Double Sampling for Informatively Missing Data in Electronic Health Record‐Based Comparative Effectiveness Research
Alexander W. Levis, Rajarshi Mukherjee, Rui Wang, Heidi Fischer, Sebastien Haneuse · 2024
Read Double Sampling for Informatively Missing Data in Electronic Health Record‐Based Comparative Effectiveness Research by Alexander W. Levis, Rajarshi Mukherjee, Rui Wang, Heidi Fischer, Sebastien Haneuse (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11639654.txt.
Description
Alexander Levis and four co-authors address a common flaw in health record studies: outcomes that go missing for reasons tied to the outcome itself. Such gaps can bias treatment comparisons, and the usual fixes rely on guesses that may be wrong. Their motivating case is a study of long-term results after weight-loss surgery. They propose double sampling, where researchers go back and gather the missing outcomes for a smaller second sample.
The paper sets out when this design can recover the data's true pattern. It then builds estimators of average treatment effect that stay valid under wide conditions, and compares them in simulations with a method that adapts when it spots trouble. The authors show the design extends to other ways data can be partly hidden. The work gives researchers a firmer route to honest answers when data go missing.
Catalog record
| CREATOR | Alexander W. Levis, Rajarshi Mukherjee, Rui Wang, Heidi Fischer, Sebastien Haneuse |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |