2022 Fiscal Year Final Research Report
Statistical methods for categorical variables used in clinical trials
| Project/Area Number |
21K20331
|
|---|
| Research Category |
Grant-in-Aid for Research Activity Start-up
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
0201:Algebra, geometry, analysis, applied mathematics,and related fields
|
|---|
| Research Institution | Yokohama City University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2021-08-30 – 2023-03-31
|
| Keywords | 分割表解析 / 順序カテゴリカルデータ / 統計モデル / 臨床試験 |
|---|
| Outline of Final Research Achievements |
To address the issues related to qualitative data analysis in clinical studies, our objective was to develop novel analytical methods. We considered the following issues: (1) An analysis method was devised to accommodate cases where clinical rating scales reclassified the original categories into three groups based on clinical interest; (2) Another analysis method was considered to address situations involving the examination of transitions at multiple time points after treatment; (3) Bayesian methods were considered for an alternative analysis approach; (4) Categorical Visualization techniques were considered to facilitate intuitive interpretation.
The findings from these studies were presented on four occasions at national conferences. Based on comments from the audience, we drafted research papers and submitted them to international peer-reviewed journals, and one of the papers was accepted for publication.
|
| Free Research Field |
カテゴリカルデータ解析
|
| Academic Significance and Societal Importance of the Research Achievements |
臨床研究で用いられる質的な臨床評価尺度に対して，量的な集計解析で評価している事例が散見される．等間隔性の無い質的データに対して，平均値を算出したり，Student's t-testを適用したりすることは，誤った結果解釈を導くことに懸念がある．
本研究では，臨床研究における質的データの解析に対して有用な結果解釈ができる解析手法の開発を検討した．本研究から得られた研究成果は従来の解析手法よりも拡張した解析手法となっており，応用面だけではなく理論面についても重要な成果である．
|
