about potential undertreatment. To align with Universal Health Coverage principles, it is essential to identify which patient groups benefit most from resource-efficient protocol-based versus brief, flexible, and individualized treatments. Our main aim is to evaluate whether a integrating guided self-help into PCBH improves outcomes compared to the core PCBH model, as well as to assess whether patients identified as suitable for protocol-based interventions benefit more from the combined model.

Methods: Patients seeking help for mental or behavioral health problems at PCBH primary care centers will be randomized to one of two arms: core PCBH, where patients receive a contextual assessment and brief interventions tailored to their needs, or an extended PCBH model, where a diagnostic assessment determines whether patients receive brief interventions or guided self-help. The primary outcome is functional impairment, assessed at baseline and followed up at 4, 8, and 12 weeks (primary endpoint), as well as at 1 year. Secondary outcomes include symptom change, cost-effectiveness, and care process factors.

Discussion: The study design allows for comparisons of patient outcomes between the two care models, with a primary focus on evaluating superiority and a secondary focus on non-inferiority, cost-effectiveness, and care process factors. Overall, the project seeks to advance understanding of effective mental health interventions in primary care settings and inform decision-making regarding treatment approaches.

Insomnia is prevalent, emphasizing the need for effective and sustainable treatments. While short-term use of sleep medication is recommended, long-term use remains common, underscoring the necessity for psychological treatments like Cognitive Behavioral Therapy for Insomnia (CBT-I) in clinical practice. This study aimed to evaluate the effectiveness of guided Internet-Based Cognitive Behavioral Therapy for insomnia (ICBT-I) when integrated into general practice. Participants (n = 177) were recruited from 33 primary health care centers (PCCs) and enrolled in an eight-week guided ICBT-I program. Eligible participants were at least 18 years old and reported sleep problems significantly affecting their daily lives. Significant reductions in insomnia were observed, with large improvements in sleep disturbances. ISI scores decreased significantly from pre- to post-treatment (beta = 9.368, p _ .001, Hedges' g = 1.40). Depression (beta = 5.496, g = 0.68) and anxiety (beta = 3.982, g = 0.56) also showed moderate improvements (p _ .001). All sleep diary measures improved significantly (p _ .001), and sleep medication use dropped from 48.6% at pretreatment to 17.5% at posttreatment (p _ .001). These findings suggest that guided ICBT-I in primary care effectively reduces insomnia and improves mental health, with outcomes comparable to specialized care.

Study Objectives: Insomnia disorder, affecting 10% of the population, poses a significant public health concern and is a risk-factorfor many health issues. Cognitive behavioral therapy is first-choice treatment, but the key component—sleep restriction therapy—presents with side effects and adherence challenges. Sleep compression therapy, suggested as a potentially gentler alternative, hasnever been directly compared to sleep restriction therapy.

Methods:  Single-blind trial at the Internet Psychiatry Clinic in Stockholm, Sweden. Patients with insomnia disorder were randomized1:1 to evaluate non-inferiority of sleep compression therapy to sleep restriction therapy in improving insomnia and to compareimportant clinical aspects. Primary outcome: self-reported Insomnia Severity Index (ISI), assessed pretreatment, weeks 1–5, and week10. Non-inferiority analysis based on intent-to-treat analyses with multiple imputation and mixed effects models.

Results: Adults with insomnia (n = 234; mean age 44.3 [SD = 13.7] years, 173 [73.4%] female) received treatment as a 10-week highly structured, therapist-guided online program, to strengthen experimental integrity and treatment fidelity. Both treatments improved insomnia severity with large effects. Sleep compression therapy failed to show non-inferiority with a conservative limit of 1.6 ISI-points (95% CI: -0.01, 1.70), gave statistically significantly smaller improvements (p = .006), and was associated with slower improvements despite better adherence and somewhat less side effects.

Conclusions: This direct comparison and well-controlled trial provides empirically based support for clinicians to prioritize sleep restriction therapy over sleep compression therapy, while the latter can be a valid alternative when sleep restriction therapy cannot be used.

Clinical Trial: CompRest-a Comparison Between Sleep Compression and Sleep Restriction for Treating Insomnia.

Objective: Predicting treatment outcome has the potential to enhance Internet-delivered Cognitive Behavioral Therapy (ICBT). One aspect of guided ICBT is patient-therapist interaction through written messages. With Natural language processing (NLP) these could be leveraged to predict outcome; however current evidence is limited. This study investigates the predictive accuracy of NLP models for treatment outcomes and evaluates whether NLP provides additional predictive value beyond symptom variables.

Methods: Patient-therapist messages from 6613 patients undergoing 12 weeks of treatment were used to train three types of NLP models: Term Frequency-Inverse Document Frequency (TF-IDF), Bidirectional Encoder Representations from transformers (BERT), and BERT for Longer Text (BELT). These were trained both with and without symptom variables from the initial treatment period to predict post-treatment symptoms. A dummy model was also used, and a linear regression model acted as a symptoms only benchmark. Multiple imputation addressed missing data, and nested cross-validation was used.

Results: The symptom only model performed best. Only BERT outperformed the dummy model, achieving a Root Mean Squared Error (RMSE) of 0.17 compared to RMSE of 0.18. Adding symptom variables to the BERT model significantly increased its accuracy, but not the RMSE metric. The best linear regression benchmark based on symptoms only had a BACC of 70 % (F1-score of 0.66) which outperformed the BERT model with 60 % (F1: 0.55) and the combined BERT plus symptoms model achieved 68 % (F1: 0.62).

Conclusion: These initial findings indicate a small predictive value from patient-therapist written message interaction but added no value beyond using only symptoms to predict post-treatment symptoms. Further research is needed to refine NLP-methods for use in psychological treatment, and more accurately assess the predictive potential of text-based interactions during ICBT.

Background: Insomnia and depression are common, comorbid conditions with cognitive consequences. Cognitive behavioral therapy for insomnia (CBT-I) improves subjective cognition, but effects on objective performance are unclear. This study aims to examine cognitive differences in insomnia with and without comorbid depression, and changes in cognition following CBT-I.

Methods: This study examined cognitive outcomes in 170 (124 for longitudinal analyses) participants from two randomized clinical trials of internet-delivered 9 or 12 weeks CBT-I for patients with insomnia (n = 78) or comorbid insomnia and depression (n = 92). Cognitive performance was assessed pre- and post-treatment using selected computerized cognitive tests from the CANTAB battery. Linear regression and mixed-effects models were used for evaluation.

Results: Following CBT-I, improvements were statistically significant on seven out of 17 outcomes across patients with and without comorbid depression: the Rapid Visual Processing task (correct hits, p < .001, misses, p < .001, latency, p = .040), Stockings of Cambridge (problems solved on first choice, p = .042), and Affective Go/No gGo (commissions, p = .005; omissions, p = .022; affective bias, p = .020). Comorbid depression was not statistically significantly associated with cognitive performance on most tasks. However, on the Spatial Span task, comorbid depression was associated with lower span length (p = .038), fewer attempts (p = .021), and longer latency (p = .030), suggesting impaired spatial working memory in depressed individuals. No statistically significant associations were found between changes in insomnia or depression severity and changes in cognitive performance after treatment.

Conclusions: Differences in cognitive performance between patients with insomnia and patients with both insomnia and depression seem very small. CBT-I may be associated with improvements in objective cognitive performance across some domains, including attention, working memory, executive function, and emotional processing. Future research should employ better control for e.g. practice effects, and explore any long-term cognitive effects of CBT-I.

Background: The 9-item self-rated Montgomery-& Aring;sberg Depression Rating Scale (MADRS-S) is commonly used for depression assessment worldwide, but its psychometric properties have not been thoroughly evaluated among psychiatric patients. Aims: To 1) evaluate the psychometric properties of the original MADRS-S and 2) derive new improved versions of MADRS-S by excluding problematic items. Methods: We included 1286 individuals (Mage = 37.8, 66 % women) diagnosed with major depressive disorder at the largest outpatient internet psychiatry unit in Sweden. Psychometric properties were evaluated and optimized using factor analysis and item response theory methods. Results: Exploratory and confirmatory factor analysis showed a one-factor or a bi-factor structure with a general factor suggesting essential one-dimensionality (common variance explained = 84 %). MADRS-S showed acceptable reliability (Molenaar-Sijtsma index [MS] = 0.77, Omega Hierarchical [OmegaH] = 0.81) but weak scalability (H = 0.31). Removing the two worst-performing items, sleep and appetite, created the MADRS-S-7 with improved scalability (H = 0.39) and reliability (MS = 0.79, OmegaH = 0.80). A further reduction formed the three-item MADRS-S-3 including only hallmark depression symptoms-mood, initiative, and emotional involvement-that demonstrated the strongest scalability (H = 0.53) and acceptable reliability (MS = 0.75, OmegaH = 0.74). Conclusions: The original MADRS-S show essential one-dimensionality and acceptable reliability but poor scalability and some low-performing items. The new MADRS-S-7 retains most of the conceptual scope with superior psychometric properties. Furthermore, the novel MADRS-S-3 is the optimal choice when focusing on core depression symptoms, with minimized rating burden for patients and maximized ability to rank respondents on the latent variable by their summed score. The present findings warrant external validation across different psychiatric contexts.