Japan's Reputation for Quality
The roots of continuous quality improvement in health care can be traced back to "Kaizen (continuous improvement involving everyone)," a core concept of quality improvement in Japanese industry (1). In 1989, in the Sounding Board of the New England Journal of Medicine, Donald Berwick introduced "Kaizen" as a potentially crucial concept for improving health care in the United States, stating that, "In the discovery of imperfection lies the chance for processes to improve" (2).
Ironically, Japan's reputation for manufacturing quality products was not reflected in the quality of the country's health care. In spite of the strong government regulation of financing to assure universal coverage and free access to health care, the issue of quality of care had been left largely to health professionals, resulting in variability of care (3). In the last 15 years, emphasis has thus been placed on the health care quality in Japan, as evidenced by the Japan Council for Quality Health Care (JCQHC) accrediting hospitals much like the Joint Commission does in the United States. Also, professional societies are becoming more actively involved in quality measurement efforts to improve quality of care (4-6).
At present, Japan has no means to implement national, routine quality measurement, but there is a growing interest among various stakeholders, including accreditors, researchers, public payers, and care organizations. For such implementation to occur, progress must be made on potential national data sources for these measures.
National Databases of Healthcare Information
In Japan, there are several national databases to collect and store the profiles of health care institutions or individuals who have used health care services; each one, however, operates independently with different purposes. For example, the sole use of the Health Insurance Claims database was for billing purposes until recently, when the government changed its policy on this resource's use. Such movement toward utilization of national databases raises expectations for quality measurement.
Health Insurance Claims Database
The Health Insurance Claims database encompasses all medical, dental, and pharmaceutical claims from both the public and private sectors in Japan. Since the late 1990's, healthcare organizations—some 8,739 hospitals and 99,635 clinics as of 2009—have transitioned from using paper-based to electronic claim systems. Currently, claims for reimbursement are processed electronically for 99.9% of inpatient and 91.0% of outpatient care encounters. The medical claims data include patient demographics, diagnosis, inpatient and outpatient encounters (amount billed), medical and surgical procedures, and imaging and laboratory tests. Researchers are now participating in a 2-year feasibility study to explore secondary use of the database for health services research.
Diagnosis Procedure Combination/Per-Diem Payment System (DPC/PDPS) Database
Another national database is the Diagnosis Procedure Combination/Per-Diem Payment System (DPC/PDPS) database, a promising information source from a quality perspective, introduced in 2003 (7). It is similar to the National Inpatient Sample database in the United States and contains data on 2,658 diagnostic procedures performed in 1,391 participating acute care hospitals in Japan, representing 50.4% of all beds other than those for chronic diseases, tuberculosis, and psychiatry.
The DPC/PDPS database stores detailed clinical information (e.g., smoking index, TNM classification of malignant tumors, Activity of Daily Living Scale, Killip classification of acute myocardial infarction), as well as administrative data, such as patient background and length of hospital stay. Furthermore, the utilization of surgical procedures, drugs, and devices is recorded daily during the hospital stay. At present, the database's primary use is for billing purposes, although it has been employed in part for research purposes, such as the assessment of hospital performance with a case-mix standardized mortality model (8).
Other databases have also added limited information on structures, processes, and outcomes that may prove useful for quality measurement:
- Inpatient utilization data on all patients in the Hospital Report (e.g., length of stay and number of hospital admissions)
- Structural information of all facilities from the Healthcare Facility Survey (e.g., the number of beds)
- A Patient Survey of a randomly selected sample of the population annually since 1953 and every three years since 1984. In addition to identifying the number of patient visits by region and admission profiles (e.g., diagnosis and use of emergency care), the data include a few process and outcome elements, such as vital statistics relevant to life expectancy and causes of death (9).
Future Opportunities and Challenges of Data Utilization
New information technology creates platforms for scalable innovation. Information can thus be aggregated across databases for monitoring and identifying trends in health care quality, utilization, and costs. Analyzing data at the national level helps to highlight priority areas in health care policies, and comparability of data across health care organizations and regions has the potential to create benchmarks for improving the quality of care.
In principle, the national claims data should provide traceability of inpatient and outpatient care for everyone in Japan. Analysis of the DPC/PDPS database will allow researchers to analyze treatment patterns, such as the use of combination therapies, treatment duration, and outcomes. By analyzing the accumulated data, the quality of acute care can be measured with adjustment for patient characteristics and disease severity.
The Health Insurance Claims and the DPC/PDPS data stored in the claims' databases are now ready for linking to individual patients by the assignment of a common identification number. With this project underway, an increase in the amount of data available in two years' time is expected. For example, longitudinal analyses of inpatient and outpatient prescribing patterns will be possible based on the identification of beneficiaries over time, allowing for the assessment of quality of care for patients with chronic conditions.
However, we must overcome several challenges to make full use of the data. As none of the databases is perfect, linkage across these resources is crucial. For example, individual patient-based information stored in the claims databases and the Patient Survey or vital statistics information are not directly linked at present, but combined analysis of all these data could provide geographically defined population-based information on resource utilization and outcomes. Yet, such linking is not easily accomplished because of inconsistencies in data definition and limited matching variables.
Another challenge is relating the Health Insurance Claims data to treatment in comorbid or complicated cases. This requires case-mix adjustment, and sufficient clinical information is not available to achieve this. Indeed, combining the currently available databases has limitations in terms of the type of data content. Information may not be what is required, as it is not originally collected for the purpose of quality assessment. Focus should therefore be placed on measurable priorities and the development of new methodologies to measure the quality of care using these databases.
The voluntary basis of DPC/PDPS ultimately limits its completeness. Only about 50% of all hospitals currently submit data, and mandatory participation is limited to advanced medical facilities such as university hospitals. Moreover, some health care institutions show reluctance to submit data because of their concern that sharing this information would influence reimbursement. Recently, a non-profit organization has been established to support appropriate use of health care information and relevant research. Hopefully, this organization will facilitate further participation.
Currently, very few researchers can handle national data due to the complexity and quantity of information. For more institutionalized and systematic quality measurement, specialists need training with sufficient analytic skills and knowledge of quality management.
Another important issue is who has access to these national databases as they are now strictly limited to a handful of researchers who have fulfilled the government's requirements. Increasing the use of data increases the risk of data breaches, warranting a review of privacy and security policies for these resources. The use of the national databases in research should be granted only to those engaged in work that will be of public benefit.
Overall, the use of the national databases does indeed have great potential to improve the quality of patient care. However, the challenges addressed here warrant a comprehensive strategy, including an establishment of a technical and authorizing environment to ensure proper use of the data.
Hiroto Ito, PhD
National Institute of Mental Health, National Center of Neurology and Psychiatry, Tokyo, Japan
The views and opinions expressed are those of the authors and do not necessarily state or reflect those of the National Quality Measures Clearinghouse™ (NQMC), the Agency for Healthcare Research and Quality (AHRQ), or its contractor, ECRI Institute. Additionally, the views and opinions expressed are those of the author and do not necessarily state or reflect those of the National Center of Neurology and Psychiatry.
The author appreciates the feedback on DPC/PDPS given by Professor Kiyohide Fushimi, Health Policy and Informatics Section, Department of Health Policy, Tokyo Medical and Dental University.
Potential Conflicts of Interest
Dr. Ito declared no potential conflicts of interest with respect to this expert commentary.
Dr. Ito reports the following business/professional interests: Visiting Professor, Graduate School of Medical Science, University of Yamanashi; Surveyor, Japan Council for Quality Health Care; Board Member, Japanese Society of Social Psychiatry; and Advisory Board Member, Japanese Association of Psychiatric Hospitals.
Dr. Ito reports the following guideline/measure development involvement: member of the International Initiative for Mental Health Leadership (IIMHL) Clinical Leaders Project (project to develop and implement quality measurement framework across 12 countries: Australia, Canada, England, Germany, Ireland, Japan, Netherlands, Norway, New Zealand, Scotland, Taiwan, and the United States); Japan Council for Quality Health Care Committee Member on development of standards and criteria for psychiatric services.
- Imai M. Kaizen: the key to Japan's competitive success. McGraw-Hill/Irwin, 1986.
- Berwick DM. Continuous improvement as an ideal in health care. N Engl J Med 1989;320:53-56.
- Hashimoto H, Ikegami N, Shibuya K, et al. Cost containment and quality of care in Japan: is there a trade-off? Lancet 2011 (Early online publication). Accessed September 19, 2011.
- Ito H, Iwasaki S, Nakano Y, et al. Directions of quality improvement activities of health care organizations in Japan. Int J Qual Health Care 1998;10:361-363.
- Ito H, Sugawara H. Relationship between accreditation scores and the public disclosure of accreditation reports: a cross sectional study. Qual Safe Health Care 2005;14:87-92.
- Greenfield D, Braithwaite J. Health sector accreditation research: a systematic review. Int J Qual Health Care 2008; 20:172-183.
- Matsuda S. Diagnosis procedure combination: the Japanese approach to casemix. Kimberly JR, de Pouvourville G, D'Aunno T (eds). The globalization of managerial innovation in health care. Cambridge: Cambridge University Press; 2008. 254-272.
- Miyata H, Hashimoto H, Horiguchi H, et al. Assessment of hospital performance with a case-mix standardized mortality model using an existing administrative database in Japan. BMC Health Serv Res 2010;10:130.
- Statistics Bureau of Ministry of Internal Affairs and Communications. Japan statistical yearbook 2011. Available at: http://qualitymeasures.ahrq.gov/disclaimer.aspx?redirect=http://www.stat.go.jp/english/data/nenkan/index.htm . Accessed June 6, 2011.