The main purpose of the MDCRC study is to investigate demographic information, medical and psychiatric history, clinical characteristics, course, and prognosis, as well as assess the risk factors or preventive factors associated with relapse or recovery or diagnostic conversion from unipolar depressive disorder to bipolar disorder targeting early-onset mood disorders. Additionally, the MDCRC study aims to facilitate translational research for mood disorders and to provide a resource for the study of mood disorders, such as bridging translational research and ICT convergence research on mood disorders by utilizing unique resources. Furthermore, the additional purpose of the MDCRC is to present a new methodology of mood disorders research. By complementing the shortcomings of the various mood disorders studies that have been conducted so far, our team aims to develop and perform specific research methods that reflect the clinical aspects of mood disorders.

The MDCRC study recruits mood disorder patients who fulfill the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) for major depressive disorder, bipolar I disorder, or bipolar II disorder.¹⁸ Generally, mood disorder patients do not receive treatment from the beginning of major mood symptoms for several reasons.^19202122 In fact, mood disorder patients initially visit psychiatric clinics and are diagnosed after a considerable period of time from the initial experience of major mood symptoms.^19202122 Early mood disorder is defined as follows: 1) under 25 years old with mood disorder; or 2) under 35 years old with mood disorder within 2 years of treatment. We exclude patients with intellectual disability or organic brain injury, and/or difficulty reading and understanding the Korean language.

The assessment schedule consists of the baseline assessment, inter-visit assessment, and follow-up assessment. The overall configuration of the MDCRC study is depicted in Table 1. The follow-up assessment is divided into formal assessment and telephone-based assessment according to the evaluation methods and into scheduled visit assessment and unscheduled visit assessment depending on the time of evaluation. Telephone-based assessment could be used only for limited cases when patients cannot visit the hospital for acceptable reasons, and it is contrived to reduce drop-out from the study. Scheduled visits for follow-up assessments are performed every 12 weeks from the baseline assessment. For inter-visit assessments, all the subjects are dictated to check a daily mood chart using the eMood chart application for smartphones or personal computers. If it is not applicable, the eMood chart is replaced by a paper mood diary. The daily mood chart is an important component of the inter-visit assessment. Patients are provided a daily mood chart that asks for mood state (−3 to +3), energy level (−3 to +3), anxiety (0-3), irritability (0-3), drinking (yes or no), prescription medication use (yes or no), total sleep time (hours), weight (once every two weeks), and menstruation (only for women). A trained psychiatrist reviews the records of the daily mood charts at every visit at an outpatient clinic or at the follow-up assessments. By reviewing daily mood charts and interviewing patients, the psychiatrist validates the daily mood chart and sometimes can modify the mood chart according to the judgment of the clinician. This process is called “inter-visit mood chart validation.” After the inter-visit mood chart validation process, a psychiatrist determines whether there have been possibilities of major or minor mood episodes during the inter-visit period. When a major or minor mood episode is suspected from the daily mood chart validation, investigators proceed to the next step called “inter-visit mood episode evaluation.” At the inter-visit mood episode evaluation step, clinicians ask patients about mood episodes according to DSM-5 criteria and establish a most recent diagnosis. If a major or minor mood episode is not suspected from the inter-visit mood chart validation, investigators skip the mood episode evaluation step and finish the follow-up assessment. Overall design of the MDCRC study is presented in Figure 1.

The MDCRC study started in September 2014 and will be sustained for at least 5 years. Since mood disorders have chronic and various clinical courses, many diagnosed cases of unipolar depression are known to change to bipolar disorder. It is very important to follow-up for long-term periods in the course of mood disorders, and we want to follow-up as long as possible.

Various assessments are performed at each visit starting from the baseline assessment. At first, screening evaluations for inclusion and exclusion criteria for the MDCRC Study are performed. After successful screening, at every visit thereafter, clinicians have to check and write down patients' most recent diagnosis according to the DSM-5.¹⁸ Assessment is composed of demographic and clinical data, psychiatric assessment by investigators and patients, and biological assessment, which is the same as mentioned below (Table 2).

The demographic data at baseline assessment consists of age, gender, height, weight, waist circumstance, handedness, and socioeconomic status, such as the average monthly income, education level, marital status with children, and kinds of jobs.

The sleep pattern was assessed by asking the sleep onset time, wake-up time, sleep latency time, total sleep time, and nap time. Almost all questions asked are divided by weekdays and weekends.

