Using the prognostic score to design stronger observational studies
Abstract/Contents
- Abstract
- The discipline of causal inference has risen to great importance - particularly in medicine - because of the increasing abundance of passively-collected observational data and the urgent burden of causal questions too numerous to be addressed with individual randomized experiments. For decades, popular causal inference approaches have involved matching or adjustment based on a propensity score, which summarizes measured baseline variation associated with treatment assignment. However, the propensity score does not capture all of the baseline variation which may be important to a causal question. The prognostic score, encapsulating baseline variation associated with the potential untreated outcome presents a promising -- though underdeveloped -- complement to the propensity score. I present several methods which leverage the prognostic score alongside the propensity score to facilitate stronger study designs: After addressing prognostic score estimation using a new pilot design approach, I introduce and characterize (1) a study design for jointly matching subjects on propensity and prognostic scores, (2) prognostic score stratification approach for large matching studies, and (3) the assignment-control plot, a versatile visualization tool for causal inference study design, methodological research, and education. The more clearly we characterize meaningful aspects of baseline variation -- such as the prognostic score -- the more effectively we can design and communicate about observational causal inference studies
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2022; ©2022 |
| Publication date | 2022; 2022 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Aikens, Rachael Caelie |
|---|---|
| Degree supervisor | Baiocchi, Michael |
| Degree supervisor | Chen, Jonathan H |
| Thesis advisor | Baiocchi, Michael |
| Thesis advisor | Chen, Jonathan H |
| Thesis advisor | Simard, Julia |
| Degree committee member | Simard, Julia |
| Associated with | Stanford University, Department of Biomedical Informatics |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Rachael C. Aikens |
|---|---|
| Note | Submitted to the Department of Biomedical Informatics |
| Thesis | Thesis Ph.D. Stanford University 2022 |
| Location | https://purl.stanford.edu/ym932yw6908 |
Access conditions
- Copyright
- © 2022 by Rachael Caelie Aikens
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...