Indiscriminate adjustment for confounders is worse than you think and what can be done about it
Harsh Sinha, Pradeep Reddy Raamana · 2025
Read Indiscriminate adjustment for confounders is worse than you think and what can be done about it by Harsh Sinha, Pradeep Reddy Raamana (2025) 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-11715647.txt.
Description
Harsh Sinha and Pradeep Raamana warn against blind tweaks in prediction models in 2025. Health data sets hold many extra factors that can bias a model. The common habit is to strip out all such factors at once. Many believe that dropping more factors can only help, or at worst do no harm.
The authors show that this habit can add bias rather than remove it. Whether a factor helps or hurts depends on the data at hand. They offer a way to measure the bias from any chosen set of factors. Their method works broadly and can guide safer model building from health records.
Catalog record
| CREATOR | Harsh Sinha, Pradeep Reddy Raamana |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 2 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |