Forecasting the onset of depression with limited baseline data only: A comparison of a person-specific and a multilevel modeling based exponentially weighted moving average approach.

The onset of depressive episodes is preceded by changes in mean levels of affective experiences, which can be detected using the exponentially weighted moving average procedure on experience sampling method (ESM) data. Applying the exponentially weighted moving average procedure requires sufficient baseline data from the person under study in healthy times, which is needed to calculate a control limit for monitoring incoming ESM data. It is, however, not trivial to obtain sufficient baseline data from a single person. We therefore investigate whether historical ESM data from healthy individuals can help establish an adequate control limit for the person under study via multilevel modeling. Specifically, we focus on the case in which there is very little baseline data available of the person under study (i.e., up to 7 days). This multilevel approach is compared with the traditional, person-specific approach, where estimates are obtained using the person's available baseline data. Predictive performance in terms of Matthews correlation coefficient did not differ much between the approaches; however, the multilevel approach was more sensitive at detecting mean changes. This implies that for low-cost and nonharmful interventions, the multilevel approach may prove particularly beneficial. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Depression recognition using voice-based pre-training model.

The early screening of depression is highly beneficial for patients to obtain better diagnosis and treatment. While the effectiveness of utilizing voice data for depression detection has been demonstrated, the issue of insufficient dataset size remains unresolved. Therefore, we propose an artificial intelligence method to effectively identify depression. The wav2vec 2.0 voice-based pre-training model was used as a feature extractor to automatically extract high-quality voice features from raw audio. Additionally, a small fine-tuning network was used as a classification model to output depression classification results. Subsequently, the proposed model was fine-tuned on the DAIC-WOZ dataset and achieved excellent classification results. Notably, the model demonstrated outstanding performance in binary classification, attaining an accuracy of 0.9649 and an RMSE of 0.1875 on the test set. Similarly, impressive results were obtained in multi-classification, with an accuracy of 0.9481 and an RMSE of 0.3810. The wav2vec 2.0 model was first used for depression recognition and showed strong generalization ability. The method is simple, practical, and applicable, which can assist doctors in the early screening of depression.

Mental health analysis of international students using machine learning techniques.

International students' mental health has become an increasing concern in recent years, as more students leave their country for better education. They experience a wide range of challenges while studying abroad that have an impact on their psychological well-being. These challenges can include language obstacles, cultural differences, homesickness, financial issues and other elements that could severely impact the mental health of international students. Given the limited research on the demographic, cultural, and psychosocial variables that influence international students' mental health, and the scarcity of studies on the use of machine learning algorithms in this area, this study aimed to analyse data to understand the demographic, cultural factors, and psychosocial factors that impact mental health of international students. Additionally, this paper aimed to build a machine learning-based model for predicting depression among international students in the United Kingdom. This study utilized both primary data gathered through an online survey questionnaire targeted at international students and secondary data was sourced from the 'A Dataset of Students' Mental Health and Help-Seeking Behaviors in a Multicultural Environment,' focusing exclusively on international student data within this dataset. We conducted data analysis on the primary data and constructed models using the secondary data for predicting depression among international students. The secondary dataset is divided into training (70%) and testing (30%) sets for analysis, employing four machine learning models: Logistic Regression, Decision Tree, Random Forest, and K Nearest Neighbor. To assess each algorithm's performance, we considered metrics such as Accuracy, Sensitivity, Specificity, Precision and AU-ROC curve. This study identifies significant demographic variables (e.g., loan status, gender, age, marital status) and psychosocial factors (financial difficulties, academic stress, homesickness, loneliness) contributing to international students' mental health. Among the machine learning models, the Random Forest model demonstrated the highest accuracy, achieving an 80% accuracy rate in predicting depression.

External validation of three diabetes prediction scores in a Spanish cohort: does adding high risk for depression improve the validation of the FINDRISC score (FINDRISC-MOOD)?

OBJECTIVES: To evaluate the external validity of the FINDRISC, DESIR and ADA risk scores for the prediction of diabetes in a Spanish population aged >45 years and to test the possible improvement of FINDRISC by adding a new variable of high risk of depression when Patient Health Questionnaire-9 (PHQ-9) questionnaire score ≥10 (FINDRISC-MOOD). DESIGN: Prospective population-based cohort study. SETTING: 10 primary healthcare centres in the north of the city of Madrid (Spain). PARTICIPANTS: A total of 1242 participants without a history of diabetes and with 2-hour oral glucose tolerance test (OGTT) plasma glucose <200 mg/dL (<11.1 mmol/L) were followed up for 7.3 years (median) using their electronic health records (EHRs) and telephone contact. PRIMARY AND SECONDARY OUTCOME MEASURES: Diabetes risk scores (FINDRISC, DESIR, ADA), PHQ-9 questionnaire and 2-hour-OGTT were measured at baseline. Incident diabetes was defined as treatment for diabetes, fasting plasma glucose ≥126 mg/dL (≥7.0 mmol/L), new EHR diagnosis or self-reported diagnosis. External validation was performed according to optimal cut-off, sensitivity, specificity and Youden Index. Comparison between diabetes risk scores, including FINDRISC-MOOD (original FINDRISC score plus five points if PHQ-9 ≥10), was measured by area under the receiver operating characteristic curve (AUROC). RESULTS: During follow-up, 104 (8.4%; 95% CI, 6.8 to 9.9) participants developed diabetes and 185 had a PHQ-9 score ≥10. The AUROC values were 0.70 (95% CI, 0.67 to 0.72) for FINDRISC-MOOD and 0.68 (95% CI, 0.65 to 0.71) for the original FINDRISC. The AUROCs for DESIR and ADA were 0.66 (95% CI, 0.63 to 0.68) and 0.66 (95% CI, 0.63 to 0.69), respectively. There were no significant differences in AUROC between FINDRISC-MOOD and the other scores. CONCLUSIONS: The results of FINDRISC-MOOD were like those of the other risk scores and do not allow it to be recommended for clinical use.

Neuroserpin, IL-33 and IL-17A as potential markers of mild symptoms of depressive syndrome in Toxoplasma gondii-infected pregnant women.

Introduction: Depressive syndrome (DS) is a common complication during pregnancy and the postpartum period, and is triggered by multiple organic/genetic and environmental factors. Clinical and biochemical follow-up is essential for the early diagnosis and prognosis of DS. The protozoan Toxoplasma gondii causes infectious damage to the fetus during parasite primary-infection. However, in long-term infections, pregnant women develop immune protection to protect the fetus, although they remain susceptible to pathological or inflammatory effects induced by T. gondii. This study aimed to investigate plasma inflammatory biomarkers in pregnant women seropositive and seronegative for T. gondii, with diagnoses of minor and moderate/severe DS. Methods: Pregnant women (n=45; age=18-39 years) were recruited during prenatal care at health centers in Ouro Preto, Minas Gerais, Brazil. Participants were asked to complete a socio-demographic questionnaire to be submitted to well-standardized DS scale calculators (Beck Depression Inventory Questionnaire, Edinburgh Postnatal Depression Scale, and Major Depressive Episode Module). Additionally, 4 mL of blood was collected for plasma neuroserpin, CCL2, IL-17A, and IL-33 analysis. Results: Pregnant volunteers with chronic T. gondii contact were all IgG+ (44%; n=21) and exhibited increased plasma IL-33, IL-17A, and neuroserpin levels, but not CCL2, compared to uninfected pregnant women. Using Beck's depression inventory, we observed an increase in plasma IL-17A and IL-33 in women with T. gondii infeCction diagnosed with mild DS, whereas neuroserpin was associated with minor and moderate/severe DS. Discussion: Our data suggest a close relationship between DS in pregnant women with chronic T. gondii infection and neurological conditions, which may be partially mediated by plasma neuroserpin, IL-33, and IL-17A levels.

The SCREENIVF Hungarian version is a valid and reliable measure accurately predicting possible depression in female infertility patients.

