Cohort profile: rationale and design of the Resource Center for Health Science (RECHS) project – a study of health hazards and medical cost burden among the Japanese population


  • This prospective cohort study investigates the associations between lifestyle and environmental factors and some surrogate markers of non-communicable diseases.

  • We initiated a cohort study among participants of health check-ups and conducted a baseline self-administered questionnaire that included items on dietary habits and stress.

  • We used a Brief-type Self-Administered Diet History Questionnaire, the Center for Epidemiologic Studies Depression Scale and the General Health Questionnaire-28 as self-administered questionnaires.

  • This project designed a community-based, prospective cohort study combined with personal health records from annual health check-ups and an integrated biobank.

  • Some limitations to the study protocol include the low response rates to the self-administered questionnaires and a selection bias resulting from losses to follow-up.


In Japan, the world’s fastest and most rapidly ageing society,1 2 medical costs are also rising dramatically because of advances in medical technology and unmet medical needs; therefore, how societies deal with health hazards and medical cost burden has become an urgent issue.3–5 All developed countries will eventually face this problem, so it must be addressed.

Epidemiological cohort studies with long-term follow-up have played an important role in identifying factors that contribute to primary prevention and early disease detection. In recent years, in developed countries, the increased incidence of not only non-communicable diseases caused by diet and lifestyle,6 but also psychiatric diseases such as depression, which leads to deteriorated health, a lower quality of life (QOL) and possibly even a loss of life for the entire population, has become a major social problem.7 8 The causes of obesity include lower physical activity levels, the increased intake of high-calorie foods, metabolic abnormalities and social factors.9 On the other hand, depression has been considered one of the main factors contributing to obesity.10 11 Depression is caused by both biological and psychological/social factors12 and can also be triggered by environmental factors such as life events and excessive stress. Depression has been on the rise in recent years, and according to the WHO, will be the second most prevalent disease by 2020.13

The causes of non-communicable diseases and mental disorders involve a complex interplay between predisposing risk factors and various lifestyle, environmental and health-related exposures.14–16 Moreover, a significant inter-relationship exists between non-communicable diseases and mental disorders. Therefore, to gain novel insights into the causes and consequences of non-communicable diseases and mental disorders, large prospective studies that comprehensively collect diverse exposures prior to the onset of diseases are essential. To address this need, the Resource Center for Health Science (RECHS) project was established as a prospective cohort study that combines a large sample size with a wide range of data on exposures and outcomes. This initiative aims to enhance the prevention, diagnosis and treatment of middle and old age-related ailments, encompassing non-communicable diseases and mental disorders.

The present cohort study is based on an integrated database (DB) of self-administered questionnaires, anonymised clinical physiological data and biobanked biological samples obtained from health examinations. It is expected to result in a unique cohort study using a DB that enables the analysis of dietary habits, mental status and omics data. A Brief-type Self-Administered Diet History Questionnaire (BDHQ),17 18 the Centre for Epidemiologic Studies Depression Scale (CES-D)19 20 and the General Health Questionnaire-28 (GHQ-28),21 22 all of which have been widely used and clinically validated in many countries around the world, will be used as self-administered questionnaires to assess and follow changes over time in dietary habits and stress levels.

The RECHS has initiated a project based on the development and utilisation of the Bio-Resources/DB (BR/DB), which contains personal health records (PHRs), such as health/medical records physically consolidated with BR (eg, serum and urine samples) taken from the same individuals. The purpose of this study is to establish a primary defence for improving QOL by applying the BR/DB to the analysis from the perspectives of epidemiology and clinical chemistry. Furthermore, this study is expected to contribute to the construction of a PHR system. In terms of regional medical services, the RECHS coordinating activities are fully dependent on as many general hospitals as possible, as well as groups in academia capable of analysing the BR/DB.

The purpose of the RECHS project is to provide a long-term repository and enable the utilisation of BR integrated with PHRs. This project is intended to contribute to primary prevention measures for improving QOL at the individual level.

Cohort description

Data source and study participants

In 2010, researchers affiliated with the RECHS designed a community-based, prospective cohort study that was combined with PHRs from annual health checkups and an integrated biobank that stores BR in Kasamatsu town, Hashima District, Gifu Prefecture, which is located in central Japan near metropolitan Nagoya (figure 1). As of 2010, Kasamatsu had an area of 10.3 km2 with a population of approximately 22 000, and the proportion of older people aged ≥65 years in Kasamatsu was about 23.5% (5400 persons); by 2015, based on census data in Japan, the proportion of the older population had increased to about 27% (6100 persons).

Figure 1
Figure 1

The location of the cohort study. The location of Gifu Prefecture (red area) in Japan is shown on the left. An enlarged map showing the location of Kasamatsu town is shown on the right.

