Identifying depression-related topics in smartphone-collected free-response speech recordings using an automatic speech recognition system and a deep learning topic model.

BACKGROUND: Prior research has associated spoken language use with depression, yet studies often involve small or non-clinical samples and face challenges in the manual transcription of speech. This paper aimed to automatically identify depression-related topics in speech recordings collected from clinical samples. METHODS: The data included 3919 English free-response speech recordings collected via smartphones from 265 participants with a depression history. We transcribed speech recordings via automatic speech recognition (Whisper tool, OpenAI) and identified principal topics from transcriptions using a deep learning topic model (BERTopic). To identify depression risk topics and understand the context, we compared participants' depression severity and behavioral (extracted from wearable devices) and linguistic (extracted from transcribed texts) characteristics across identified topics. RESULTS: From the 29 topics identified, we identified 6 risk topics for depression: 'No Expectations', 'Sleep', 'Mental Therapy', 'Haircut', 'Studying', and 'Coursework'. Participants mentioning depression risk topics exhibited higher sleep variability, later sleep onset, and fewer daily steps and used fewer words, more negative language, and fewer leisure-related words in their speech recordings. LIMITATIONS: Our findings were derived from a depressed cohort with a specific speech task, potentially limiting the generalizability to non-clinical populations or other speech tasks. Additionally, some topics had small sample sizes, necessitating further validation in larger datasets. CONCLUSION: This study demonstrates that specific speech topics can indicate depression severity. The employed data-driven workflow provides a practical approach for analyzing large-scale speech data collected from real-world settings.

Trends in burnout and related sick leave duration in working-age adults followed in general practices in Germany between 2012 and 2022.

AIMS: This study investigated burnout trends and related sick leave duration in working-age adults followed in general practices in Germany. METHODS: Participants were aged 20-65 years and were diagnosed with burnout in one of 442 general practices in this country in 2012-2014, 2016-2018, and 2020-2022. The prevalence of burnout and the duration of sick leave in all practices were compared between 2012-2014 and 2016-2018 and between 2016-2018 and 2020-2022. The association between diagnosis years (i.e., 2012-2014, 2016-2018, and 2020-2022) and long-term sick leave (i.e., more than 42 days of sick leave) was investigated using adjusted logistic regression. RESULTS: The study included 39,793, 46,708, and 50,721 patients diagnosed with burnout in the periods from 2012 to 2014, 2016-2018, and 2020-2022, respectively, and the prevalence of burnout in all patients was 3.6%, 3.8%, and 3.6% during these three periods. This prevalence increased significantly in people aged 61-65 years and those with a Charlson Comorbidity Index of ≥1. The mean (SD) sick leave duration in the population of people with burnout rose from 24.1 (41.9) days in the period of 2012-2014 to 36.2 (65.6) days in the period from 2020 to 2022. Finally, compared with 2012-2014, burnout diagnosed in 2016-2018 and 2020-2022 was associated with higher odds of long-term sick leave. CONCLUSIONS: In view of these findings, it is clear that public health measures are urgently needed to improve burnout prevention in Germany.

Health service and psychotropic medication use for mental health conditions among healthcare workers active during the Spain Covid-19 Pandemic - A prospective cohort study using web-based surveys.

Little is known about healthcare workers' (HCW) use of healthcare services for mental disorders. This study presents data from a 16-month prospective cohort study of Spanish HCW (n = 4,809), recruited shortly after the COVID-19 pandemic onset, and assessed at four timepoints using web-based surveys. Use of health services among HCW with mental health conditions (i.e., those having a positive screen for mental disorders and/or suicidal thoughts and behaviours [STB]) was initially low (i.e., 18.2 %) but increased to 29.6 % at 16-month follow-up. Service use was positively associated with pre-pandemic mental health treatment (OR=1.99), a positive screen for major depressive disorder (OR=1.50), panic attacks (OR=1.74), suicidal thoughts and behaviours (OR=1.22), and experiencing severe role impairment (OR=1.33), and negatively associated with being female (OR = 0.69) and a higher daily number of work hours (OR=0.95). Around 30 % of HCW with mental health conditions used anxiolytics (benzodiazepines), especially medical doctors. Four out of ten HCW (39.0 %) with mental health conditions indicated a need for (additional) help, with most important barriers for service use being too ashamed, long waiting lists, and professional treatment not being available. Our findings delineate a clear mental health treatment gap among Spanish HCW.

Association between tuberculosis and psychotic experiences: Mediating factors and implications for patient care in low- and middle-income countries.

Background: Tuberculosis may play a role in the aetiology of psychosis. However, little is known about the association between tuberculosis and psychotic experiences (PEs) or the mediating factors of this association. Methods: We analysed cross-sectional data from 48 low- and middle-income countries of the World Health Survey (WHS). Tuberculosis assessment was based on past 12-month symptoms of active tuberculosis. We assessed four types of past 12-month PEs with the Composite International Diagnostic Interview. We performed multivariable multinomial logistic regression and mediation analysis. Results: We analysed data on 224 842 individuals aged ≥18 years (mean age = 38.3 years, standard deviation = 16.0; 50.7% women). Tuberculosis was associated with 1.84 (95% confidence interval (CI) = 1.41-2.40), 2.18 (95%CI = 1.58-3.03), and 3.79 (95%CI = 2.88-4.98) times higher odds for 1, 2, and ≥3 PEs, respectively. The mediation analysis showed that the association between tuberculosis and at least one PE is mainly explained by anxiety (31.5%), sleep/energy (27.8%), and pain/discomfort (23.5%). Conclusions: Tuberculosis was associated with increased odds of PEs. Factors such as affect, sleep, and pain should be considered in tuberculosis patients to target those who might be particularly vulnerable to PEs, and consequently, to psychotic disorders and other adverse effects of PE.

Physical Activity Participation Decreases the Risk of Depression in Older Adults: The ATHLOS Population-Based Cohort Study.

BACKGROUND: To which extent physical activity is associated with depression independent of older adults' physical and cognitive functioning is largely unknown. This cohort study using harmonised data by the EU Ageing Trajectories of Health: Longitudinal Opportunities and Synergies consortium, including over 20 countries, to evaluate the longitudinal association of physical activity (light-to-moderate or vigorous intensity) with depression in older adults (aged ≥ 50 years). RESULTS: We evaluated 56,818 participants (light-to-moderate models; 52.7% females, age 50-102 years) and 62,656 participants (vigorous models; 52.7% females, age 50-105 years). Compared to never, light-to-moderate or vigorous physical activity was associated with a lower incidence rate ratio (IRR) of depression (light-to-moderate model: once/week: 0.632, 95% CI 0.602-0.663; twice or more/week: 0.488, 95% CI 0.468-0.510; vigorous model: once/week: 0.652, 95% CI 0.623-0.683; twice or more/week: 0.591, 95% CI 0.566-0.616). Physical activity remained associated with depression after adjustment for the healthy ageing scale, which is a scale that incorporated 41 items of physical and cognitive functioning (light-to-moderate model: once/week: 0.787, 95% CI 0.752-0.824; twice or more/week: 0.711, 95% CI 0.682-0.742; vigorous model: once/week: 0.828, 95% CI 0.792-0.866; twice or more/week: 0.820, 95% CI 0.786-0.856). CONCLUSIONS: Physical activity, of any intensity and weekly frequency, was a strong protective factor against depression, independent of physical and mental functioning. Health policies could stimulate the incorporation of lower physical activity intensity to protect against depression, which might be more feasible at the population level.

Dynapenic Abdominal Obesity Increases Risk for Falls Among Adults Aged ≥50 Years: A Prospective Analysis of the Irish Longitudinal Study on Ageing.

