Mobile mental health (MH) apps are increasingly promoted as tools for self-care. However, less is known about how users integrate them into everyday life and how they combine with other forms of support.

This qualitative study examined what motivates individuals to adopt MH apps and how these tools are used alongside other care practices.

We conducted semistructured interviews with 31 Czech MH app users aged 18-56 (23 women), recruited through social media, researchers' networks, and snowball sampling. Eligible participants had used an MH app for at least 2 weeks; individuals currently experiencing a psychological crisis were excluded. Interviews were conducted online or face-to-face and analyzed using reflexive thematic analysis by Braun and Clarke. The analysis was inductive and informed by a critical realist epistemological stance. Initial analytic engagement was conducted independently, followed by collaborative theme development and refinement.

We identified the following three motivational patterns for MH app use: (1) regular use to maintain well-being and support preventive self-care; (2) reactive use during periods of acute distress; and (3) temporary, problem-focused use aimed at understanding or addressing a specific difficulty. Beyond motivations for adoption, our findings revealed the following four distinct trajectories of app engagement and disengagement: (1) emancipation, with apps acting as ongoing support for self-care when therapy is reduced or ended; (2) independence, as techniques learned through apps are transferred into everyday life and used without the app; (3) reliance, where users persist with app-based self-care despite insufficient benefit, occasionally delaying or limiting other forms of help; and (4) episodic use, characterized by intermittent engagement during specific life periods. These findings extend existing research by showing that disengagement is not necessarily a marker of failure but can reflect successful self-care and the transfer of learned skills. We also identified tensions and unintended consequences, including the burden of sustaining engagement, risks of overreliance, emotional costs associated with sustained self-monitoring, and tensions between digital self-care and efforts to reduce screen use. Participants generally viewed self-monitoring as beneficial when guided by clear personal goals and embedded within a broader system of self-care.

Our findings suggest that appropriate disengagement can signal a successful step in self-care. App design and policy should acknowledge episodic use and strengthen connections to human support with clear guidance and accessible pathways to additional care when needed.

Older adults commonly experience coexisting mental and physical multimorbidity, alongside frailty and polypharmacy. This clinical complexity is associated with poorer health outcomes, including higher rates of acute hospital admission, longer lengths of stay, increased dependency at discharge and higher mortality. Despite this, physical and mental health services for older adults are fragmented across Europe, often operating in a parallel rather than integrated manner. As a result, many patients receive poorly coordinated care. In response, there is growing interest in the development of integrated care services for these patients across Europe. Although a range of service models have been proposed, what constitutes "best practice" remains unclear and is likely to depend on local population needs, service configuration, healthcare context and healthcare workers education. The European Geriatric Medicine Society and European Association of Geriatric Psychiatry have initiated a Task and Finish Group to develop expert-based recommendations for integrated physical-mental health services for older adults across Europe. In this perspectives paper we consider the challenges associated with delivering integrated physical-mental healthcare for older adults, drawing on clinical and service level observations. We suggest that current approaches to integrated care are often limited in scope and variability and may not fully address the needs of older adults with combined physical and mental illness. A broader conceptualisation of integration may be required. We outline implications for future research, service development, and policy, with a focus on improving alignment between healthcare systems to address the needs of this growing population. This work was endorsed by the European Geriatric Medicine Society.

Medication adherence remains a major challenge in the long-term management of rheumatoid arthritis (RA). While clinical determinants of adherence have been widely studied, the psychological mechanisms through which individual differences influence adherence behavior are less well understood. The present study aimed to examine a theoretically informed model linking personality traits, coping-related humor styles, mental health-related quality of life, and medication adherence in patients with RA.

A cross-sectional study was conducted in a sample of adults diagnosed with RA. Participants completed established self-report measures assessing personality traits (Extraversion, Neuroticism), humor styles (self-enhancing, self-defeating), mental health-related quality of life (SF-36 Mental Component Summary), and medication adherence (Morisky Medication Adherence Scale). Ordinal logistic regression was initially performed to examine direct predictors of adherence. Subsequently, path analysis was used to test a mediation model in which mental health-related quality of life was specified as a proximal determinant of adherence.

Mental health-related quality of life emerged as the only significant direct predictor of medication adherence. Personality traits and humor styles were not directly associated with adherence. Higher Neuroticism and greater use of self-defeating humor were associated with poorer mental health-related quality of life, whereas Extraversion was positively associated with mental health-related quality of life. Path analysis demonstrated that the associations between personality traits, humor styles, and adherence were statistically accounted for by indirect pathways through mental HRQoL.

The findings indicate that mental health-related quality of life represents a central psychological pathway linking dispositional characteristics to medication adherence in rheumatoid arthritis. Rather than exerting direct effects, personality traits and coping-related humor styles influence adherence indirectly through their impact on patients' mental health. These results underscore the importance of integrating psychological assessment and support into routine rheumatology care to promote sustained adherence and optimize long-term outcomes. These findings have direct implications for routine rheumatology care, highlighting the importance of addressing mental health alongside pharmacological management. Key Points • Mental health-related quality of life (HRQoL) was the only significant direct predictor of medication adherence in patients with rheumatoid arthritis. • Personality traits and humor styles were not directly associated with adherence; their associations operated indirectly through mental HRQoL. • Higher Neuroticism and self-defeating humor were associated with poorer mental HRQoL, whereas Extraversion was associated with better mental HRQoL. • Brief psychological assessment within routine rheumatology care may help identify patients at risk for suboptimal adherence.

Psychosis exhibits marked neurobiological heterogeneity that is not adequately captured by symptom-based diagnoses. Although prior efforts have identified subgroups, it remains unclear how latent disease variability manifests within the brain's multiscale intrinsic functional organization and whether these patterns can be leveraged to systematically identify biologically driven subtypes of psychosis. Here, we aim to identify Psychosis Imaging Neurosubtypes (PINs)-homogeneous subgroups defined solely from neurobiological features of brain function, independent of clinical or cognitive definitions, and whether these subgroups show meaningful phenotypic differentiation. Resting-state fMRI data from 2103 B-SNIP 1&2 participants were used to compute subject-specific multiscale functional network connectivity (msFNC). We then derived a low-dimensional neurobiological subspace, termed Latent Network Connectivity, which captured system-wide interconnected multiscale information across three components. Projections of probands msFNC onto this subspace revealed three PINs through unsupervised learning that spanned DSM diagnoses (Schizophrenia, Bipolar, Schizoaffective Disorder). Although defined purely from intrinsic neurobiology, PINs exhibited distinct cognitive, clinical, and connectivity profiles. PIN-1, the most cognitively impaired, showed Cerebellar-Subcortical and Visual-Sensorimotor hypoconnectivity, alongside Visual-Subcortical hyperconnectivity. Most cognitively preserved PIN-2 showed Visual-Subcortical, Subcortical-Sensorimotor, and Subcortical-Higher Cognition hypoconnectivity. PIN-3 exhibited intermediate cognition, showing Cerebellar-Subcortical hypoconnectivity alongside Cerebellar-Sensorimotor and Subcortical-Sensorimotor hyperconnectivity. Although covariate-adjusted analyses did not reveal significant clinical differences among PINs, exploratory analyses identified selective differences in YMDR, PANSS, and Age of Onset. Notably, 55% of relatives aligned with the same neurosubtype as their probands, a rate significantly higher than chance and than DSM-based alignment, suggesting potential heritability. Cognitive performance reliably aligns with distinct brain connectivity patterns, which are also evident in relatives, supporting their construct validity. Our PINs partially converged with previously reported BSNIP Biotypes with some correspondence (e.g. PIN1 with Biotype 1) but not with the other Biotypes, which were determined using integrated electrophysiological, cognitive, and oculomotor data, supporting convergent validity, while also indicating that intrinsic rsfMRI derived multiscale connectivity partitions psychosis heterogeneity somewhat differently from multimodal biomarker approaches. These findings underscore the limitations of DSM-based classifications in capturing the biological complexity of psychotic disorders and highlight the potential of imaging-based neurosubtypes to enhance our understanding of the psychosis spectrum.

Musculoskeletal (MSK) pain is common among women veterans and may be exacerbated during pregnancy and postpartum. However, little is known regarding pregnant veterans' engagement with chiropractic, complementary and integrative health (CIH), and Whole Health (e.g., health and wellness coaching) services in this population.

Pregnant and postpartum veterans were identified from a longitudinal, prospective, multisite cohort study. Demographic and clinical characteristics were collected via Department of Veterans Affairs (VA) electronic health record and study survey data. Visits to VA-funded chiropractic, CIH, and Whole Health services during pregnancy, the first 9 postpartum months, and the 5 years preceding pregnancy were identified. We described participants' engagement with these services and assessed factors associated with use of any of these services during pregnancy and/or postpartum using logistic regression.

Of the 1,139 veterans included, 109 (10%) engaged with chiropractic, CIH, and/or Whole Health services during pregnancy and/or postpartum. Chiropractic (53%), acupuncture (20%), and massage therapy (10%) were the most used services. Visits to services were similar between pregnant (54%) and postpartum (46%) timeframes. Engagement with these services in the 5 years preceding pregnancy (odds ratio [OR]: 6.90; 95% confidence interval [CI] [4.24, 11.25]) and presence of a mental health diagnosis (anxiety, depression, and/or posttraumatic stress disorder) (OR: 1.79; 95% CI [1.02, 3.17]) were significantly associated with chiropractic, CIH, and/or Whole Health service use during pregnancy and/or postpartum. All other demographic and clinical variables were nonsignificant.

Results from this study may inform future implementation and administrative investigations of pregnant and postpartum veterans' use of chiropractic, CIH, and Whole Health services.

Limited recent literature has evaluated the use of large language models (LLMs) in the qualitative analysis of health data; more research is needed to expand the generalizability of LLM use and to evaluate potential ethical considerations.

Our research sought to (1) describe the process of using AI to analyze qualitative in-depth interview data, (2) identify similarities and differences between the human and AI-generated analyses to compare the quality and rigor of the two techniques and describe the strengths and weaknesses of each approach, and (3) make recommendations regarding the bounds of ethics and the role of researcher bias in AI-assisted qualitative research.

Nested within a larger mixed methods study, 17 Ugandan women recovering from female genital fistula surgery participated in hour-long semistructured interviews exploring their mental, physical, and overall health trajectories. Each interview lasted about an hour and was audio-recorded. Following translation and transcription, the data underwent thematic analysis by humans and by Versa, a University of California, San Francisco (UCSF) developed LLM powered by ChatGPT-4o. The AI analysis was conducted using 2 strategies: inductively and deductively. Finally, the analytic outputs were compared by MP and reviewed by AME and MG to evaluate the code frequency and alignment (most frequently used codes and equivalent concepts vs unique concepts between human and AI codes), thematic robustness (differentiation of distinct concepts and labels that convey substantive findings), analysis quality (narrative depth and excerpt accuracy), and efficiency of each method (total person-hours spent on comparable analytic tasks).

A comparative analysis revealed significant thematic overlap between the human-synthesized and AI-generated outputs, though notable differences in granularity and efficiency emerged. When inductively coding, Versa identified 39 codes, whereas human researchers used a more expansive set of 54 codes. While Versa's thematic analyses were generally accurate regarding how the interview data were represented in the analyses, the human-synthesized analysis was more robust because it incorporated compelling excerpts, and the summary content provided greater depth and range that the Versa analyses lacked. The disparity in efficiency was stark: the human analysis took approximately 15 hours to complete, while Versa produced the analysis in about 2.5 hours.

While Versa significantly expedited the initial coding phase, it still relied heavily on human researchers to create appropriate prompts and input all the data into the chat. Versa's inability to replicate the narrative depth and description of human synthesis suggests that LLMs currently lack the interpretive sensitivity required to capture the lived experiences. Using Versa alone does not currently yield a high-quality analysis; significant human engagement is needed to maintain ethical and interpretive rigor.

ClinicalTrials.gov NCT05437939; https://clinicaltrials.gov/study/NCT05437939.

Large language model-based chatbots (LLM-CBs) are increasingly used as mental health support tools. Risks and harms are discussed especially for unsupervised use and for vulnerable groups. A naturalistic characterization of the use of LLM-CBs by patients with mental health disorders is currently lacking.

The goal of this cross-sectional study is to investigate demographic and clinical characteristics of users of LLM-CB for mental health and their associations with use behavior.

Across 6 months, qualitative and quantitative data on LLM-CB use for mental health were collected in a consecutive sample of psychotherapy outpatients in routine treatment with at least 1 diagnosed mental health disorder. Results are reported descriptively and supported by exploratory inferential analyses.

