서브메뉴
검색
Some Statistical Methods to Deal With Patient Heterogeneity in Clinical Trial Design and Causal Inference
Some Statistical Methods to Deal With Patient Heterogeneity in Clinical Trial Design and Causal Inference
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202103550
- ISBN
- 9798315795599
- DDC
- 574
- 저자명
- Chen, Yi.
- 서명/저자
- Some Statistical Methods to Deal With Patient Heterogeneity in Clinical Trial Design and Causal Inference
- 발행사항
- [Sl] : The University of Wisconsin - Madison, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 112 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-12, Section: B.
- 주기사항
- Advisor: Chen, Guanhua.
- 학위논문주기
- Thesis (Ph.D.)--The University of Wisconsin - Madison, 2025.
- 초록/해제
- 요약Heterogeneous patient data offers both unique opportunities and challenges in biomedical research. Effectively integrating this information while addressing the complexities it introduces is a crucial area of study. My research focuses on two scenarios where heterogeneous patient data are encountered, with the goal to develop methods that improve the reliability and efficiency of the statistical analysis. The first scenario is about incorporating patients' natural ordering information in randomized phase II studies. The exploratory nature of phase II trials makes it quite common to include heterogeneous patient subgroups with different prognoses in the same trial. Incorporating such patient heterogeneity or stratification into statistical calculation can improve efficiency and reduce sample sizes in single-arm phase II trials with binary outcomes. However, such consideration is lacking in randomized phase II trials. In Chapter 1, we propose methods that can utilize some natural order information which may exist in stratified population to gain statistical efficiency for randomized phase II designs. We consider both binary and time-to-event outcomes in our development. Compared with methods that do not use ordering information, our method is shown to improve the probabilities of correct selection and reduce sample size in our simulation and real examples. We also developed its related R package constrselect and we discuss its key functions and implementation in Chapter 2.The second scenario addresses the problem of causal generalization, where differences in the distribution of treatment effect modifiers across populations, known as covariate shift, can result in varying ATEs. Chen et al. (2023) introduced a weighting method to estimate the target ATE using only summary-level information from a target sample while accounting for the possible covariate shifts. However, the asymptotic variance of the estimate was shown to depend on individual-level data from the target sample, hindering statistical inference. In Chapter 3, we propose a resampling-based perturbation method for confidence interval construction for the estimated target ATE, utilizing additional summary-level information. We demonstrate the effectiveness of our approach through simulation and real data settings. We also developed its related R package EBalGen and we discuss its key functions and implementation in Chapter 4.
- 일반주제명
- Biostatistics
- 일반주제명
- Statistics
- 일반주제명
- Bioinformatics
- 키워드
- Clinical trials
- 키워드
- Causal inference
- 키워드
- R package
- 기타저자
- The University of Wisconsin - Madison Biomedical Data Science
- 기본자료저록
- Dissertations Abstracts International. 86-12B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017357713
■00520260202103550
■006m o d
■007cr#unu||||||||
■020 ▼a9798315795599
■035 ▼a(MiAaPQ)AAI32041741
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a574
■1001 ▼aChen, Yi.
■24510▼aSome Statistical Methods to Deal With Patient Heterogeneity in Clinical Trial Design and Causal Inference
■260 ▼a[Sl]▼bThe University of Wisconsin - Madison▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a112 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-12, Section: B.
■500 ▼aAdvisor: Chen, Guanhua.
■5021 ▼aThesis (Ph.D.)--The University of Wisconsin - Madison, 2025.
■520 ▼aHeterogeneous patient data offers both unique opportunities and challenges in biomedical research. Effectively integrating this information while addressing the complexities it introduces is a crucial area of study. My research focuses on two scenarios where heterogeneous patient data are encountered, with the goal to develop methods that improve the reliability and efficiency of the statistical analysis. The first scenario is about incorporating patients' natural ordering information in randomized phase II studies. The exploratory nature of phase II trials makes it quite common to include heterogeneous patient subgroups with different prognoses in the same trial. Incorporating such patient heterogeneity or stratification into statistical calculation can improve efficiency and reduce sample sizes in single-arm phase II trials with binary outcomes. However, such consideration is lacking in randomized phase II trials. In Chapter 1, we propose methods that can utilize some natural order information which may exist in stratified population to gain statistical efficiency for randomized phase II designs. We consider both binary and time-to-event outcomes in our development. Compared with methods that do not use ordering information, our method is shown to improve the probabilities of correct selection and reduce sample size in our simulation and real examples. We also developed its related R package constrselect and we discuss its key functions and implementation in Chapter 2.The second scenario addresses the problem of causal generalization, where differences in the distribution of treatment effect modifiers across populations, known as covariate shift, can result in varying ATEs. Chen et al. (2023) introduced a weighting method to estimate the target ATE using only summary-level information from a target sample while accounting for the possible covariate shifts. However, the asymptotic variance of the estimate was shown to depend on individual-level data from the target sample, hindering statistical inference. In Chapter 3, we propose a resampling-based perturbation method for confidence interval construction for the estimated target ATE, utilizing additional summary-level information. We demonstrate the effectiveness of our approach through simulation and real data settings. We also developed its related R package EBalGen and we discuss its key functions and implementation in Chapter 4.
■590 ▼aSchool code: 0262.
■650 4▼aBiostatistics
■650 4▼aStatistics
■650 4▼aBioinformatics
■653 ▼aPatient heterogeneity
■653 ▼aClinical trials
■653 ▼aCausal inference
■653 ▼aOrder constrained strata
■653 ▼aR package
■690 ▼a0308
■690 ▼a0769
■690 ▼a0715
■690 ▼a0463
■71020▼aThe University of Wisconsin - Madison▼bBiomedical Data Science.
■7730 ▼tDissertations Abstracts International▼g86-12B.
■790 ▼a0262
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17357713▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