Infertility patients, often in high distress, are entitled to being informed about their mental status compared to normative data. The objective of this study was to revalidate and test the accuracy of the SCREENIVF, a self-reported tool for screening psychological maladjustment in the assisted reproduction context. A cross-sectional, questionnaire-based online survey was carried out between December 2019 and February 2023 in a consecutive sample of female patients (N = 645, response rate 22.9%) in a university-based assisted reproduction center in Hungary. Confirmatory factor analysis and cluster and ROC analyses were applied to test validity, sensitivity and specificity in relation to Beck Depression Inventory (BDI) scores. Model fit was optimal (chi-square = 630.866, p < 0.001; comparative fit index = 0.99; root-mean-square error of approximation = 0.018 (90% CI 0.013-0.023); standardized-root-mean-square-residual = 0.044), and all dimensions were reliable (α > 0.80). A specific combination of cutoffs correctly predicted 87.4% of BDI-scores possibly indicative of moderate-to-severe depression (χ2(1) = 220.608, p < 0.001, Nagelkerke R2 = 0.462, J = 66.4). The Hungarian version of the SCREENIVF is a valid and reliable tool, with high accuracy in predicting BDI-scores. Low response rate may affect generalizability. The same instrument with different cutoffs can serve various clinical goals.

Factors that indicate performance on the MoCA 7.3 in healthy adults over 50 years old.

Human aging is a physiological, progressive, heterogeneous global process that causes a decline of all body systems, functions, and organs. Throughout this process, cognitive function suffers an incremental decline with broad interindividual variability.The first objective of this study was to examine the differences in the performance on the MoCA test (v. 7.3) per gender and the relationship between the performance and the variables age, years of schooling, and depressive symptoms .The second objective was to identify factors that may influence the global performance on the MoCA test (v. 7.3) and of the domains orientation, language, memory, attention/calculation, visuospatial and executive function, abstraction, and identification.A cross-sectional study was carried out in which five hundred seventy-three (573) cognitively healthy adults ≥ 50 years old were included in the study. A sociodemographic questionnaire, the GDS-15 questionnaire to assess depression symptoms and the Spanish version of the MoCA Test (v 7.3) were administered. The evaluations were carried out between the months of January and June 2022. Differences in the MoCA test performance per gender was assessed with Student's t-test for independent samples. The bivariate Pearson correlation was applied to examine the relationship between total scoring of the MoCA test performance and the variables age, years of schooling, and depressive symptoms. Different linear multiple regression analyses were performed to determine variables that could influence the MoCA test performance.We found gender-related MoCA Test performance differences. An association between age, years of schooling, and severity of depressive symptoms was observed. Age, years of schooling, and severity of depressive symptoms influence the MoCA Test performance, while gender does not.

Feature group partitioning: an approach for depression severity prediction with class balancing using machine learning algorithms.

In contemporary society, depression has emerged as a prominent mental disorder that exhibits exponential growth and exerts a substantial influence on premature mortality. Although numerous research applied machine learning methods to forecast signs of depression. Nevertheless, only a limited number of research have taken into account the severity level as a multiclass variable. Besides, maintaining the equality of data distribution among all the classes rarely happens in practical communities. So, the inevitable class imbalance for multiple variables is considered a substantial challenge in this domain. Furthermore, this research emphasizes the significance of addressing class imbalance issues in the context of multiple classes. We introduced a new approach Feature group partitioning (FGP) in the data preprocessing phase which effectively reduces the dimensionality of features to a minimum. This study utilized synthetic oversampling techniques, specifically Synthetic Minority Over-sampling Technique (SMOTE) and Adaptive Synthetic (ADASYN), for class balancing. The dataset used in this research was collected from university students by administering the Burn Depression Checklist (BDC). For methodological modifications, we implemented heterogeneous ensemble learning stacking, homogeneous ensemble bagging, and five distinct supervised machine learning algorithms. The issue of overfitting was mitigated by evaluating the accuracy of the training, validation, and testing datasets. To justify the effectiveness of the prediction models, balanced accuracy, sensitivity, specificity, precision, and f1-score indices are used. Overall, comprehensive analysis demonstrates the discrimination between the Conventional Depression Screening (CDS) and FGP approach. In summary, the results show that the stacking classifier for FGP with SMOTE approach yields the highest balanced accuracy, with a rate of 92.81%. The empirical evidence has demonstrated that the FGP approach, when combined with the SMOTE, able to produce better performance in predicting the severity of depression. Most importantly the optimization of the training time of the FGP approach for all of the classifiers is a significant achievement of this research.

Decision tree-Markov model of perinatal depression screening: a cost-utility analysis.