Within a total of 812 included outpatients who reported on LLM-CB use for mental health, 245 (30.1%) reported having used LLM-CBs for mental health (once: 69/217, 31.8%; weekly: 118/217, 54.4%; and daily: 30/217, 13.8%). Users were more likely to be younger, female, better educated, students, or unemployed. In total, 61.4% (121/197) have used LLM-CBs in self-reported personal crisis situations, of whom, 67.7% (82/121) found the experience unhelpful. LLM-CB use showed a weak, imprecise association with the number of diagnoses in the included diagnosis groups and was higher for anxiety and eating disorder diagnoses and lower for somatoform disorders. Post hoc multivariable analyses suggest that higher use in eating disorders and lower use in somatoform disorders were attenuated after adjusting for age, while higher use for anxiety disorders persisted. LLM-CBs were generally evaluated positively across various dimensions, especially when self-reported personal crisis situation use was experienced as helpful. Qualitative analysis of an open-ended question on use experience shows LLM-CB use across 4 domains reported in the selected subsample: general psychological assistance, disorder-specific assistance, information assistance, and everyday assistance.

This naturalistic characterization of LLM-CB use in mental health outpatients implies that harms and benefits of LLM-CB use are associated with demographic and clinical profiles. The results suggest that asking about LLM-CB use may be a useful consideration in psychosocial assessment, particularly regarding crisis use, symptom-reinforcing use, and replacement of human support.

Cognitive-behavioural therapy (CBT) is a widely used intervention for depression and anxiety. However, its effectiveness is typically evaluated using subjective self-report symptom scales, which are susceptible to bias. It is hypothesised that improvements in clinical symptoms following CBT may be accompanied by improvements in physiological regulation, particularly heart rate variability (HRV), which has been proposed as a marker of emotion regulation. We conducted a systematic review with meta-analysis to examine the effects of CBT on HRV in adults with depression and/or anxiety disorders.

Following a pre-registered protocol (PROSPERO: CRD420251040957), we searched Embase, PsycINFO, PubMed, and Web of Science for randomised controlled trials (RCTs) and non-randomised interventional studies involving adults with depression and/or anxiety in CBT, and reporting HRV before and after the intervention. Random-effects meta-analyses in R estimated pooled effect sizes (Hedges' g).

After screening 9713 records, we identified 12 studies; 5 RCTs were included in the meta-analysis, and the remaining 7 in the narrative review because they did not provide calculable effect sizes. Meta-analysis results found that adults randomised to CBT had significantly greater pre-to-post intervention increases in HF-HRV, LF-HRV, and SDNN-HRV. No significant changes were observed for the LF/HF ratio, RMSSD-HRV, and the HRV Triangular Index.

Our meta-analysis provides emerging evidence that CBT might significantly increase selected HRV indices, suggesting that physiological changes may accompany improvements in psychological symptoms. However, further high-quality research is needed to determine the clinical utility of HRV as a predictor of treatment response and a biomarker for treatment effectiveness.

Sleep disturbances are common in severe mental disorders (SMD), but community-based epidemiological data in China are scarce. This study examined the prevalence and correlates of poor sleep quality and tested the indirect pathways of psychological distress in the association between social support and sleep.

Using 2022 baseline data from a community survey in Shanghai, we included 3797 SMD patients (mean age 54.9 years; 51.6% female; 64.1% schizophrenia-spectrum). Sleep quality was assessed with the Pittsburgh Sleep Quality Index (PSQI; poor defined as >5), social support with PSSS, anxiety with GAD-7, and depression with PHQ-9. Hierarchical multivariable regression, bootstrap mediation (5000 resamples), and K-means clustering were applied.

Mean PSQI was 5.14, and 39.3% had poor sleep. The full model explained 27.8% of variance. PHQ-9 was the strongest independent correlate of poor sleep. Comorbidity (OR = 1.48, 95% CI 1.23-1.77) and lower household income increased risk, whereas no hospitalization history was associated with lower odds of poor sleep. The indirect effect via depression accounted for 81.4% of the total association between social support and sleep, while that via anxiety accounted for 57.8%; the PSSS×PHQ-9 interaction was significant. Three sleep phenotypes were identified: mild (36.1%, mean PSQI 2.7), moderate (43.4%, mean 4.8), and severe (20.5%, mean 10.2), with PHQ-9 as the strongest discriminator.

Poor sleep is highly prevalent in community SMD patients. Depression appears central in the indirect pathway from social support to sleep. Integrating depression management, social support enhancement, and phenotype-based stratification warrants further investigation to improve sleep.