BACKGROUND: There is a scarcity of studies examining the longitudinal relationship between dynapenic abdominal obesity (DAO; ie, impairment in muscle strength and high waist circumference) and future fall risk. Therefore, we aimed to investigate the prospective association between DAO at baseline and falls occurring during 2 years of follow-up in a nationally representative sample of middle-aged and older individuals from Ireland. METHODS: Data from 2 consecutive waves of the Irish Longitudinal Study on Ageing survey were analyzed. Dynapenia was defined as handgrip strength of <26 kg for men and <16 kg for women. Abdominal obesity was defined as a waist circumference of >88 cm for women and >102 cm for men. DAO was assessed at Wave 1 (2009-2011) and was defined as having both dynapenia and abdominal obesity. Falls occurring between Wave 1 and Wave 2 (2012-2013) were self-reported. Multivariable logistic regression analysis was conducted. RESULTS: Data on 5 275 individuals aged ≥50 years were analyzed (mean [standard deviation {SD}] age 63.2 [8.9] years; 48.8% males). After adjustment for potential confounders, compared to no dynapenia and no abdominal obesity at baseline, DAO was significantly associated with 1.47 (95% confidence interval [CI]: 1.14-1.89) times higher odds for falls at 2-year follow-up. Dynapenia alone (odds ratio [OR] = 1.08; 95% CI: 0.84-1.40) and abdominal obesity alone (OR = 1.09; 95% CI: 0.91-1.29) were not significantly associated with falls at follow-up. CONCLUSIONS: DAO increased the risk for falls among middle-aged and older adults in Ireland. Interventions to prevent or reverse DAO may be beneficial for fall reduction.

Patterns, predictors, and patient-reported reasons for antidepressant discontinuation in the WHO World Mental Health Surveys.

BACKGROUND: Despite their documented efficacy, substantial proportions of patients discontinue antidepressant medication (ADM) without a doctor's recommendation. The current report integrates data on patient-reported reasons into an investigation of patterns and predictors of ADM discontinuation. METHODS: Face-to-face interviews with community samples from 13 countries (n = 30 697) in the World Mental Health (WMH) Surveys included n = 1890 respondents who used ADMs within the past 12 months. RESULTS: 10.9% of 12-month ADM users reported discontinuation-based on recommendation of the prescriber while 15.7% discontinued in the absence of prescriber recommendation. The main patient-reported reason for discontinuation was feeling better (46.6%), which was reported by a higher proportion of patients who discontinued within the first 2 weeks of treatment than later. Perceived ineffectiveness (18.5%), predisposing factors (e.g. fear of dependence) (20.0%), and enabling factors (e.g. inability to afford treatment cost) (5.0%) were much less commonly reported reasons. Discontinuation in the absence of prescriber recommendation was associated with low country income level, being employed, and having above average personal income. Age, prior history of psychotropic medication use, and being prescribed treatment from a psychiatrist rather than from a general medical practitioner, in comparison, were associated with a lower probability of this type of discontinuation. However, these predictors varied substantially depending on patient-reported reasons for discontinuation. CONCLUSION: Dropping out early is not necessarily negative with almost half of individuals noting they felt better. The study underscores the diverse reasons given for dropping out and the need to evaluate how and whether dropping out influences short- or long-term functioning.

Design and development of a digital intervention for workplace stress and mental health (EMPOWER).

Purpose: We describe the design and development of the European Platform to Promote health and wellbeing in the workplace (EMPOWER) digital intervention that provides an integrative user programme meeting the needs of employees and employers in addressing work stress. Results: A user-centred design process was followed from January 2020 until November 2021. A tailored algorithm was developed to provide support at the individual employee level and the company level. Each element of the digital intervention was developed in English and then translated in Spanish, English, Polish and Finnish. The digital intervention consists of a website and a mobile application (app) that provides algorithm-based personalised content after assessing a user's somatic and psychological symptoms, work functioning, and psychosocial risk factors for work stress. It has a public section and an employer portal that provides recommendations to reduce psychosocial risks in their company based upon clustered input from employees. Usability testing was conducted and showed high ease of use and completion of tasks by participants. Conclusion: The EMPOWER digital intervention is a tailored multimodal intervention addressing wellbeing, work stress, mental and physical health problems, and work productivity. This will be used in a planned RCT in four countries to evaluate its effectiveness.

Chronic and Transient Loneliness in Western Countries: Risk Factors and Association With Depression. A 2-Year Follow-Up Study.

