The DPC (Diagnosis Procedure Combination) database is a national inpatient database in Japan, which includes approximately 7 million inpatients per year from more than 1000 hospitals. DPC data include administrative claims data and some clinical data. Recently, studies using the DPC database are increasing. In this report, we explain three of the DPC studies about (i) outcomes of argatroban treatment in patients with atherothrombotic stroke, (ii) comparison of short-term mortality and morbidity between parenteral and enteral nutrition for adults without cancer, and (iii) validation of the prognostic burn index. Also, several ongoing and future issues on large healthcare databases including the DPC database are discussed.

View full abstract

The characteristic and limitation as an example of medical big data about National Database (NDB), archives of the e-claim data were discussed. The NDB accumulates all e-claim data (medical, DPC, dentistry, and pharmacy) and specific medical examination data from 2009 by the Ministry of Health, Labor and Welfare. NDB is data accumulated for rationalization of medical expenses, but the door is left open to the local governments and researchers.
There is the limitation peculiar to NDB in addition to limitation of the e-claim data. However, an electronic rate of the claim data approaching to 98% today, it may be said that NDB can almost grasp the medical care situation of our country. An example of the usage of NDB, the data book which we offered for the local health care planning of the metropolis and districts, a community medicine design is demonstrated.
The big expectation for the NDB is to become the superior source of information of the cohort study for all nations. Because it is poking each other of the receipt information based on insurance information, combination stops it when insurance is changed, but combination of the e-claim data to cross by the introduction such as medical individual numbers for a long term is enabled and becomes able to grasp a series of episodes from the onset of the disease to convergence in future.

View full abstract

As health related big data, there are DPC and National database（NDB）in Japan. Annually the former gathers over 11 million discharge data and the latter accumulates more than 1.7 billion claim data only for medical services. The recent development of information technology has made it possible to process these big data easier and the government tries to advance its use for health policy making. Compared with similar data of other countries, the DPC and NDB are very superior for its exhaustiveness and detail. The advance in use of these data is expected to contribute the further development of health service research in Japan. To do so, we have to train more human resources for health related big data analyses.

View full abstract

Recently, several databases in health care field have been developing in Japan. Among them, medical claims (receipts) have various advantages and many epidemiological studies have been conducted using such data. The Japan Medical Data Center (JMDC), under the contract with health insurance societies, constructs and provides a database which comprises data of medical claims and special health check-up examinations. This article introduces some epidemiological research based on this database. Research on evidence-practice gap were conducted for the following three subjects : adherence with clinical practice guidelines which recommend the prophylactic use of anti-osteoporosis medications for long-term users of corticosteroid, compliance with the safety measures indicated in the package insert of ergot-derived anti-Parkinson drugs, and implementation of cardiac rehabilitation for patients with ischemic heart disease. These analyses showed that only one fourth of potentially eligible patients were treated according to recommendations or indications. A linkage analysis of health check-up data and claims can be used to identify how many people who are screened for medical examination really see a physician after the check-up. The research revealed proportion of people who failed to see a physician after the check-up, was thirds for possible diabetes and was approximately 90 percent for possible hypertensives, respectively. The national database (NDB) of health check-up data and claims of all Japanese people is now developing with a great expectations. However, nongovernmental database including the JMDC's one have some advantages in terms of sufficient size and usability for epidemiological research.

View full abstract

Large clinical database is able to be captured maximum value when it collaborates clinical professionals' quality improvement initiatives. Professional autonomy and evidence based practices are essential factor for quality improvement initiative using large clinical database. High value databases are also useful for health policy research which intends to sustainable and high quality health care system, and industry-academia partnership which intends to develop innovative diagnosis/treatment technology.

View full abstract

Huge amount of clinical data is generated every day through provision of healthcare services. Such data, recorded in hospital information systems, exported in standardized format and accumulated under various initiatives, are rapidly becoming available to research community in Japan. These data are not only used in clinical epidemiologic studies aimed at progress in medical science and technology, but in socioeconomic studies directed towards analysis of healthcare costs, efficiency and sustainability. In one of large-scale medical databases, the DPC data, inclusion of patient address zip code and availability of hospital performance statistics published as an open data by the Ministry of Health Labor and Welfare, raised wide interest in analysis of regional healthcare markets. This paper gives basic introduction to geospatial analysis of DPC data using geographic information system (GIS), showing sample analytic results and issues faced in the use of GIS.

View full abstract

Utilization of electronic health record (EHR) database has recently increased for regulatory purpose such as post-marketing drug safety measures. Pharmaceuticals and Medical Devices Agency (PMDA) of Japan has initiated a new project called as MIHARI Project (Medical Information for Risk Assessment Initiative) to implement pharmacoepidemiological approach on drug safety evaluation since FY 2009. In the MIHARI project, many pilot studies have been conducted to establish the framework of pharmacoepidemiological approach for regulatory purpose and to characterize Japanese EHR databases. In FY2011, a new initiative, Medical Information Database Network (MID-NET), was started to establish large-scale hospital information system database in Japan. PMDA has accumulated regulatory experiences on pharmacoepidemiological safety assessment of drugs through these initiatives, resulting in increasing quality of drug safety measures. These would also promote standardization and utilization of EHR database in Japan.
This article describes background, past and future activities of MIHARI and MID-NET projects in PMDA.

View full abstract

In Japan, large scale health databases were constructed in a few years, such as National Claim insurance and health checkup database (NDB). But there are some legal issues for making adequate balance between privacy and public benefit by using such databases. NDB is carried based on the act for elderly person's health care but in this act, nothing is mentioned for using this database for general public benefit. Therefore, researchers who use this database are forced to pay much concern about anonymization and information security that may disturb the research work itself. Generally speaking, researchers of study for public benefit will not infringe patient's privacy, but vague and complex requirements of legislation about personal data protection may disturb the researches. Medical science does not progress without using clinical information, therefore the adequate legislation that is simple and clear for both researchers and patient is strongly required. In Japan, new personal data protection act was enacted in Sep. 2015 and will be enrolled in 2017. Cabinet order about this act and guidelines for specific fields should be discussed in a couple of years. Author recommend that health database researchers should pay enough attention about such discussions and should make rational proposals where necessary.

View full abstract