| Project/Area Number |
17H04142
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Medical and hospital managemen
|
|---|
| Research Institution | Niigata University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2017-04-01 – 2021-03-31
|
| Project Status |
Completed (Fiscal Year 2022)
|
| Budget Amount *help |
¥9,880,000 (Direct Cost: ¥7,600,000、Indirect Cost: ¥2,280,000)
Fiscal Year 2020: ¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
Fiscal Year 2019: ¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
Fiscal Year 2018: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2017: ¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
|
|---|
| Keywords | リスクマネジメント / 医療安全管理 / 自然言語処理 / グローバルトリガーツール / ディープラーニング / 有害事象 / 機械学習 |
|---|
| Outline of Final Research Achievements |
To determine the exact frequency of adverse events, we planned to construct a system that automatically detects events that lead to the discovery of adverse events from the hospital information system and determines whether the detected information is truly an adverse event by referring to the medical information surrounding the trigger. It was easy to handle Group A triggers, which can be detected from quantitative information such as order entry and test result records. However, the Group B trigger, which detects adverse events from textual information such as progress records and hospital admission summaries, was difficult to implement because the positive predictive value was extremely low. The use of various text mining methods (Transformer, BERT, GPT-3, etc.), whose development has been accelerated recently, improved the positive predictive value, but the research period was terminated in the middle of the study.
|
| Academic Significance and Societal Importance of the Research Achievements |
WHOによれば医療による重大な有害事象は入院患者の10％で発生している。重大でない事象やニアミス例を含めれば、膨大な数が発生していると推測される。これらの事例は一般的にはインシデントレポートを使って把握されているが、インシデントレポート報告は強制ではなく、報告されない事例が多数ある。本邦の病院における電子カルテの導入率は、令和2年の時点で400床以上の病院では91.2%にも上る。電子的かつ自動的に有害事象やニアミス事例を検出することができれば、その意義は極めて高い。今回は最終的な目的を達成することはできなかったが、今後の研究の礎としての意義はあったと考えている。
|