INTRODUCTION: Our aim was to test risk factors for chronic and transient loneliness as well as cross-sectional and longitudinal associations of courses of loneliness with depression. METHODS: Responses from participants in Wave 5 (T1, 2013) and Wave 6 (T2, 2015) of The Survey of Health, Ageing and Retirement in Europe (SHARE) (N = 45,490) were analyzed. The existence of clinically significant symptoms of depression was defined as reporting a value greater than or equal to 4 on the Euro-D scale. Loneliness was measured through the 3-item UCLA loneliness scale and a single question. Both measures were tested in separate regression models to identify risk factors for transient (loneliness at T1) and chronic (loneliness at T1 and T2) loneliness as well as their associations with depression. RESULTS: Chronic loneliness was observed in 47%-40% of the cases of loneliness, according to the UCLA scale and the single question, respectively. Risk factors for chronic loneliness in both models were being female, not being married, having a low educational level, having poor mental and physical health, being limited in activities, having a poor social network, and living in a culturally individualistic country. Risk factors for transient loneliness were less robust and no significant effects were found for variables such as sex and physical health in both models, education level in the UCLA measure model, and social network size in the single question model. Chronic loneliness also showed a strong association with depression in the cross-sectional model and a marked one in the longitudinal model. CONCLUSION: The courses of loneliness are relevant in the study of its risk factors and association with depression.

International Classification of Diseases-10th Revision Surrogates of the Modified Frailty Index and 12-Month Referral to the Hospital in an Older Population from Germany.

Background: The International Classification of Diseases-10th revision (ICD-10) surrogates of the modified frailty index (mFI) have been defined in recent research. This study aimed to investigate the prevalence of these ICD-10 surrogates and their association with hospital referral in an older population from Germany. Methods: The present sample included adults aged ≥65 years followed in German general practices between 2010 and 2021. The index date was the most recent visit date. There were 11 ICD-10 surrogates of the mFI, including a total of 52 diagnoses. These surrogates were assessed in the 12 months prior to the index date. Referral to the hospital was analyzed in the 12 months following the index date. Covariates included age and sex. Results: There were 1,406,038 patients included in the study (mean (standard deviation) age 77.0 (7.9) years; 56.2% women). The prevalence of the ICD-10 surrogates of the mFI ranged from 3.3% for a "history of transient ischemic attack or stroke without neurological deficit" to 68.1% for a "history of hypertension requiring medication". In addition, 24%, 37%, and 23% of participants had 1, 2-3, and >3 ICD-10 surrogates of the mFI, respectively. There was a positive and significant relationship between the number of ICD-10 surrogates and 12-month incident hospital referral (reference: 0 surrogate; 1 surrogate: HR = 1.37, 95% CI = 1.31-1.42; 2-3 surrogates: HR = 2.00, 95% CI = 1.93-2.08; >3 surrogates: HR = 3.32, 95% CI = 3.19-3.44). Conclusions: ICD-10 surrogates of the mFI were relatively frequent and were significantly associated with 12-month incident hospital referral in this sample of older adults from general practices in Germany.

Impact of perceived discrimination and coping strategies on well-being and mental health in newly-arrived migrants in Spain.

OBJECTIVES: To explore how perceived discrimination impacts the emotional well-being and mental health of newly-arrived migrants in Spain; and to identify the coping strategies and behavioral changes used to deal with perceived discrimination. DESIGN: 102 individual audio-recorded in-depth qualitative interviews were conducted. The interviews were transcribed and analyzed through content analysis. RESULTS: Negative emotions related to perceived discrimination included disgust, sadness, fear, loneliness, humiliation, sense of injustice, rage, feeling undervalued or vulnerable, and mixed emotions. Change in behaviors due to perceived discrimination comprised westernization or cultural assimilation, creating a good image, avoiding going out or leaving alone, hypervigilance, stop participating in politics, self-sufficiency, a positive adaptation, and paradoxically, becoming an oppressor. The identified coping strategies to deal with perceived discrimination were ignoring or not responding, isolation, self-medication, engagement in intellectual activities, leisure and sport, talking or insulting the oppressor, denouncement, physical fight or revenge, seeking comfort, increasing solidarity with others, crying, or using humor. Discrimination-related stress and related mental health problems were conveyed, as challenges related to substance abuse and addictive behaviors, mood, and anxiety. CONCLUSIONS: Findings establish initial evidence of the great impact of perceived discrimination on the health, emotional well-being, and behavior of newly-arrived migrants in Spain, alerting to the need for targeted policies and services to address the effects of discrimination in this population. Further research is needed to explore more closely the causes and effects of perceived discrimination on mental health, to develop more targeted and effective interventions.

Effectiveness of a scalable, remotely delivered stepped-care intervention to reduce symptoms of psychological distress among Polish migrant workers in the Netherlands: study protocol for the RESPOND randomised controlled trial.

BACKGROUND: The COVID-19 pandemic has negatively affected the mental health of international migrant workers (IMWs). IMWs experience multiple barriers to accessing mental health care. Two scalable interventions developed by the World Health Organization (WHO) were adapted to address some of these barriers: Doing What Matters in times of stress (DWM), a guided self-help web application, and Problem Management Plus (PM +), a brief facilitator-led program to enhance coping skills. This study examines whether DWM and PM + remotely delivered as a stepped-care programme (DWM/PM +) is effective and cost-effective in reducing psychological distress, among Polish migrant workers with psychological distress living in the Netherlands. METHODS: The stepped-care DWM/PM + intervention will be tested in a two-arm, parallel-group, randomized controlled trial (RCT) among adult Polish migrant workers with self-reported psychological distress (Kessler Psychological Distress Scale; K10 > 15.9). Participants (n = 212) will be randomized into either the intervention group that receives DWM/PM + with psychological first aid (PFA) and care-as-usual (enhanced care-as-usual or eCAU), or into the control group that receives PFA and eCAU-only (1:1 allocation ratio). Baseline, 1-week post-DWM (week 7), 1-week post-PM + (week 13), and follow-up (week 21) self-reported assessments will be conducted. The primary outcome is psychological distress, assessed with the Patient Health Questionnaire Anxiety and Depression Scale (PHQ-ADS). Secondary outcomes are self-reported symptoms of depression, anxiety, posttraumatic stress disorder (PTSD), resilience, quality of life, and cost-effectiveness. In a process evaluation, stakeholders' views on barriers and facilitators to the implementation of DWM/PM + will be evaluated. DISCUSSION: To our knowledge, this is one of the first RCTs that combines two scalable, psychosocial WHO interventions into a stepped-care programme for migrant populations. If proven to be effective, this may bridge the mental health treatment gap IMWs experience. TRIAL REGISTRATION: Dutch trial register NL9630, 20/07/2021, https://www.onderzoekmetmensen.nl/en/trial/27052.

Is There an Association between Post-Traumatic Stress Disorder and the Incidence of Chronic Low Back Pain?

BACKGROUND: Preliminary research suggests post-traumatic stress disorder (PTSD) is a risk factor for chronic low back pain (CLBP). However, this literature displays some limitations. Therefore, this study aimed to investigate the association between PTSD and the 10-year cumulative incidence of CLBP in adults from Germany. METHODS: The present retrospective cohort study included adults diagnosed with PTSD in 1 of 1284 general practices in Germany in 2005-2020 (index date). Individuals without PTSD were matched to those with PTSD (1:1) using a propensity score based on age, sex, index year, duration of follow-up, and the mean number of consultations during follow-up. In patients without PTSD, the index date was a randomly selected visit date. RESULTS: There were 60,664 patients included in the study. After adjusting for frequent comorbidities, there was a positive but non-significant association between PTSD and incident CLBP in the overall population (HR = 1.07, 95% CI = 0.99-1.15). Nonetheless, the relationship between PTSD and CLBP was statistically significant in the age group >60 years (HR = 1.24, 95% CI = 1.05-1.46). CONCLUSIONS: Conversely to previous research, PTSD was not associated with incident CLBP in this large German sample. Further longitudinal studies are warranted to corroborate these findings before drawing any firm conclusions.

Traumatic stress symptoms among Spanish healthcare workers during the COVID-19 pandemic: a prospective study.