Medication history is precious information to understand patients with mood disorders. However, it is very difficult to investigate prescribed past medication history for treatment of mood disorders. We decided to investigate only whether the patients have been prescribed lithium for treatment instead of conducting a full investigation of medication history. If the patients have experiences of lithium medication, the clinician checks the ALDA scale for evaluating the responsiveness of lithium treatment.²³²⁴ All psychiatric medication is recorded from the baseline assessment and at every follow-up assessment by checking the patients' medication compliance, which is the percentage of real medicated drugs among prescribed drugs.

We investigate whether the patients have history of suicide attempts. If there has been history of suicide attempts, then patients are asked the age of the first attempted suicidal event and the number of suicidal attempts with or without mood episodes. Suicidal history is also assessed at every follow-up visit for investigation of inter-visit suicidal events.

Familial psychiatric history is assessed by asking for psychiatric treatment history, such as bipolar disorder, depressive disorder, alcohol and substance use disorders, anxiety disorders, schizophrenic spectrum disorders, and other psychiatric problems, including suicide.

The medical and surgical history is asked for assessment of past and current comorbidities. Moreover, average monthly cost of treatment is assessed for psychiatric, medical, dental, and oriental medicinal managements as well as medical usage patterns at every assessment. Smoking and alcohol history including duration and quantity of the uses is also checked for all patients at every assessment. Evaluation of menopause state is performed only for female participants.

The Korean version of the Mini-International Neuropsychiatric Interview (MINI) was used for psychiatric diagnosis by clinicians who are expert in psychiatry at the baseline assessment.²⁵ Patients with mood disorders often suffer from comorbid psychiatric symptoms. Psychiatric comorbidity was assessed by the same clinicians at every visit from the baseline assessment to catch newly developed psychiatric problems.

Routine mood symptom scales including the Montgomery-Åsberg Depression Rating Scale (MADRS) for depressive symptoms²⁶ and the Young Mania Rating Scale (YMRS) for manic symptoms²⁷ are applied. The Biological Rhythms Interview of Assessment in Neuropsychiatry (BRIAN) is checked to evaluate circadian rhythm tendencies in participants.²⁸ The Clinical Global Impression-Bipolar (CGI-BP) Scale is also investigated by clinicians at every assessment.²⁹

There are various self-report scales for evaluating patients' psychiatric state and history. We selected the self-report forms and subjective scales that are considered to be important for the study of patients with mood disorders. The Quick Inventory of Depressive Symptomatology,³⁰ Hypomania Symptom Checklist-32,³¹ and Mood Disorder Questionnaire³² are provided to patients for assessing mood symptoms such as depression, anxiety, hypomania, or bipolarity phenotypes. The Barratt Impulsiveness Scale³³ is applied for evaluating other important states closely related to mood problems. The Korean Version of Drug Attitude Inventory-10 is checked once for evaluating patient's compliance with biological treatment at baseline assessment.³⁴ For evaluating patients' seasonality and chronotype, the Seasonal Pattern Assessment Questionnaire³⁵ and Composite Scale of Morningness³⁶ are provided at every visit. The Korean version of WHO Quality of Life Scale abbreviated version is asked for subjective perception of quality of life in patients with mood disorders.³⁷ General physical activity is another important aspect of mood disorders, so we apply The International Physical Activity Questionnaire-Short Form.³⁸ Since various events occur in routine life and perceived stress from the events could affect the prognosis of mood disorders, the Life Experiences Survey is checked by each patient at every assessment.³⁹

Blood sample collection from all participants is performed once at the baseline enrollment of the study. Blood samples (10 mL) are collected into ethylenediaminetetraacetic acid (EDTA) bottles and stored in a −30℃ freezer. They will be used for genomic analyses or other biological analyses.

Clinical data are registered electronically via an electronic case report form based on the Internet-based Clinical Research and Trial Management System (iCReaT) established by the Centers for Disease Control and Prevention, Ministry of Health and Welfare, Republic of Korea (iCReaT, Study No. C150010). The data will be encrypted, and it will have limited access that is managed by iCReaT data managers.

There could be various objects for analysis from the MDCRC Study. We will use various statistical analysis methods depending on the characteristics of data and each hypothesis. For the analysis of demographic, clinical features, prognosis (including predictors and medical usage between different clinical groups of mood disorders), we will use paired t-test, logistic regression analysis, repeated measures ANOVA, Cox proportional hazard regression, and so on. For the analysis of longitudinal data of clinical symptom changes, mixed linear models taking into account the correlation between the different assessment variables of mood disorder patients in the course of the cohort study assessment schedule will be used.