An observational study in Anxiety Disorders, Depressive Disorder and Bipolar Disorder (BD), sponsored by Second Affiliated Hospital, Zhejiang University, School of Medicine. Recruiting at 1 site in China. Open to participants aged 12 Years to 55 Years, including healthy volunteers. Per ClinicalTrials.gov, last updated 2026-10-29.
Sponsored by Second Affiliated Hospital, Zhejiang University, School of Medicine · Observational
This study aims to understand how anxiety disorders, depressive disorders, and bipolar disorder change over time and to identify patterns that may help with earlier recognition, diagnosis, and prediction of future symptoms.
The study will enroll approximately 2,690 participants aged 12 to 55 years across multiple hospitals in China. Participants will include people with anxiety disorders, depressive disorders, or bipolar disorder, as well as healthy volunteers for comparison.
Participants will be followed for 3 years after enrollment, with assessments at baseline and at 1, 6, 12, 18, 24, 30, and 36 months. Study assessments will collect information about symptoms, treatment, daily functioning, and quality of life. Other study data will include biological samples, brain imaging, recordings of brain electrical activity, and measurements from wearable devices and tasks involving speech, facial expressions, and eye movements.
Researchers will examine how these measurements relate to symptoms and changes in health over time. They will also develop and evaluate artificial intelligence models intended to help identify these disorders and predict treatment outcomes, relapse, and suicide or self-harm risk.
This is an observational study. The researchers will not assign treatments or change participants' treatment as part of the study. Treatments received during routine clinical care will be recorded.
This multicenter prospective cohort study will establish a standardized research database integrating clinical assessments, biological samples, neuroimaging, electroencephalography, and digital measurements in participants with anxiety disorders, depressive disorders, bipolar disorder, and healthy controls.
Enrollment will occur on a rolling basis, and follow-up visits will be scheduled relative to each participant's baseline assessment. Each participant will be followed for 36 months unless they withdraw or otherwise discontinue participation.
The study comprises two stages. The model development cohort will include 1,590 participants: 380 with anxiety disorders, 380 with depressive disorders, 380 with bipolar disorder, and 450 healthy controls. A separate independent validation cohort will include 1,100 participants, with 275 in each of the four groups. This second cohort will be used to evaluate model discrimination, sensitivity, specificity, calibration, and generalizability across centers.
The research database will integrate clinical and psychosocial information with biological samples for multi-omics analyses, structural and functional magnetic resonance imaging, diffusion tensor imaging, 64-channel electroencephalography, and digital measurements from wearable devices, speech, facial expressions, and eye movements. Standardized data collection procedures and longitudinal tracking will support comparisons across groups, centers, and assessment time points.
Analyses will investigate associations between these measurements and clinical characteristics, symptom trajectories, treatment outcomes, relapse, and suicide or self-harm risk. Longitudinal analyses will also explore how biological changes interact with environmental and psychosocial factors. Statistical approaches will include multivariable analyses, mixed-effects models, survival analyses, structural equation modeling, and machine learning, as appropriate to the research question.
The study will develop and independently validate multimodal artificial intelligence models for early recognition and diagnostic support, and explore models for predicting treatment outcomes, relapse, and suicide or self-harm risk. Model development and evaluation are research activities within this observational cohort; no treatment is assigned by the study protocol.
5,113 studies on the registry are indexed under Anxiety Disorders; 1,418 are open to participants now.
This study's planned enrollment of 2,690 is above the median of 150 across 636 observational studies indexed under Anxiety Disorders.
Browse Anxiety Disorders studies →Second Affiliated Hospital, Zhejiang University, School of Medicine is the lead sponsor of 97 studies on the registry; 65 are open to participants now.
Counted across the registry records on this site, refreshed daily.
Participants aged 12-55 years will be recruited across participating hospitals in Zhejiang Province, China, through outpatient referrals and public recruitment announcements, including WeChat public-account posts. The study population comprises individuals with DSM-5 anxiety disorders, depressive disorders, or bipolar I or II disorder, and healthy volunteers meeting the control-group eligibility criteria. Enrollment will occur on a rolling basis. The planned sample includes 1,590 participants in the model development cohort and 1,100 participants in a separate independent validation cohort. Each participant will be followed for 36 months after baseline.
Go through the criteria one by one — yes, no or not sure — and take the list to your doctor or the study team. Kept in this browser only.
General criteria for all groups:
Anxiety disorders group:
Depressive disorders group:
Bipolar disorder group:
Healthy control group:
Exclusion Criteria:
Participants aged 12-55 years who meet DSM-5 criteria for one or more anxiety disorders and have a Hamilton Anxiety Rating Scale (HAMA-14) total score of at least 14. Planned enrollment is 655 participants: 380 in the model development cohort and 275 in the independent validation cohort. Participants will undergo baseline assessments and follow-up at 1, 6, 12, 18, 24, 30, and 36 months. No treatment is assigned by the study; routine clinical care and treatment received will be recorded.
Participants aged 12-55 years who meet DSM-5 criteria for a depressive disorder and have a 17-item Hamilton Depression Rating Scale (HAMD-17) total score of at least 17. This group includes participants with and without a history of suicide attempts or self-harm. Planned enrollment is 655 participants: 380 in the model development cohort and 275 in the independent validation cohort. Participants will undergo baseline assessments and follow-up at 1, 6, 12, 18, 24, 30, and 36 months. No treatment is assigned by the study; routine clinical care and treatment received will be recorded.
Participants aged 12-55 years who meet DSM-5 criteria for bipolar I or bipolar II disorder and are not experiencing a manic episode at enrollment. Planned enrollment is 655 participants: 380 in the model development cohort and 275 in the independent validation cohort. Participants will undergo baseline assessments and follow-up at 1, 6, 12, 18, 24, 30, and 36 months. No treatment is assigned by the study; routine clinical care and treatment received will be recorded.
Healthy participants aged 12-55 years who meet the protocol-defined eligibility criteria for the control group, including normal cognitive functioning, no serious physical or chronic disease, and no family history of mental illness across three generations on either the maternal or paternal side. Planned enrollment is 725 participants: 450 in the model development cohort and 275 in the independent validation cohort. Participants will undergo baseline assessments and follow-up at 1, 6, 12, 18, 24, 30, and 36 months. No treatment is assigned by the study.
Macro-Averaged Area Under the ROC Curve of the Multimodal Digital Phenotyping-Based Diagnostic Model in the Independent Validation Cohort
The model will use multimodal digital phenotyping data to identify three diagnostic labels: depressive disorders, anxiety disorders, and bipolar disorder. Reference diagnoses will be established by psychiatrists using MINI interviews and clinical diagnostic assessment. Co-occurring diagnoses will be represented according to diagnostic criteria, allowing anxiety disorders to coexist with depressive or bipolar disorders. An area under the receiver operating characteristic curve (AUC) will be calculated separately for each label. The primary measure will be the unweighted mean of the three label-specific AUCs. Healthy controls will be negative for all three labels. AUC ranges from 0 to 1, with higher values indicating better discrimination. A 95% confidence interval will be calculated using bootstrap resampling. Feature selection, tuning, and training will be completed in the development cohort, and the model will be locked before independent validation.
Time frame: Baseline (at enrollment in the independent validation cohort).
Change From Baseline in 17-Item Hamilton Depression Rating Scale (HAMD-17) Score
Depressive symptom severity will be assessed using the 17-item Hamilton Depression Rating Scale (HAMD-17). Total scores range from 0 to 52, with higher scores indicating more severe depressive symptoms. The outcome is the change in total score at each follow-up assessment relative to baseline, calculated as the follow-up score minus the baseline score. Results will be analyzed separately for the depressive disorders group and the bipolar disorder group. Negative changes indicate improvement.
Time frame: Baseline and 1, 6, 12, 18, 24, 30, and 36 months after baseline.
Change From Baseline in 14-Item Hamilton Anxiety Rating Scale (HAMA-14) Score
Anxiety symptom severity will be assessed using the 14-item Hamilton Anxiety Rating Scale (HAMA-14). Each item is scored from 0 to 4, yielding a total score from 0 to 56. Higher scores indicate more severe anxiety symptoms. The outcome is the change in total score at each follow-up assessment relative to baseline, calculated as the follow-up score minus the baseline score. Negative changes indicate improvement.
