Grant: Discovery Grants Program - Individual
McGill University
$37,000.00
Flexible modeling of time-to-event data with time-varying covariates and data imperfections
FY 2023-24
Summary
The overall objective of my research program is to develop new methods for analyzing medical data that will help unravel complex, dynamic processes underlying the occurrence, progression, and clinical outcomes of many diseases. Such processes involve typically many factors that interact together to affect the final health outcomes of individual patients, and dissociating their separate roles requires sophisticated methods of data analysis. In addition, the importance of some risk factors may vary with the person's age or disease duration, and their relationship with the risk of the outcome may be complex. For example, both very low and high body weights are associated with higher mortality among people of the same age. Over the past two decades, thanks to continuous support from consecutive NSERC Discovery Grants, I have developed several new statistical methods that helped address these analytical challenges. Now I propose to extend this methodology further, making use of recent progress in both the mathematical foundations of statistical methods and computational resources available to efficiently perform complex analyses of large health databases. These new methods, to be developed within the current NSERC grant application, will allow data analysts to address further complications of the data collected in important long-term longitudinal studies of human health. Specifically, the values of many prognostic and risk factors (e.g., blood pressure, body weight, viral load) and many environmental or occupational hazards (e.g., air pollution levels, workplace noise) change over time. Thus, to understand the role of a particular prognostic factor, we must assess how important are its values observed, say, 5 years ago versus its values measured just a few months ago. This will allow clinicians to identify more accurately the profiles of patients who are at particularly high short-term risks of a serious disease exacerbation or death and, thus, require immediate medical attention. On the other hand, for other patients, found to be more likely to experience these adverse events only after a few years, the results of our analyses will permit assessing potential benefits of different preventive interventions. For example, they may indicate that a substantive decrease in the risk of stroke will occur about three years after an important change of diet aimed at lowering the patient's high cholesterol to the recommended level, or that the benefits of a vaccine will become evident within three months after vaccination but will start to gradually decrease about 18 months later. Based on several successful applications of our novel NSERC-supported methods in impactful clinical research studies over the past two decades, I expect that the proposed new methods will help understand the evolution of many diseases and, ultimately, reduce their negative impact, resulting in improved health of the Canadian population.
Program
Discovery Grants Program - Individual
Recipient
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
RGPIN-2023-04119
Name nom
Abrahamowicz, Michal
Department département
Department of Epidemiology, Biostatistics and Occupational Health
Organization id
61
Institution établissement
McGill University
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
37000
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
1101
Area of application group en
Health, education and social services
Area of application group fr
Santé publique, éducation et services sociaux
Area of application en
Human health (including medically-related psychological research)
Area of application fr
Santé publique (y compris recherche psychologique liée à la médecine)
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
Flexible modeling of time-to-event data with time-varying covariates and data imperfections
Keyword
Accelerated failure time model; Additive hazards; Biostatistics; Cumulative effects; Measurement errors; Non-linear effects; Simulations; Tensor splines; Time-to-event analysis; Time-varying covariates
Application summary
The overall objective of my research program is to develop new methods for analyzing medical data that will help unravel complex, dynamic processes underlying the occurrence, progression, and clinical outcomes of many diseases. Such processes involve typically many factors that interact together to affect the final health outcomes of individual patients, and dissociating their separate roles requires sophisticated methods of data analysis. In addition, the importance of some risk factors may vary with the person's age or disease duration, and their relationship with the risk of the outcome may be complex. For example, both very low and high body weights are associated with higher mortality among people of the same age. Over the past two decades, thanks to continuous support from consecutive NSERC Discovery Grants, I have developed several new statistical methods that helped address these analytical challenges. Now I propose to extend this methodology further, making use of recent progress in both the mathematical foundations of statistical methods and computational resources available to efficiently perform complex analyses of large health databases. These new methods, to be developed within the current NSERC grant application, will allow data analysts to address further complications of the data collected in important long-term longitudinal studies of human health. Specifically, the values of many prognostic and risk factors (e.g., blood pressure, body weight, viral load) and many environmental or occupational hazards (e.g., air pollution levels, workplace noise) change over time. Thus, to understand the role of a particular prognostic factor, we must assess how important are its values observed, say, 5 years ago versus its values measured just a few months ago. This will allow clinicians to identify more accurately the profiles of patients who are at particularly high short-term risks of a serious disease exacerbation or death and, thus, require immediate medical attention. On the other hand, for other patients, found to be more likely to experience these adverse events only after a few years, the results of our analyses will permit assessing potential benefits of different preventive interventions. For example, they may indicate that a substantive decrease in the risk of stroke will occur about three years after an important change of diet aimed at lowering the patient's high cholesterol to the recommended level, or that the benefits of a vaccine will become evident within three months after vaccination but will start to gradually decrease about 18 months later. Based on several successful applications of our novel NSERC-supported methods in impactful clinical research studies over the past two decades, I expect that the proposed new methods will help understand the evolution of many diseases and, ultimately, reduce their negative impact, resulting in improved health of the Canadian population.