Background: Perinatal depression affects the physical and mental health of pregnant women. It also has a negative effect on children, families, and society, and the incidence is high. We constructed a cost-utility analysis model for perinatal depression screening in China and evaluated the model from the perspective of health economics. Methods: We constructed a Markov model that was consistent with the screening strategy for perinatal depression in China, and two screening strategies (screening and non-screening) were constructed. Each strategy was set as a cycle of 3 months, corresponding to the first trimester, second trimester, third trimester, and postpartum. The state outcome parameters required for the model were obtained based on data from the National Prospective Cohort Study on the Mental Health of Chinese Pregnant Women from August 2015 to October 2016. The cost parameters were obtained from a field investigation on costs and screening effects conducted in maternal and child health care institutions in 2020. The cost-utility ratio and incremental cost-utility ratio of different screening strategies were obtained by multiplicative analysis to evaluate the health economic value of the two screening strategies. Finally, deterministic and probabilistic sensitivity analyses were conducted on the uncertain parameters in the model to explore the sensitivity factors that affected the selection of screening strategies. Results: The cost-utility analysis showed that the per capita cost of the screening strategy was 129.54 yuan, 0.85 quality-adjusted life years (QALYs) could be obtained, and the average cost per QALY gained was 152.17 yuan. In the non-screening (routine health care) group, the average cost was 171.80 CNY per person, 0.84 QALYs could be obtained, and the average cost per QALY gained was 205.05 CNY. Using one gross domestic product per capita in 2021 as the willingness to pay threshold, the incremental cost-utility ratio of screening versus no screening (routine health care) was about -3,126.77 yuan, which was lower than one gross domestic product per capita. Therefore, the screening strategy was more cost-effective than no screening (routine health care). Sensitivity analysis was performed by adjusting the parameters in the model, and the results were stable and consistent, which did not affect the choice of the optimal strategy. Conclusion: Compared with no screening (routine health care), the recommended perinatal depression screening strategy in China is cost-effective. In the future, it is necessary to continue to standardize screening and explore different screening modalities and tools suitable for specific regions.

Reliability and validity of the short version of the childhood abuse self report scale in Chinese college students.

BACKGROUND: The reliability and validity of the current scale for measuring childhood abuse in China are worrying. The development of the Short Version of the Childhood Abuse Self Report Scale (CASRS-12) helps to change this situation, but the effectiveness of the tool has not yet been tested in Chinese participants. This study aims to test the reliability and validity of the CASRS­12 in Chinese college students. METHODS: A total of 932 college students were investigated, of whom 418 were investigated for the first time, and only the CASRS­12 was filled out. In the second survey, 514 participants filled out the CASRS­12, Depression Scale, Self-esteem Scale and Subjective Well-being Scale in turn. After 4 weeks, 109 participants were selected for retest. RESULTS: Each item of the CASRS­12 had good discrimination. Exploratory factor analysis and confirmatory factor analysis (χ2/df = 4. 18, RMSEA = 0. 079, CFI = 0. 95, TLI = 0. 94, IFI = 0. 95, NFI = 0. 94) all supported the four-factor structure of the scale, and the cumulative contribution rate of variance was 76.05%. Cronbach's α coefficient and retest reliability were 0.86 and 0.65, respectively. Childhood abuse was positively correlated with depression (r = 0. 42, p < 0.01), and negatively correlated with self-esteem (r=-0. 33, p < 0.01) and subjective well-being (r=-0. 32, p < 0.01). CONCLUSION: The Chinese version of CASRS­12 meets the measurement standard and could be used to measure the level of childhood abuse of Chinese college students.

The association of depressive symptoms with handgrip strength and gait speed in community-dwelling older adults: data from the baseline phase of Birjand Longitudinal Aging Study.

