2019 Fiscal Year Final Research Report
Establishment of effective strategy for classification of individuals who need to receive health guidance using data mining technique
| Project/Area Number |
16K09062
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Epidemiology and preventive medicine
|
|---|
| Research Institution | Sapporo Medical University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
斉藤 重幸 札幌医科大学, 保健医療学部, 教授 (60253994)
赤坂 憲 大阪大学, 医学部附属病院, 助教 (70468081)
|
|---|
| Project Period (FY) |
2016-04-01 – 2020-03-31
|
| Keywords | 特定健診・特定保健指導 / 保健事業 / 生活習慣病罹患ハイリスク者階層化 / データマイニング / 地域一般住民コホート |
|---|
| Outline of Final Research Achievements |
We attempted to effectively stratify groups of health checkup examinees for lifestyle-related disease prevention by applying a decision tree analysis and a cluster analysis in our community-based cohort study. We were able to classify health checkup examinee into several groups with similar characteristics of the subject such as risk for future occurrence of hypertension and type 2 diabetes from the results of longitudinal analyses and one’s current lifestyle including diet and demand for health service. It was suggested that health workers in local governments could implement effective lifestyle-related disease prevention with less human resource when using these classification methods.
|
| Free Research Field |
疫学・公衆衛生学
|
| Academic Significance and Societal Importance of the Research Achievements |
市町村国保がデータヘルス計画の中の保健事業を立案する上で、被保険者集団を疾病罹患リスクや、年代や生活スタイル、生活習慣改善意欲などに基づいてターゲットを絞ることは重要であるが、どのような方法で効果的に対象を分類・階層化するかの方法論はまだ確立していない。本研究によって、過去に蓄積された健診データをデータマイニングの手法を用いて分析することで、受診者の健診結果に基づき将来の疾病罹患リスクや対象特性の近い集団に効果的に分類・階層化し、市町村の保健従事者の限られた人的・経済的資源においても効果的な保健事業が運営できる可能性が示唆された。
|