AIM: To investigate the occurrence of traumatic stress symptoms (TSS) among healthcare workers active during the COVID-19 pandemic and to obtain insight as to which pandemic-related stressful experiences are associated with onset and persistence of traumatic stress. METHODS: This is a multicenter prospective cohort study. Spanish healthcare workers (N = 4,809) participated at an initial assessment (i.e., just after the first wave of the Spain COVID-19 pandemic) and at a 4-month follow-up assessment using web-based surveys. Logistic regression investigated associations of 19 pandemic-related stressful experiences across four domains (infection-related, work-related, health-related and financial) with TSS prevalence, incidence and persistence, including simulations of population attributable risk proportions (PARP). RESULTS: Thirty-day TSS prevalence at T1 was 22.1%. Four-month incidence and persistence were 11.6% and 54.2%, respectively. Auxiliary nurses had highest rates of TSS prevalence (35.1%) and incidence (16.1%). All 19 pandemic-related stressful experiences under study were associated with TSS prevalence or incidence, especially experiences from the domains of health-related (PARP range 88.4-95.6%) and work-related stressful experiences (PARP range 76.8-86.5%). Nine stressful experiences were also associated with TSS persistence, of which having patient(s) in care who died from COVID-19 had the strongest association. This association remained significant after adjusting for co-occurring depression and anxiety. CONCLUSIONS: TSSs among Spanish healthcare workers active during the COVID-19 pandemic are common and associated with various pandemic-related stressful experiences. Future research should investigate if these stressful experiences represent truly traumatic experiences and carry risk for the development of post-traumatic stress disorder.

Multilingual markers of depression in remotely collected speech samples: A preliminary analysis.

BACKGROUND: Speech contains neuromuscular, physiological and cognitive components, and so is a potential biomarker of mental disorders. Previous studies indicate that speaking rate and pausing are associated with major depressive disorder (MDD). However, results are inconclusive as many studies are small and underpowered and do not include clinical samples. These studies have also been unilingual and use speech collected in controlled settings. If speech markers are to help understand the onset and progress of MDD, we need to uncover markers that are robust to language and establish the strength of associations in real-world data. METHODS: We collected speech data in 585 participants with a history of MDD in the United Kingdom, Spain, and Netherlands as part of the RADAR-MDD study. Participants recorded their speech via smartphones every two weeks for 18 months. Linear mixed models were used to estimate the strength of specific markers of depression from a set of 28 speech features. RESULTS: Increased depressive symptoms were associated with speech rate, articulation rate and intensity of speech elicited from a scripted task. These features had consistently stronger effect sizes than pauses. LIMITATIONS: Our findings are derived at the cohort level so may have limited impact on identifying intra-individual speech changes associated with changes in symptom severity. The analysis of features averaged over the entire recording may have underestimated the importance of some features. CONCLUSIONS: Participants with more severe depressive symptoms spoke more slowly and quietly. Our findings are from a real-world, multilingual, clinical dataset so represent a step-change in the usefulness of speech as a digital phenotype of MDD.

Association between carpal tunnel syndrome and the five-year incidence of anxiety disorder and depression in adults followed in general practices in Germany.

OBJECTIVE: Little is known about the longitudinal relationship between carpal tunnel syndrome (CTS) and mental health. Therefore, this retrospective cohort study investigated the association between CTS and the five-year incidence of anxiety disorder and depression in adults from Germany. METHODS: Data from the Disease Analyzer database (IQVIA) were used for the present study. Patients aged ≥18 years diagnosed for the first time with CTS in one of 1284 general practices in Germany in 2005-2020 were included (index date). Individuals without CTS were matched to those with CTS using a propensity score based on age, sex, the mean number of consultations per year during the follow-up, and the index year. In people without CTS, the index date was a randomly selected consultation in 2005-2020. RESULTS: There were 75,135 patients with and 75,135 patients without CTS included in the study (mean [SD] age 57.2 [16.5] years; 59.7% women). Within five years of the index date, the incidence of anxiety disorder was 3.9% and 3.6% in the group with and the group without CTS, respectively (log-rank p-value<0.001), while figures for depression were 14.8% and 11.5% (log-rank p-value<0.001). These findings were corroborated in the Cox regression analyses adjusted for multiple physical conditions, as CTS was associated with anxiety disorder (HR = 1.14, 95% CI = 1.08-1.21) and depression (HR = 1.29, 95% CI = 1.25-1.33) in the overall sample. CONCLUSION: CTS was associated with an increased incidence of anxiety disorder and depression in Germany. Further research should identify the mediators involved in these relationships.