BACKGROUND: Depression is a multifaceted condition with a high prevalence and burden to society. Handgrip strength (HGS) and gait speed (GS) are indices of physical health, which is linked to mental health. Previous studies have shown heterogeneity among countries in the association of physical parameters and depression. In this study, we aimed to investigate the association of HGS and GS with depressive symptoms in older adults. METHODS: This is a cross-sectional study analyzing data from the Birjand Longitudinal Aging Study, a cohort of community-dwelling older adults (≥ 60 years old). Depressive symptoms were assessed by the nine-item Patient Health Questionnaire. HGS was measured with a hand dynamometer in a sitting position, and GS was estimated by a 15-foot walk test at usual pace. RESULTS: Compared to participants in the first quartile, those in the second quartile of HGS had significantly lower odds of suffering from depressive symptoms, while GS was not significantly associated with depressive symptoms. A higher HGS was associated with a lower risk of moderate depressive symptoms, while a higher GS was related to a lower risk of moderately severe and severe symptoms. CONCLUSIONS: Our findings suggest that older people residing in Birjand, Iran with a moderate HGS are less likely to suffer from depressive symptoms than those with lower HGS.

A qualitative study on the adaptation of community programmes for the promotion of early detection and health-seeking of perinatal depression in Nepal.

BACKGROUND: Despite the high burden of perinatal depression in Nepal, the detection rate is low. Community-based strategies such as sensitization programmes and the Community Informant Detection Tool (CIDT) have been found to be effective in raising awareness and thus promoting the identification of mental health problems. This study aims to adapt these community strategies for perinatal depression in the Nepalese context. METHODS: We followed a four-step process to adapt the existing community sensitization program manual and CIDT. Step 1 included in-depth interviews with women identified with perinatal depression (n=36), and focus group discussions were conducted with health workers trained in community mental health (n=13), female community health volunteers (FCHVs), cadre of Nepal government for the prevention and promotion of community maternal and child health (n=16), and psychosocial counsellors (n=5). We explored idioms and understanding of depression, perceived causes, and possible intervention. Step 2 included draft preparation based on the qualitative study. Step 3 included a one-day workshop with the psychosocial counsellors (n=2) and health workers (n=12) to assess the understandability and comprehensiveness of the draft and to refine the content. A review of the CIDT and community sensitization program manual by a psychiatrist was performed in Step 4. RESULTS: The first step led to the content development for the CIDT and community sensitization manual. Multiple stakeholders and experts reviewed and refined the content from the second to fourth steps. Idioms of depression and commonly cited risk factors were incorporated in the CIDT. Additionally, myths of perinatal depression and the importance of the role of family were added to the community sensitization manual. CONCLUSION: Both the CIDT and community sensitization manual are grounded in the local context and are simple, clear, and easy to understand.

The association of symptomatic vitreous floaters with depression and anxiety.

PURPOSE: To evaluate the levels of anxiety and depression in patients with symptomatic vitreous floaters and to determine the possible correlations of psychological implications with the symptoms duration and possible improvement, the degree of posterior vitreous detachment, and the discomfort severity. METHODS: Ninety patients complaining for floaters and fifty-seven age- and gender-matched healthy-control subjects were recruited. Every participant underwent a complete ophthalmological examination, including funduscopy and optical coherence tomography scans, while clinical and demographic data were also gathered. The Patient Health Questionnaire-9 (PHQ-9), the Zung Depression Inventory-Self-Rating Depression Scale (Zung SDS), and the Hospital Anxiety and Depression Scale (HADS) were completed by everyone. RESULTS: Between the studied groups, no significant differences were detected regarding the clinical and demographic data (p > 0.05). The patients with floaters had significantly higher scores of PHQ-9, Zung SDS, HADS Anxiety, and HADS Depression (p < 0.001). After adjustment for several confounders, PHQ-9 (p = 0.041), Zung SDS (p = 0.003), and HADS Anxiety (p = 0.036) values remained significantly impaired. Among the patients, PHQ-9 and Zung SDS scores were significantly elevated in the patients with floaters duration less than 4 weeks (p < 0.05). Finally, anxiety and depression were significantly correlated with the symptoms duration and intensity, with the floater-associated discomfort, and with the stage of posterior vitreous detachment. CONCLUSION: Vitreous floaters have a negative impact on patients' psychological status, by the terms of enhanced depressive and anxiety levels. To the best of our knowledge, our study is the first in the literature to elaborate the aforementioned association, by assessing three different questionnaires simultaneously.

Prevalence of depression and associated symptoms among patients attending primary healthcare facilities: a cross-sectional study in Nepal.

BACKGROUND: Depression is a prevalent mental health condition worldwide but there is limited data on its presentation and associated symptoms in primary care settings in low- and middle-income countries like Nepal. This study aims to assess the prevalence of depression, its hallmark and other associated symptoms that meet the Diagnostic and Statistical Manual (DSM-5) criteria in primary healthcare facilities in Nepal. The collected information will be used to determine the content of a mobile app-based clinical guidelines for better detection and management of depression in primary care. METHODS: A total of 1,897 adult patients aged 18-91 (63.1% women) attending ten primary healthcare facilities in Jhapa, a district in eastern Nepal, were recruited for the study between August 2, 2021, and March 25, 2022. Trained research assistants conducted face-to-face interviews in private spaces before the consultation with healthcare providers. Depression symptoms, including hallmark symptoms, was assessed using the validated Nepali version of the Patient Health Questionnaire (PHQ-9). RESULTS: One in seven (14.5%) individuals attending primary health care facilities in Jhapa met the threshold for depression based on a validated cut-off score ( > = 10) on the PHQ-9. The most commonly reported depressive symptoms were loss of energy and sleep difficulties. Approximately 25.4% of women and 18.9% of men endorsed at least one of the two hallmark symptoms on the PHQ-9. Using a DSM-5 algorithm (at least one hallmark symptom and five or more total symptoms) to score the PHQ-9, 6.3% of women and 4.3% of men met the criteria for depression. The intra-class correlation coefficient for PHQ-9 total scores by health facility as the unit of clustering was 0.01 (95% confidence interval, 0.00-0.04). CONCLUSION: Depression symptoms are common among people attending primary healthcare facilities in Nepal. However, the most common symptoms are not the two hallmark criteria. Use of total scores on a screening tool such as the PHQ-9 risks overestimating the prevalence and generating false positive diagnoses. Compared to using cut off scores on screening tools, training health workers to first screen for hallmark criteria may increase the accuracy of identification and lead to better allocation of treatment resources.

Exploring prenatal depression and postpartum depression: Findings from a prospective cohort study at University Hospital Obstetrics in Damascus.

The perinatal period is crucial for both mother and newborn, and mental health, including prenatal and postpartum depression (PPD), is a significant aspect. Screening for these disorders allows for early treatment and helps prevent risks to both mother and child. This prospective cohort study was carried out at University Hospital Obstetrics in Damascus City. The first phase was during the third trimester of pregnancy and the second phase involved a follow-up assessment after 6 weeks of delivery. The Arabic-validated version of the Edinburgh Postnatal Depression Scale questionnaire (EPDS) was used. A cutoff of 13 or higher was used to determine the presence of probable depression in both assessments. Of 347 pregnant women, 38.6% had prenatal depression (PND). 295 patients have achieved the second assessment, of which 30.2% had PPD. Furthermore, 42.6% who had PND developed PPD on follow-up. Binary logistic regression indicated that PND was predicted by non-Syrian nationality, paternal absence, poor financial status, number of previous pregnancies, and a history of depression independent of pregnancy. PPD was predicted by a history of PPD, and work status. Findings underscore potential value of early screening for depressive symptoms as a predictive measure. It is recommended that women with a history of depression receive heightened attention and care, irrespective of the timing of their depressive episodes.

A novel machine learning-based prediction method for patients at risk of developing depressive symptoms using a small data.

The prediction of depression is a crucial area of research which makes it one of the top priorities in mental health research as it enables early intervention and can lead to higher success rates in treatment. Self-reported feelings by patients represent a valuable biomarker for predicting depression as they can be expressed in a lower-dimensional network form, offering an advantage in visualizing the interactive characteristics of depression-related feelings. Furthermore, the network form of data expresses high-dimensional data in a compact form, making the data easy to use as input for the machine learning processes. In this study, we applied the graph convolutional network (GCN) algorithm, an effective machine learning tool for handling network data, to predict depression-prone patients using the network form of self-reported log data as the input. We took a data augmentation step to expand the initially small dataset and fed the resulting data into the GCN algorithm, which achieved a high level of accuracy from 86-97% and an F1 (harmonic mean of precision and recall) score of 0.83-0.94 through three experimental cases. In these cases, the ratio of depressive cases varied, and high accuracy and F1 scores were observed in all three cases. This study not only demonstrates the potential for predicting depression-prone patients using self-reported logs as a biomarker in advance, but also shows promise in handling small data sets in the prediction, which is critical given the challenge of obtaining large datasets for biomarker research. The combination of self-reported logs and the GCN algorithm is a promising approach for predicting depression and warrants further investigation.