The health-check cohort consisted of all patients who had received a health check at the Matsunami Health Promotion Clinic from February 2010. Recruitment teams inform potential participants of the study objectives and associated benefits of their participation. The research nature of the study is emphasised, including the information it is designed to provide and the impact the study results may have on policy-making and healthcare for future generations. Written informed consent to participate in the cohort was repeatedly confirmed at each annual health check-up. The exclusion criteria for the RECHS are as follows: (1) persons who do not consent to participate in the study and (2) persons who are not able to complete the study questionnaires. As of April 2020, the enrolment of cohort members and follow-up health checks are ongoing. The target cumulative cohort population for the survey is about 10 000. In the RECHS cohort study, we are planning to have a sample size of 10 000 participants. The sample size was calculated by PS: Power and Sample Size Calculation (Department of Biostatistics and Vanderbilt University School of Medicine, USA), which is a programme used to calculate the required sample size and power under the following assumptions: 80% power, a type I error rate of 0.05, an accrual interval of 156 months, an additional follow-up period after the accrual interval of 60 months and a median survival time on the control treatment of 48 months. Based on the entire study cohort of 10 000, an HR of 0.941 or 1.064 would be able to detect an event occurring during follow-up and equally sized low-risk and high-risk groups. This size would also be sufficient to screen for the marginal effects of more strongly related factors or factors with a higher number of events.

The RECHS project consists of one basic and four detailed surveys, namely, a comprehensive health check-up and the BDHQ, CES-D and GHQ-28.

Baseline survey

The baseline questionnaire items included the area of residence, birth year, sex, health condition at that time, smoking and alcohol habits, past and family history of various diseases, health check-up/cancer screening history, frequency of food intake and physical exercise, and reproductive history (for female participants). Question items also included the number of holidays, means of transportation to daily activities, school or work (eg, walking, driving, public transportation such as trains or buses), information about socioeconomic status, sleep duration and household workload. Anthropometric measurements were conducted by trained medical staff following standardised criteria, including blood pressure, height, weight and body mass index. Fasting blood samples obtained via antecubital venipuncture underwent comprehensive laboratory analysis encompassing a complete blood count, evaluating red blood cells, haematocrit, haemoglobin, platelet count and white cell count, in addition to assessments for aspartate aminotransferase, alanine aminotransferase, γ-glutamyl transpeptidase, total bilirubin, alkaline phosphatase, total protein, albumin, lactate dehydrogenase, cholinesterase, total cholesterol, triglyceride, high-density lipoprotein-cholesterol, low-density lipoprotein-cholesterol, haemoglobin A1c and fasting blood glucose. Serum creatinine, blood urea nitrogen and uric acid levels were also included in the blood tests. Urine samples were assessed for protein, sugar levels and occult blood. Chest X-ray, ultrasound sonography and electrocardiography were also assessed. Concurrently, several mini-tubes (100–500 µL per tube) of serum from each examinee were stored at –80°C in freezers at Matsunami Research Park for future analyses.


Brief-type Self-Administered Diet History Questionnaire

Dietary intake was assessed using a BDHQ previously validated for the assessment of the Japanese diet.17 18 The BDHQ is a four-page, structured, self-administered questionnaire. The BDHQ asks the participants about the consumption and intake frequency of 46 listed food and non-alcoholic beverage items and the daily intake of rice and miso soup. The frequency of drinking alcoholic beverages and amount per drink for five alcoholic beverages, typical cooking methods and general dietary behaviour during the preceding month are queried prior to the questionnaire survey. The validity of the BDHQ was assessed by comparing 4-day and 16-day diet records; the details of the validation study have been described elsewhere.17 18

The CES-D and the GHQ-28

Depressive status and some questions regarding sleep habits were also assessed among the cohort participants. Depressive status was measured using the CES-D and GHQ-28. The CES-D, which was developed for epidemiological studies by the National Institute of Mental Health,19 20 is a self-assessment scale for depression that is widely employed around the world in community-based and population-based surveys, and has been shown to be an adequate measure of depression with sufficient reliability and validity.23 The CES-D is composed of 16 depressive symptom items and four positive affect items. The degree of depression is measured by a score from 0 to 60. The cut-off point is set at 16 points in the original version, and a reliability and validity study conducted in Japan concluded that 19 points was also an appropriate cut-off.23

The GHQ-28, developed by Goldberg,21 22 is a 28-item self-administered questionnaire that has been adopted and validated as a screening test for psychological well-being by the WHO. The reliability of the GHQ-28 has been confirmed.23 The GHQ-28 score ranges from 0 to 84, with lower scores indicating lower levels of distress. The GHQ-28 score does not have an absolute cut-off value. The GHQ-28 is closely correlated with the GHQ-60.22 24 Data were obtained from the questionnaire surveys conducted at baseline and the time of the annual health check-ups.


