서브메뉴
검색
Statistical Approaches to Heterogeneity in Alzheimer's Disease and Related Dementias
Statistical Approaches to Heterogeneity in Alzheimer's Disease and Related Dementias
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260311091514.5
- ISBN
- 9798286425334
- DDC
- 610.72
- 서명/저자
- Statistical Approaches to Heterogeneity in Alzheimers Disease and Related Dementias / Elizabeth Pirraglia
- 발행사항
- [Sl] : New York University, 2025
- 형태사항
- 1 electronic resource (150 pages)
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-12, Section: B.
- 주기사항
- Advisors: Shao, Yongzhao; Troxel, Andrea Committee members: Anthopolos, Rebecca; Betensky, Rebecca; Glodzik, Lidia; Hu, Jiyuan; Osorio, Ricardo.
- 학위논문주기
- - Ph.D. : New York University, 2025.
- 초록/해제
- 요약Alzheimer's Disease and Related Dementias (ADRD) is an umbrella term for numerous progressive neurocognitive disorders that impair memory, cognition, and behavior. While research has advanced our general understanding of these disorders, important challenges remain, particularly in accurately diagnosing subtypes, identifying specific disease mechanisms, and designing effective, personalized treatments. These difficulties are compounded by the clinical and biological heterogeneity of ADRD, as well as the limitations of available data and conventional analytic methods. This dissertation seeks to address these gaps by improving diagnostic precision, leveraging flexible statistical models to better capture disease complexity, and introducing innovative approaches for integrating data across diverse research cohorts.A major challenge in ADRD research stems from the diagnostic inconsistencies across ADRD subtypes, including the underdiagnosis of coexisting pathologies in mixed dementia. These diagnostic gaps obstruct efforts to identify subtype-specific biomarkers and hinder the development of interventions tailored to individual disease trajectories. This dissertation quantifies the extent of underdiagnosis and investigates how key risk factors, such as age and the variance of the APOE gene, differentially contribute to the development of various ADRD subtypes, offering new insights into the complexity of disease presentation and progression.In addition to diagnostic challenges, most studies focus exclusively on individual dementia types, neglecting the significant comorbidity and heterogeneity inherent to ADRD. This model misspecification risks producing biased findings, particularly in research on mortality and aging-related diseases where heterogeneous comorbid conditions are prevalent and non-negligible. Flexible statistical models, such as competing risks survival models, are necessary to address this complexity by analyzing multiple time-to-event outcomes simultaneously. However, the development and application of competing risks and other flexible statistical models in dementia research remain underutilized. By employing competing risks survival models, this study evaluates the varying effects of APOE on mortality across ADRD subtypes identified through analysis of post-mortem neuropathology data. The findings reveal that the influence of APOE on mortality can differ substantially between subgroups, with significantly opposing trends that would be masked in all-cause mortality analyses.The focus on individual dementia types also hinders the identification of biomarkers of different ADRD subtypes, particularly in cases with mixed dementia pathologies. To reliably diagnose mixed dementia, exploring combinations of specific biomarkers is crucial. However, comprehensive datasets containing post-mortem neuropathology in addition to imaging and fluid biomarker data, are scarce. To bridge this gap, an approach called "Multi-Cohort Bridging" is introduced. This approach is designed to link large reference cohorts with neuropathological outcomes to smaller cohorts with diverse biomarker data, enabling more robust analyses.Together, these contributions advance the understanding of mixed dementia, enhance methods for capturing ADRD heterogeneity, and support the development of more accurate diagnostic tools and personalized treatment strategies.
- 언어주기
- English
- 일반주제명
- Biostatistics
- 일반주제명
- Aging
- 일반주제명
- Pathology
- 일반주제명
- Neurosciences
- 키워드
- APOE gene
- 키워드
- Biomarkers
- 키워드
- Competing risks
- 키워드
- Etiology
- 키워드
- Mixed dementia
- 기타저자
- New York University Basic Medical Science
- 기본자료저록
- Dissertations Abstracts International. 86-12B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260311s2025 us eng d■001000017356798
■00520260311091514.5
■006m o d
■007cr|nu||||||||
■020 ▼a9798286425334
■040 ▼aMiAaPQD▼beng▼cMiAaPQD▼erda
■082 ▼a610.72
■1001 ▼aPirraglia, Elizabeth▼eauthor.