Digital Phenotyping for Stress, Anxiety, and Mild Depression: Systematic Literature Review.

BACKGROUND: Unaddressed early-stage mental health issues, including stress, anxiety, and mild depression, can become a burden for individuals in the long term. Digital phenotyping involves capturing continuous behavioral data via digital smartphone devices to monitor human behavior and can potentially identify milder symptoms before they become serious. OBJECTIVE: This systematic literature review aimed to answer the following questions: (1) what is the evidence of the effectiveness of digital phenotyping using smartphones in identifying behavioral patterns related to stress, anxiety, and mild depression? and (2) in particular, which smartphone sensors are found to be effective, and what are the associated challenges? METHODS: We used the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) process to identify 36 papers (reporting on 40 studies) to assess the key smartphone sensors related to stress, anxiety, and mild depression. We excluded studies conducted with nonadult participants (eg, teenagers and children) and clinical populations, as well as personality measurement and phobia studies. As we focused on the effectiveness of digital phenotyping using smartphones, results related to wearable devices were excluded. RESULTS: We categorized the studies into 3 major groups based on the recruited participants: studies with students enrolled in universities, studies with adults who were unaffiliated to any particular organization, and studies with employees employed in an organization. The study length varied from 10 days to 3 years. A range of passive sensors were used in the studies, including GPS, Bluetooth, accelerometer, microphone, illuminance, gyroscope, and Wi-Fi. These were used to assess locations visited; mobility; speech patterns; phone use, such as screen checking; time spent in bed; physical activity; sleep; and aspects of social interactions, such as the number of interactions and response time. Of the 40 included studies, 31 (78%) used machine learning models for prediction; most others (n=8, 20%) used descriptive statistics. Students and adults who experienced stress, anxiety, or depression visited fewer locations, were more sedentary, had irregular sleep, and accrued increased phone use. In contrast to students and adults, less mobility was seen as positive for employees because less mobility in workplaces was associated with higher performance. Overall, travel, physical activity, sleep, social interaction, and phone use were related to stress, anxiety, and mild depression. CONCLUSIONS: This study focused on understanding whether smartphone sensors can be effectively used to detect behavioral patterns associated with stress, anxiety, and mild depression in nonclinical participants. The reviewed studies provided evidence that smartphone sensors are effective in identifying behavioral patterns associated with stress, anxiety, and mild depression.

Profiles of centenarians' functioning: linking functional and cognitive capacity with depressive symptoms.

BACKGROUND: Despite most centenarians facing age-related declines in functional and cognitive capacities, the severity of these declines varies among individuals, as does the maintenance of good mental health (e.g., depressive symptoms) despite these declines. This study aims to examine this heterogeneity in centenarians from the Second Heidelberg Centenarian Study, which collected data from 112 centenarians living in Germany. In our study, we focus on a subsample of 73 centenarians who provided self-reports for our measures of interest (M age = 100.4, SD age = 0.55). METHODS: We examined correlations between functional capacity (i.e., PADL, IADL), cognitive capacity (i.e., MMSE), and depressive symptoms (i.e., GDS), and the existence of different profiles using hierarchical clustering. RESULTS: Higher functional capacity was related to higher cognitive capacity and to fewer depressive symptoms. Yet, higher cognitive capacity was associated with more depressive symptoms. Hierarchical clustering analysis elucidated this contradiction by identifying three profiles: low-capacity individuals (i.e., 24 individuals had low functional and cognitive capacities, with low depressive symptoms), high-capacity individuals (i.e., 33 individuals with high functional and cognitive capacities, with low depressive symptoms), and low-functional-high-cognitive-capacity individuals (i.e., 16 individuals showed low functional but high cognitive capacity, with high depressive symptoms). Our post-hoc analyses highlighted arthritis and pain as risk factors for functional dependence and depression. CONCLUSIONS: Our findings emphasize the importance of identifying centenarian subgroups with specific resource- and risk profiles to better address their needs, and of treating pain to improve functional capacity and mental health in centenarians.