JPMA Media Forum. Current Status and Issues for Realization of "Japanese EHDS" - Toward Realization of Next Generation Medical Care

Mar. 18, 2026

2026年3月18日、室町三井ホール＆カンファレンス（東京都中央区）にて「製薬協メディアフォーラム」を開催しました。今回は「“日本版EHDS”実現に向けた現状と課題－次世代医療の実現に向けて－」をテーマに、製薬協産業政策委員会の安中良輔委員、The Office of Pharmaceutical Industry Researchの渡邉奈都子主任研究員、東京大学名誉教授／一般社団法人次世代基盤政策研究所代表理事の森田朗氏による講演を実施しました。当日は会場およびウェブ配信にて、メディアや会員会社含め70名以上が参加しました。

Background of the Forum

With the enactment of the Medical Care Act Amendment Law in December 2025, we expect to see progress in efforts to promote the secondary use of medical information. While there are high expectations for progress in efforts to promote the secondary use of medical information following the enactment of the revised Medical Service Act in December 2025, there are still issues that need to be resolved, and the road to resolution is not smooth. In this forum, we will discuss the Japanese version of EHDS (European Health Data Space).^* In addition to lectures by Mr. Yasunaka and Mr. Morita, members of the Cabinet Office's "Study Group on Promotion of Utilization of Medical and Other Information," on the current status and issues toward the realization of the "Act on the Utilization of Medical and Other Information," a lecture was given on the outline of the results of a "public awareness survey on the utilization of health and medical information" conducted by the National Institute for Policy Studies. The following is a transcript of the lectures.

Lecture 1

Discussions at the Study Group on Promotion of Utilization of Medical and Other Information and Future Issues

Pharmaceutical companies want to deliver innovative drugs to patients as quickly as possible through efficient drug discovery using health and medical data. However, it is not realistic to ask pharmaceutical companies to build a highly comprehensive database just for the sake of pharmaceutical companies, considering the burden on the busy medical field. Ideally, we would like to create an integrated grand design in which data generated for "primary use," which is used to improve patient health and treatment, is also used for "secondary use," such as improving the efficiency of healthcare costs for government and research and development for pharmaceutical companies and academia like ours.

There are three main benefits that pharmaceutical companies can offer to the public through the use of medical data: First is the reduction of the risk of adverse drug reactions. The first is to reduce the risk of adverse drug reactions, and we recognize that the most important thing is to use large-scale medical data to analyze drug safety information in real time and link it to appropriate use. Third, the use of genomic and other data and the improvement of the quality of research plans will reduce the number of clinical development failures and increase the probability of success in creating new drugs.

 

Specifically, there is a need to improve the efficiency of data collection by storing some or all of the control groups for clinical trials in medical data. There is also a need to track side effects that do not appear until a long period of time, such as carcinogenicity, for example, and a need for data linkage between parents and children for research on genetic diseases. To this end, it is necessary to build a foundation for "life course data," such as Japan's world-class maternal and child health handbooks and school health checkups, which link data from before birth to death.

 

The Ministry of Health, Labor and Welfare (MHLW) is currently trying to build a national medical information platform, aiming to provide better medical care to patients by linking medical, nursing, and municipal data. Most recently, the revision of the Medical Care Act, etc., enacted last December, will allow secondary use of public databases held by the MHLW in a safe, pseudonymized and anonymized form in three years' time. I am very encouraged.

In this new infrastructure, various databases can be linked, centering on the National Database (NDB). Of particular note is the fact that various medical data and "mortality data" will be linked for analysis. In terms of evaluating efficacy and analyzing the most serious side effects, it is this death data that pharmaceutical companies have needed the most, and until now, this data has not been sufficiently available from private databases.

In addition, a very high level mechanism has been designed in terms of preventing unauthorized use. Users can analyze data on the cloud, but they cannot download the data to be analyzed, which is personal information. Since the system is designed to take out only the analyzed results (non-personal information), it is structurally impossible for personal information to be leaked or distributed.

The significance of being able to analyze these databases "in conjunction" rather than in isolation is immeasurable. For example, linking the NDB, which contains prescribing and mortality data, with the National Cancer Registry, which has detailed information, will enable the most important assessments, such as overall survival (OS) after anticancer drug administration. We also expect that if the immunization database is expanded, it will also be possible to assess the side effects and efficacy of vaccines.

The European "EHDS" is truly an excellent model for promoting database collaboration in Japan. The Japanese government also showed its unwavering determination to study the EHDS as a reference and to submit a bill in next year's ordinary Diet session in the 2025 Cabinet decision. In response to this, a new study group chaired by Mr. Morita has been set up, and the JPMA will be sincerely involved in the study group.