■24510▼aStatistical Approaches to Heterogeneity in Alzheimer's Disease and Related Dementias ▼cElizabeth Pirraglia
■260 ▼a[Sl]▼bNew York University▼c2025
■264 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a1 electronic resource (150 pages)
■336 ▼atext▼btxt▼2rdacontent
■337 ▼acomputer▼bc▼2rdamedia
■338 ▼aonline resource▼bcr▼2rdacarrier
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-12, Section: B.
■500 ▼aAdvisors: Shao, Yongzhao; Troxel, Andrea Committee members: Anthopolos, Rebecca; Betensky, Rebecca; Glodzik, Lidia; Hu, Jiyuan; Osorio, Ricardo.
■5021 ▼bPh.D.▼cNew York University▼d2025.
■520 ▼aAlzheimer's Disease and Related Dementias (ADRD) is an umbrella term for numerous progressive neurocognitive disorders that impair memory, cognition, and behavior. While research has advanced our general understanding of these disorders, important challenges remain, particularly in accurately diagnosing subtypes, identifying specific disease mechanisms, and designing effective, personalized treatments. These difficulties are compounded by the clinical and biological heterogeneity of ADRD, as well as the limitations of available data and conventional analytic methods. This dissertation seeks to address these gaps by improving diagnostic precision, leveraging flexible statistical models to better capture disease complexity, and introducing innovative approaches for integrating data across diverse research cohorts.A major challenge in ADRD research stems from the diagnostic inconsistencies across ADRD subtypes, including the underdiagnosis of coexisting pathologies in mixed dementia. These diagnostic gaps obstruct efforts to identify subtype-specific biomarkers and hinder the development of interventions tailored to individual disease trajectories. This dissertation quantifies the extent of underdiagnosis and investigates how key risk factors, such as age and the variance of the APOE gene, differentially contribute to the development of various ADRD subtypes, offering new insights into the complexity of disease presentation and progression.In addition to diagnostic challenges, most studies focus exclusively on individual dementia types, neglecting the significant comorbidity and heterogeneity inherent to ADRD. This model misspecification risks producing biased findings, particularly in research on mortality and aging-related diseases where heterogeneous comorbid conditions are prevalent and non-negligible. Flexible statistical models, such as competing risks survival models, are necessary to address this complexity by analyzing multiple time-to-event outcomes simultaneously. However, the development and application of competing risks and other flexible statistical models in dementia research remain underutilized. By employing competing risks survival models, this study evaluates the varying effects of APOE on mortality across ADRD subtypes identified through analysis of post-mortem neuropathology data. The findings reveal that the influence of APOE on mortality can differ substantially between subgroups, with significantly opposing trends that would be masked in all-cause mortality analyses.The focus on individual dementia types also hinders the identification of biomarkers of different ADRD subtypes, particularly in cases with mixed dementia pathologies. To reliably diagnose mixed dementia, exploring combinations of specific biomarkers is crucial. However, comprehensive datasets containing post-mortem neuropathology in addition to imaging and fluid biomarker data, are scarce. To bridge this gap, an approach called "Multi-Cohort Bridging" is introduced. This approach is designed to link large reference cohorts with neuropathological outcomes to smaller cohorts with diverse biomarker data, enabling more robust analyses.Together, these contributions advance the understanding of mixed dementia, enhance methods for capturing ADRD heterogeneity, and support the development of more accurate diagnostic tools and personalized treatment strategies.
■546 ▼aEnglish
■590 ▼aSchool code: 0146
■650 4▼aBiostatistics
■650 4▼aAging
■650 4▼aPathology
■650 4▼aNeurosciences
■653 ▼aAPOE gene
■653 ▼aBiomarkers
■653 ▼aCompeting risks
■653 ▼aEtiology
■653 ▼aMixed dementia
■653 ▼aMulti-Cohort Bridging
■7102 ▼aNew York University▼bBasic Medical Science.▼edegree granting institution.
■7201 ▼aShao, Yongzhao▼edegree supervisor.
■7201 ▼aTroxel, Andrea▼edegree supervisor.
■7730 ▼tDissertations Abstracts International▼g86-12B.
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17356798▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