Since 2010, we have been conducting annual health check-ups. The study participants who were still able to participate in these health check-ups and their study outcomes were then obtained from the relevant measurements and questionnaires. To follow the cohort participants and obtain information regarding morbidity and mortality, we will adopt the following three strategies: (1) repeated questionnaires at the site of the health checkups by mail or telephone, (2) official death records and (3) confirmation through a review of medical charts by attending and family physicians.


Locally and globally, the results of this study will be communicated through article submissions in peer-reviewed journals and conference presentations at national and international venues.

Patient and public involvement

No patient is involved.

Data validation

Trained staff have been handling data entry, checking and cleaning regularly. Inconsistent records have been manually checked and corrected. The quality and completeness of the data entry have been checked regularly. Data checks have also been performed by research staff who differ from the data entry staff. Data have been collected electronically, including on laptops and desktops provided to the staff. Electronic data entry devices have been accessible by authorised staff only. A central DB has been developed to ensure secure data management.

Participant characteristics

As of 31 March 2015, RECHS includes 5137 participants. The preliminary demographic profile of the participants at baseline, 5 years after the start of enrolment, is shown in table 1. The mean age of the participants was 49.6±9.3 years, and 61.0% were men. Based on the initial recruiting centres, we experienced about 2800 new cases yearly during the first 3 years of inclusion. Considering an improving recruitment rate, we estimate that the number of cases that can be recruited in the RECHS is around 10 000.

Table 1

Patient demographics at baseline

Findings to date

In Japan, where the birth rate is rapidly decreasing and the population rapidly ageing, medical costs are rising dramatically because of advances in medical technology and unmet medical needs; thus, how to deal with the rising medical cost burden has become an urgent issue.3–5 Implementing therapeutic and/or preventive measures at an appropriate time before symptom onset or the occurrence of serious tissue damage is necessary to prevent or delay the onset of disease.

To realise primary preventive medicine (measures against predisease) at the individual level and contribute to national healthcare, including the improvement of QOL, the purpose of this project is to store and manage a long-term continuous personalised DB to promote useful clinical research.

The key to maintaining good health is attending regular health check-ups and making improvements in daily lifestyle. To achieve a long and healthy life or lifelong vitality, it is also important to maintain a personal health history record. Current clinical tests are powerful tools for diagnosing diseases but are not used for primary prevention, such as maintaining physical and mental health. Laboratory tests that target primary prevention (some current lipid-related items are effective for primary prevention25) are therefore needed.

It is important to conduct comprehensive analyses of blood components that can track the process of transition from a healthy to a diseased state at the individual level. The construction of a properly managed and chronological DB containing information about the results of health checkups and medical examinations from the stored BR of healthy people and patients with specific diseases would, therefore, be helpful. It is important to understand how lifestyle-related diseases can affect health. Among lifestyle habits, the influence of diet is significant, and thus, investigating the causal relationship between these factors is also important.

In anticipation of this trend, the search for and development of biomarkers useful for early primary prevention for the purpose of health promotion at the individual level are being vigorously pursued. However, to validate the composition of such biomarkers, it is essential to collect BR and PHRs that can be used for prospective cohort studies. To solve the aforementioned problems, clinical laboratory tests, dietary habit surveys (eg, the BDHQ) and stress surveys have been introduced, and the construction of a BR/DB is underway.

Potential research methodologies include descriptive studies on nutritional indicators, depression scales, explorations of factors associated with the onset of non-communicable diseases and mental disorders, investigations into novel biomarkers using cryopreserved samples, and the development of future disease prediction models based on presurveyed data. The utilisation of continuously measured data for examining causal inferences such as interrupted time series and regression discontinuity designs is feasible. Moreover, leveraging foundational data on diagnostic performance, such as cachexia, overweight or depression, can also be anticipated.

Through an analysis of the cohort’s health records encompassing both lifestyle-related diseases and mental health, this study indicates potential areas for improvement. The utilisation of electronic and standardised PHRs provides valuable insights and presents opportunities for targeted interventions. This approach demonstrates the potential to identify and address key factors vital for enhancing health outcomes across lifestyle-related and mental health domains.


All RECHS data reported in this cohort profile are held and managed confidentially by the authors’ research group at Matsunami Research Park in Japan. The data are currently accessible only to the study team. Researchers interested in collaborating may contact Tetsuji Morishita, Vice Director of Internal Medicine at Matsunami General Hospital ([email protected]).

Data availability statement

Data are available on reasonable request. The datasets generated and/or analysed during the present study are not publicly available due to ethical/privacy reasons but are available from the corresponding author on reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and the study protocol was approved by the Ethics Committees of Matsunami General Hospital (No. 558, 2010 and No. 487, 2021) and Kyoto University (No. E761, 2009). Participants gave informed consent to participate in the study before taking part.


We are greatly indebted to the staffs of Matsunami General Hospital and Matsunami Health Promotion Clinic for their generous cooperation in establishing and following the cohort. We also gratefully acknowledge the special efforts of Matsunami Research Park. The authors thank FORTE Science Communications ( for English language editing.

This post was originally published on