Use of directed acyclic graphs (DAGs) to identify confounders in applied health research: review and recommendations.
Tennant PWG, Murray EJ, Arnold KF, Berrie L, Fox MP, Gadd SC, Harrison WJ, Keeble C, Ranker LR, Textor J, Tomova GD, Gilthorpe MS, Ellison GTH. · 2021
Read Use of directed acyclic graphs (DAGs) to identify confounders in applied health research: review and recommendations. by Tennant PWG, Murray EJ, Arnold KF, Berrie L, Fox MP, Gadd SC, Harrison WJ, Keeble C, Ranker LR, Textor J, Tomova GD, Gilthorpe MS, Ellison GTH. (2021) 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/europmc-pmc8128477.txt.
Description
Tennant and co-authors review how health researchers actually use directed acyclic graphs, and find wide inconsistency. These diagrams map assumed causes before any statistics run. Done well, they name exactly which confounders to adjust for. The team gathered 234 articles from 1999 to 2017 that used the method. Reporting habits varied at every step they checked.
Only a fifth of the articles, 48 in total, stated their target estimand. Half, 115 papers, reported the adjustment set their graph implied. Two-thirds, 144 articles, shared at least one of their diagrams. The average graph held 12 nodes and 29 connecting arcs. Over a third of the graphs included unmeasured variables. The authors close with concrete reporting recommendations. Estimands, diagrams, and adjustment sets should all appear in print. A tool for clarity is too often used opaquely. Better habits would make causal claims easier to audit.
Catalog record
| CREATOR | Tennant PWG, Murray EJ, Arnold KF, Berrie L, Fox MP, Gadd SC, Harrison WJ, Keeble C, Ranker LR, Textor J, Tomova GD, Gilthorpe MS, Ellison GTH. |
|---|---|
| YEAR | 2021 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 49 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |