Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

University of Waterloo

$27,000.00

Statistical issues and methods in causal survival analysis

FY 2023-24

Summary
Randomized controlled trials are the gold standard for establishing causal relationships between a treatment strategy and an outcome of interest. However, when randomized controlled trials are not possible (e.g., too expensive, unethical to conduct), observational studies become essential to compare the effectiveness of different treatment strategies. Despite the rich literature on causal survival analysis, there still remain many open problems that stem from data structure complications: How can multiple competing events be handled for survival outcomes? How can rare events and low treatment uptake be handled for survival outcomes? How should we handle interval-censored outcomes? In addition to those, there remain open questions with the cutting-edge estimation procedures often used in causal survival analysis: How can we create monotonic survival curves using doubly and multiply robust estimators? How can we incorporate machine learning algorithms into these estimators while still guaranteeing model double-robustness? This program aims to address commonly overlooked issues in causal survival analysis, and the proposed methods will advance the state-of-the-art doubly and multiply robust methodologies in the area. The application of these new methods will have real world impact by potentially influencing policy decisions in health research worldwide. For instance, one applied area of interest is to understand the effect of antiretroviral therapies on comorbidities of HIV such as cancer and cardiovascular diseases. Another area is to evaluate strategies for treating long-term HIV comorbidities, co-infections and complications on all-cause mortality in those with HIV and Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) co-infection. The long-term goal of this research program is to develop principled statistical methods to compare the effectiveness of treatment strategies on clinical survival outcomes. One tangible outcome from this program will be the development of readily available software packages which will allow clinical researchers and epidemiologists to integrate our proposed methodologies into their work. A central focus of the proposal is to support up to ten HQPs who will become skilled statisticians and independent researchers by the end of their degree program. In addition to gaining expertise in their respective areas of research and developing novel state-of-the-art methodologies, they will also have plenty of opportunities to acquire strong communication, technical and critical thinking skills which will prepare them well for careers post graduation.
Program
Discovery Grants Program - Individual
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
RGPIN-2023-03641
Name nom
Wen, Lan
Department département
Statistics and Actuarial Science
Organization id
33
Institution établissement
University of Waterloo
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
27000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1508
Committee name en
Mathematics and Statistics
Committee name fr
Mathématiques et statistique
Area of application code
1211
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Medical and health sciences
Area of application fr
Sciences médicales et de la santé
Research subject code
9999
Research subject group en
Not available
Research subject group fr
Not available
Research subject en
Non disponible
Research subject fr
Non disponible
Application title
Statistical issues and methods in causal survival analysis
Keyword
Causal inference; Causal survival analysis; Competing events; Double robustness; Longitudinal data; Machine learning; Multiple robustness; Observational studies
Application summary
Randomized controlled trials are the gold standard for establishing causal relationships between a treatment strategy and an outcome of interest. However, when randomized controlled trials are not possible (e.g., too expensive, unethical to conduct), observational studies become essential to compare the effectiveness of different treatment strategies. Despite the rich literature on causal survival analysis, there still remain many open problems that stem from data structure complications: How can multiple competing events be handled for survival outcomes? How can rare events and low treatment uptake be handled for survival outcomes? How should we handle interval-censored outcomes? In addition to those, there remain open questions with the cutting-edge estimation procedures often used in causal survival analysis: How can we create monotonic survival curves using doubly and multiply robust estimators? How can we incorporate machine learning algorithms into these estimators while still guaranteeing model double-robustness? This program aims to address commonly overlooked issues in causal survival analysis, and the proposed methods will advance the state-of-the-art doubly and multiply robust methodologies in the area. The application of these new methods will have real world impact by potentially influencing policy decisions in health research worldwide. For instance, one applied area of interest is to understand the effect of antiretroviral therapies on comorbidities of HIV such as cancer and cardiovascular diseases. Another area is to evaluate strategies for treating long-term HIV comorbidities, co-infections and complications on all-cause mortality in those with HIV and Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) co-infection. The long-term goal of this research program is to develop principled statistical methods to compare the effectiveness of treatment strategies on clinical survival outcomes. One tangible outcome from this program will be the development of readily available software packages which will allow clinical researchers and epidemiologists to integrate our proposed methodologies into their work. A central focus of the proposal is to support up to ten HQPs who will become skilled statisticians and independent researchers by the end of their degree program. In addition to gaining expertise in their respective areas of research and developing novel state-of-the-art methodologies, they will also have plenty of opportunities to acquire strong communication, technical and critical thinking skills which will prepare them well for careers post graduation.