Time frame: Baseline and 1, 6, 12, 18, 24, 30, and 36 months after baseline.
Change From Baseline in Bech-Rafaelsen Mania Rating Scale (BRMS) Score in the Bipolar Disorder Group
Manic symptom severity in participants with bipolar disorder will be assessed using the 11-item Bech-Rafaelsen Mania Rating Scale (BRMS). Each item is scored from 0 to 4, yielding a total score from 0 to 44. Higher scores indicate more severe manic symptoms. The outcome is the change in total score at each follow-up assessment relative to baseline, calculated as the follow-up score minus the baseline score. Negative changes indicate improvement.
Time frame: Baseline and 1, 6, 12, 18, 24, 30, and 36 months after baseline.
Sensitivity of the Diagnostic Model in the Independent Validation Cohort
Sensitivity will be calculated separately for the depressive disorder, anxiety disorder, and bipolar disorder labels, with reference diagnoses established by psychiatrists using MINI interviews and clinical diagnostic assessment. Each label will be evaluated as a separate binary outcome, allowing co-occurring diagnoses according to diagnostic criteria. Sensitivity is the number of true positives divided by the sum of true positives and false negatives. Values range from 0 to 1, with higher values indicating better sensitivity. Each estimate will be reported with a 95% confidence interval. The model and label-specific classification thresholds will be fixed using the development cohort before independent validation and will not be adjusted using validation data.
Time frame: Baseline (at enrollment in the independent validation cohort).
Specificity of the Diagnostic Model in the Independent Validation Cohort
Specificity will be calculated separately for the depressive disorder, anxiety disorder, and bipolar disorder labels, with reference diagnoses established by psychiatrists using MINI interviews and clinical diagnostic assessment. Each label will be evaluated as a separate binary outcome, allowing co-occurring diagnoses according to diagnostic criteria. Specificity is the number of true negatives divided by the sum of true negatives and false positives. Values range from 0 to 1, with higher values indicating better specificity. Each estimate will be reported with a 95% confidence interval. The model and label-specific classification thresholds will be fixed using the development cohort before independent validation and will not be adjusted using validation data.
Time frame: Baseline (at enrollment in the independent validation cohort).
Label-Specific Accuracy of the Diagnostic Model in the Independent Validation Cohort
Accuracy will be calculated separately for the depressive disorder, anxiety disorder, and bipolar disorder labels, with reference diagnoses established by psychiatrists using MINI interviews and clinical diagnostic assessment. Each label will be evaluated as a separate binary outcome, allowing co-occurring diagnoses according to diagnostic criteria. Label-specific accuracy is the sum of true positives and true negatives divided by the number of participants evaluated for that label. Values range from 0 to 1, with higher values indicating better accuracy. Each estimate will be reported with a 95% confidence interval. The model and label-specific classification thresholds will be fixed using the development cohort before independent validation and will not be adjusted using validation data.
Time frame: Baseline (at enrollment in the independent validation cohort).
Positive Predictive Value of the Diagnostic Model in the Independent Validation Cohort
Positive predictive value will be calculated separately for the depressive disorder, anxiety disorder, and bipolar disorder labels, with reference diagnoses established by psychiatrists using MINI interviews and clinical diagnostic assessment. Each label will be evaluated as a separate binary outcome, allowing co-occurring diagnoses according to diagnostic criteria. Positive predictive value is the number of true positives divided by the sum of true positives and false positives. Values range from 0 to 1, with higher values indicating better positive predictive value. Each estimate will be reported with a 95% confidence interval. The model and label-specific classification thresholds will be fixed using the development cohort before independent validation and will not be adjusted using validation data.
Time frame: Baseline (at enrollment in the independent validation cohort).
Negative Predictive Value of the Diagnostic Model in the Independent Validation Cohort
Negative predictive value will be calculated separately for the depressive disorder, anxiety disorder, and bipolar disorder labels, with reference diagnoses established by psychiatrists using MINI interviews and clinical diagnostic assessment. Each label will be evaluated as a separate binary outcome, allowing co-occurring diagnoses according to diagnostic criteria. Negative predictive value is the number of true negatives divided by the sum of true negatives and false negatives. Values range from 0 to 1, with higher values indicating better negative predictive value. Each estimate will be reported with a 95% confidence interval. The model and label-specific classification thresholds will be fixed using the development cohort before independent validation and will not be adjusted using validation data.
Time frame: Baseline (at enrollment in the independent validation cohort).
Change From Baseline in Patient Health Questionnaire-9 (PHQ-9) Score
Depressive symptom severity will be assessed using the 9-item Patient Health Questionnaire (PHQ-9). Total scores range from 0 to 27, with higher scores indicating more severe depressive symptoms. The outcome is the change in total score at each follow-up assessment relative to baseline, calculated as the follow-up score minus the baseline score. Negative changes indicate improvement.
Time frame: Baseline and 1, 6, 12, 18, 24, 30, and 36 months after baseline.
Change From Baseline in Generalized Anxiety Disorder-7 (GAD-7) Score
Anxiety symptom severity will be assessed using the 7-item Generalized Anxiety Disorder scale (GAD-7). Total scores range from 0 to 21, with higher scores indicating more severe anxiety symptoms. The outcome is the change in total score at each follow-up assessment relative to baseline, calculated as the follow-up score minus the baseline score. Negative changes indicate improvement.
Time frame: Baseline and 1, 6, 12, 18, 24, 30, and 36 months after baseline.
Mean Absolute Error of Baseline HAMD-17 Score Estimation in the Independent Validation Cohort
The model will use baseline multimodal digital phenotyping data to estimate the concurrently assessed 17-item Hamilton Depression Rating Scale (HAMD-17) total score. Performance will be evaluated in the independent validation cohort using clinician-rated HAMD-17 scores as the reference. Mean absolute error (MAE) will be calculated as the mean absolute difference between model-estimated and observed HAMD-17 total scores, expressed in HAMD-17 score points. Lower values indicate better estimation accuracy. Model development, feature selection, and parameter tuning will be completed in the development cohort, and the model will be locked before independent validation.
Time frame: Baseline (at enrollment in the independent validation cohort).
Mean Absolute Error of Future HAMD-17 Score Prediction Using Baseline Digital Phenotyping Data
Exploratory models will use baseline multimodal digital phenotyping data to predict HAMD-17 total scores at subsequent follow-up visits. Predictive performance will be evaluated separately at each follow-up time point using mean absolute error (MAE), calculated as the mean absolute difference between predicted and observed HAMD-17 scores. MAE will be expressed in HAMD-17 score points, with lower values indicating better prediction accuracy. Feature selection, parameter tuning, and model training will be performed within the development cohort. Models will be fixed before evaluation in the independent validation cohort. Only information available at baseline will be used as predictor inputs.
Time frame: 1, 6, 12, 18, 24, 30, and 36 months after baseline.
Plasma Metabolite Abundance Profiles
Exploratory plasma metabolomic profiling will be performed using untargeted liquid chromatography-mass spectrometry (LC-MS). The outcome will be the normalized abundance of individual metabolite features after analytical quality control and data preprocessing. Metabolite abundances will be reported separately rather than combined into a composite score. Exploratory analyses will examine associations between metabolite abundances and concurrently assessed HAMD-17 and HAMA-14 total scores.
Time frame: Baseline.
Relative Abundance Profiles of Gut Microbial Species
Exploratory gut microbiome profiling will be performed. The outcome will be the relative abundance of individual microbial species after sequencing quality control and taxonomic profiling, expressed as the proportion of the microbial community assigned to each species. Species-level abundances will be reported separately rather than combined into a composite score. Exploratory analyses will examine associations between microbial species relative abundances and concurrently assessed HAMD-17 and HAMA-14 total scores.
Time frame: Baseline
Other open studies at The Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou
Open a list, then choose Watch this search to get new studies there in a weekly email.
Plan to share: Undecided
No publications or documents are linked to this record.
From the registry record's own update history. This site started tracking changes on Sep 25, 2026; for anything earlier, see the record history on ClinicalTrials.gov ↗
Eligibility is decided by the study team. Share this record with your doctor or contact the team directly.
Contact study teamGet an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.
Second Affiliated Hospital, Zhejiang University, School of Medicine