TOPPAN Holdings Adds Free-Text Data to Medical Information Analysis and Provision Service "DATuM IDEA®"

TOPPAN Holdings Inc. (President & COO: Satoru Oyama, hereinafter TOPPAN Holdings) developed the medical information analysis and provision service "DATuM IDEA®" (pronounced: Day-tum Idea) in April 2022, based on electronic health record data, and is currently providing it mainly to pharmaceutical companies. Effective July 1, 2026, free-text data held by 55 medical facilities nationwide will be added. The number of target facilities for this data linkage will be the largest in Japan for text data held by certified operators under the Next Generation Medical Infrastructure Act (Note 1).

The data to be added this time consists of anonymized text information recorded in the free-description fields of electronic health records and referral letters. Information such as doctors' treatment decisions, patient needs (chief complaints, etc.), changes in symptoms, occurrence of side effects, and reasons for drug changes/discontinuation will be extracted from the text data. By combining this with structured data with unified items and formats (disease names, drug names, drug dosages, laboratory test data, etc.), highly reliable and high-quality analysis is possible, linking not only objective treatment outcomes but also the reasons behind them.

TOPPAN will continue to support the identification of medical needs and the creation of evidence aligned with the realities of medical practice in disease areas where there are significant medical needs but no established treatments, thereby contributing to the acceleration of drug discovery.

Flow to Text Data Provision

Background

In recent years, the importance of utilizing real-world data (Note 2) based on actual clinical practice has been increasing among pharmaceutical companies. Previously, objective numerical analysis using structured data with unified items and formats (disease names, drug names, drug dosages, laboratory test data, etc.) and medical receipt data (medical fee statements) was mainstream. However, as data utilization deepens, the need for more reliable research has become apparent.

TOPPAN Holdings has designated "Healthcare" as a strategic focus business in its medium-term management plan and provides the medical information analysis and provision service "DATuM IDEA®". This service utilizes anonymized electronic health record data pre-processed by certified operators under the Next Generation Medical Infrastructure Act to analyze treatment progress in actual clinical practice, contributing to the acceleration of drug development and the identification of medical needs.

By adding text data recorded in the free-description fields of electronic health records and referral letters to "DATuM IDEA®", it will enable more reliable real-world data analysis, including the clinical background of doctors and patients, in addition to analysis using structured data with unified items and formats. This will contribute to the identification of medical needs in disease areas where treatments are not well-established and to the acceleration of drug discovery.

Features

- Analysis of doctors' treatment decisions and patient needs, which were difficult to grasp with structured data with unified items and formats

Text data recorded in electronic health records, referral letters, etc., is structured using natural language processing and keyword extraction technologies (Note 3). This allows for the analysis of more detailed clinical information that was difficult to grasp with structured data alone, which has unified items and formats. It enables the analysis of the basis for doctors' treatment decisions leading to drug changes/discontinuation, patient symptoms such as pain and nausea, and the occurrence of side effects.

This is particularly beneficial for advanced analysis in rare disease areas where the number of patients is limited and a detailed understanding of disease status and treatment progression is required.

- Efficient provision of vast amounts of patient data

Text data held by 55 medical facilities nationwide will be added. This number of target facilities is the largest in Japan for text data held by certified operators under the Next Generation Medical Infrastructure Act.

TOPPAN obtains data that has been anonymized in advance by ICI Corporation, commissioned by the Japan Medical Association Information Management Organization (J-MIMO), so no new anonymization processing is required, shortening the time until text data can be provided.

In particular, for cancer and inflammatory bowel disease, which have high demand from pharmaceutical companies, text data has been pre-structured for numerous items and can be provided smoothly. The company plans to expand the target diseases according to needs.

Future Goals

TOPPAN Holdings will widely deploy "DATuM IDEA®" with added text data, primarily to pharmaceutical companies. Furthermore, in disease areas with significant medical importance but no established treatments, the company will continue to support the identification of medical needs and the creation of evidence aligned with the realities of medical practice, contributing to the acceleration of drug discovery.

*Note 1: Next Generation Medical Infrastructure Act: Officially titled the "Act on the Law Concerning Anonymously Processed Medical Information and Fictitiously Processed Medical Information for the Promotion of Research and Development in the Medical Field," it stipulates the certification of entities engaged in the creation of anonymously processed medical information and fictitiously processed medical information, and regulations concerning the handling of medical information, anonymously processed medical information, fictitiously processed medical information, etc., for the promotion of research and development in the medical field.

*Note 2: Real-world data: A general term for medical data obtained in daily clinical practice.

*Note 3: The analysis logic using natural language processing is in the process of patent application.

* Product and service names mentioned in this press release are trademarks or registered trademarks of their respective companies.

* The content of this press release is current as of the date of announcement. It is subject to change without notice.

End