There are two points we are particularly emphasizing in designing this system: The first is the use of generative AI (large-scale language models (LLMs)) for data structuring. In response to the current situation where data structuring is not progressing well, the use of AI should also be considered in both directions. In fact, there is a case in which "human-assessed results" and "AI-assessed results" for changes in tumor size after administration of anticancer drugs were almost identical, suggesting that AI can also be fully utilized.

Second is "international collaboration. Research and development in the medical field is no longer possible without international collaboration. Currently, the perspective of economic security is becoming very important, but looking at the number of papers using actual real world data (RWD), there is also solid collaboration between the United States and China. I believe that Japan should also balance security and international collaboration and promote the formulation of rules with international standards that can be used in cooperation with other countries.

 

Lecture 2

Overview of the results of the public awareness survey on the utilization of health and medical information

 

Based on these five perspectives, a web-based survey was conducted on 2,900 men and women aged 20 or older in 47 prefectures throughout Japan, to ensure that the survey was not biased in accordance with Japan's population composition.

The most important result is that 57.2% of respondents were in favor of "primary use," in which their medical data is used to "improve their own medical care and medical services," and 54.8% were in favor of "secondary use," in which their medical data is used to "research and develop new medicines and improve public health. Both of these categories accounted for a majority of the respondents, and there was no significant difference in acceptance between primary and secondary use. About 90% of the respondents who were in favor of primary use were also in favor of secondary use, indicating a strong overlap in acceptance awareness. On the other hand, about 30% of the respondents were undecided, indicating that there is a certain segment of the population that is not clearly opposed but is reserving judgment.

 

An international comparison of this figure shows that Japan is in the middle group, on the same level as France and the UK, but below the Organization for Economic Cooperation and Development (OECD) average of 64%.

What factors, then, increase the willingness to favor secondary use? The analysis revealed that the factor with the strongest influence was "the large number of expected benefits. Next in line were "lack of anxiety and resistance," "usual trust in the entity utilizing the data," and prerequisites such as "knowledge of anonymous processing" and "high level of health literacy and digital literacy.

In other words, the willingness to agree to the provision of data is not determined solely by fixed values, but is an aspect that can be raised well after the fact by understanding the benefits, dispelling concerns, and accumulating knowledge.

Digging deeper into the substance of the benefits, we found that the largest number of respondents simultaneously value both personal benefits, such as "access to medical care that is right for me and my family," and social benefits, such as "leading to the development of new drugs and treatments. In addition, only about 18% of those who perceive none of the benefits are in favor, while more than 80% of those who expect four or more of the benefits are in favor. A trend was clearly confirmed in which the more benefits were perceived, the more the respondents' intention to approve increased step by step.

 

We also analyzed the factors that cause anxiety about the use of data. The majority of respondents (42.7%) were "undecided," with less than 30% of the respondents feeling strongly about it. Furthermore, it is noteworthy that only less than 20% of those who feel anxious responded that they have "concrete concerns," while approximately 80% have "vague anxiety. This suggests that the public's concerns are not so much about some specific risk, but rather vague concerns stemming from a lack of information and trust in the system as a whole.

 

Finally, there is the level of trust in the accessing entity. As a result, medical institutions, universities, and research institutes were the most trusted, while government and administrative agencies and IT companies remained at relatively low levels, and we, pharmaceutical companies, were somewhere in between.

 

When asked why they were willing to provide data, "public nature" was cited for the government and "expertise" for medical institutions and pharmaceutical companies. Conversely, the main reasons for refusing to provide the data were "concerns about commerciality" for pharmaceutical and IT companies, and "lack of transparency and accountability" for the government. The most common reason given by all entities was "concern about privacy protection. This result indicates that it is essential for the recipients of data to design transparent governance based on "for what purposes, how data is handled, and how privacy is protected.

Through this survey, we found that a certain percentage of the public is in an "undecided" state, where they do not clearly express their approval or disapproval. This segment of the population is not strongly opposed, but rather lacks the information necessary to make a decision, or is not yet highly interested. In order to increase the awareness of acceptance, it is important to arouse interest and promote concrete understanding of the significance and value of utilization among this attitude reserved group.

Lecture 3

Current status and issues toward realization of Japanese version of EHDS

Secondary use," which is the subject of today's discussion, is the collection and analysis of health and medical data from the entire nation to statistically clarify "the efficacy or risk of this drug for this type of person," which can then be used for drug discovery, medical research, and optimal medical policy.

For this reason, it is important to preserve a "life log," a record of a person's health status from birth to death. This record of health status would also enable preventive medicine to find conditions before they reach a "threshold" for disease onset and delay the onset of the disease.

However, in order to connect data, it is necessary to have a consistent format (standardization) and a common "identifier (ID)" that allows for reliable identification. Japan has world-class data such as maternal and child health handbooks and school health checkups, but because each administrative department has created a partially optimized system for its own purposes, the data is not standardized and cannot be connected. In addition, the protection of personal information, with its excessive focus on "consent" and "anonymization" as a way to protect individual rights, has hindered the utilization of data. In other words, Japan lacked an overall "grand design" for how to use and connect medical information.

The sense of crisis over such delays has led to a strong policy to promote the use of data in the 2025 "Framework Policy" and the Regulatory Reform Implementation Plan, among others. In order to materialize this policy, the "Study Group on Promotion of Utilization of Medical and Other Information" chaired by me has been launched, with the aim of reaching a conclusion and submitting a bill by the summer of this year.

In this discussion, the EHDS Regulation, which entered into force in the EU in March 2025, is considered the model to be referred to in the government's cabinet decision. It is rare for a Japanese cabinet decision document to state so clearly, "Let's refer to a foreign system. Of course, it is important to create something unique to Japan, but it is also an indication of the expectation that we should take what is good and use it as a reference to create something even better, and the EHDS is truly a system worthy of reference.

The reason why the EHDS was created in the first place is because of the pandemic of the new coronavirus from 2020. In the EU, where people can come and go freely without a passport, medical data systems in each of the 27 EU member states were disparate, despite the fact that both infected people and the virus were moving across borders. Although a system was hastily created to deal with this emergency situation, this bitter experience made us keenly aware of the need to accurately grasp the infection situation in the entire EU region and make the data available. The EU is currently working on a number of initiatives to promote the use of data, including the EU General Data Protection Regulation (GDPR), the AI Act, and other cross-sectoral laws, as well as the Health and Medical Care Act, in anticipation of an era in which data will be used to drive social change. The EU is now trying to create a grand matrix legal system that combines cross-sectoral laws such as the EU General Data Protection Regulation (GDPR) and the AI Law, and domain-specific laws such as health, agriculture, and environment.

The EHDS is to accumulate a lifelog for each individual from birth to death and utilize it for daily health management and treatment (primary use), and for medical research and drug discovery (secondary use). The core of the system is the Electronic Health Record (EHR), a highly standardized and secure electronic health record. MyHealth@EU" for primary use and "HealthData@EU" for secondary use have been established as a platform to link national databases that meet the EHR standards.

The most important emphasis in primary use (MyHealth@EU) is the "patient's right to better healthcare. For example, if a Swede becomes suddenly ill in France while traveling, he or she can ask a French doctor to look at his or her electronic medical record in Sweden and request the best possible treatment. If you do not want your medical history to be known to doctors in other countries, you are also granted an opt-out right not to have your data shown, but the system says, "That is your responsibility," even if you do not receive optimal treatment as a result of the missing information. Thus, the system is designed with the patient's best interest in mind.

The key is that each country has a system in place that is obligatory. The key is the "Health Data Access Bodies" that each country is required to establish, which are contact points or regulatory bodies. The process of secondary use is as follows.

Because of this rigorous procedure, the EU Commission initially proposed that "no individual consent is required for secondary use. However, during the discussion process, "some countries have constitutionally guaranteed the right to opt-out," and although Denmark and Finland, which do not require opt-out at all, opposed the proposal, the right to opt-out was eventually recognized. However, the opt-out right is not recognized as an exception for research for public purposes, policy formation, etc. Technological approaches are also being promoted to minimize the risk of personal information being leaked without opting out by utilizing "Privacy-Enhancing Technologies" (PETs), which use sophisticated mathematical processing to make personal identification nearly impossible.

 

Data can be used for medical research, drug discovery, public health, and policy making, etc. However, the use of data for "discrimination" or "product sales and advertising" such as targeting is strictly prohibited and will be strictly monitored by access authorities.

Looking at the current situation in Japan, there are several challenges in creating the system. First, while the EU proceeds with top-down decisions regardless of whether they can or cannot be done, Japan tends to start only after it is agreed that "it can be done for sure," which inevitably leads to slow progress. Also, in terms of system structure, the EU has a "common information infrastructure on which various applications are mounted," whereas Japan has a "stove-piped" system for each administrative purpose, which is later connected by a common infrastructure, making it a very difficult task.
And, after all, "public trust" is very important. There are still people who have vague concerns about whether they will allow pharmaceutical manufacturers, who are commercial enterprises, to use their data. It is essential to explain to these people the benefits of data utilization and the strict system to ensure safety, and to gain their trust.

Question and Answer Session

(Public Affairs Committee)

Back to 2026 Announcements from Public Affairs Committee

TOP