Sex differences in the symptom network structure of depression: Findings from a nationwide sample of the Spanish adult population.

BACKGROUND: Sex differences in the prevalence and clinical features of depression have been widely described. However, some authors argue that categorical diagnostic systems do not adequately capture the complexity of depression. The aim of this study was to examine sex differences in the symptom network structure of depressive symptoms among individuals with a major depressive episode. METHODS: The study sample consisted of 510 participants (age 62.17 ± 14.43, 71.96 % women) from a nationwide study of the Spanish non-institutionalised adult population (Edad con Salud). To estimate the presence of a 12-month major depressive episode according to DSM-IV criteria, participants were administered an adapted version of the Composite International Diagnostic Interview (CIDI 3.0). A network analysis was carried out to determine possible interrelationships between different depressive symptoms by sex. RESULTS: Men and women showed a similar overall structure and network strength. However, sex-specific variations emerged in relation to individual symptom associations and symptom centrality. Specifically, for individual symptom associations "loss of confidence" and "suicide attempts" were more strongly related in women, and "suicidal ideation" and "impaired thinking" in men. For symptom centrality, "anxiety" played a central role in men's symptomatology, whereas "hopelessness", "loss of confidence", "distress" and "slowness of movement" were the most central symptoms in the women's group. LIMITATIONS: Reliance on cross-sectional data precludes us from determining the direction and temporality of the association between different symptoms. CONCLUSIONS: This study suggests that specific symptoms should be prioritised in the prevention, diagnosis assessment and treatment of depressed patients based on sex.

Challenges in Using mHealth Data From Smartphones and Wearable Devices to Predict Depression Symptom Severity: Retrospective Analysis.

BACKGROUND: Major depressive disorder (MDD) affects millions of people worldwide, but timely treatment is not often received owing in part to inaccurate subjective recall and variability in the symptom course. Objective and frequent MDD monitoring can improve subjective recall and help to guide treatment selection. Attempts have been made, with varying degrees of success, to explore the relationship between the measures of depression and passive digital phenotypes (features) extracted from smartphones and wearables devices to remotely and continuously monitor changes in symptomatology. However, a number of challenges exist for the analysis of these data. These include maintaining participant engagement over extended time periods and therefore understanding what constitutes an acceptable threshold of missing data; distinguishing between the cross-sectional and longitudinal relationships for different features to determine their utility in tracking within-individual longitudinal variation or screening individuals at high risk; and understanding the heterogeneity with which depression manifests itself in behavioral patterns quantified by the passive features. OBJECTIVE: We aimed to address these 3 challenges to inform future work in stratified analyses. METHODS: Using smartphone and wearable data collected from 479 participants with MDD, we extracted 21 features capturing mobility, sleep, and smartphone use. We investigated the impact of the number of days of available data on feature quality using the intraclass correlation coefficient and Bland-Altman analysis. We then examined the nature of the correlation between the 8-item Patient Health Questionnaire (PHQ-8) depression scale (measured every 14 days) and the features using the individual-mean correlation, repeated measures correlation, and linear mixed effects model. Furthermore, we stratified the participants based on their behavioral difference, quantified by the features, between periods of high (depression) and low (no depression) PHQ-8 scores using the Gaussian mixture model. RESULTS: We demonstrated that at least 8 (range 2-12) days were needed for reliable calculation of most of the features in the 14-day time window. We observed that features such as sleep onset time correlated better with PHQ-8 scores cross-sectionally than longitudinally, whereas features such as wakefulness after sleep onset correlated well with PHQ-8 longitudinally but worse cross-sectionally. Finally, we found that participants could be separated into 3 distinct clusters according to their behavioral difference between periods of depression and periods of no depression. CONCLUSIONS: This work contributes to our understanding of how these mobile health-derived features are associated with depression symptom severity to inform future work in stratified analyses.