Observational studies of early versus late salvage therapies in critical care exhibit intrinsic selection bias: two meta-analyses
Authors' conclusion
There were significant differences in the results of observational and experimental studies looking at timing of salvage therapies, partly due to selection bias in observational studies. This issue was worsened by only including subjects who receive the therapy.
Study context
- Population
- Critical care patients in observational and experimental studies of timing of renal replacement therapies (RRT) and invasive mechanical ventilation (IMV)72 RRT and 50 IMV studies
- Comparison
- Early vs late initiation of RRT and of invasive mechanical ventilation (IMV); observational vs experimental studies
- Outcome
- Mortality
Methodological detailstwo meta-analyses
- Two meta-analyses
Pooled result
As reported in the abstract
Mortality, early vs late RRT and IMV, by study design
Early vs LateOdds ratio (95% CI)
RRT, observational studies
0.52(0.42–0.63)
RRT, experimental studies
0.94(0.76–1.17)
IMV, observational studies
1.25(1.03–1.52)
IMV, experimental studies
1.86(0.90–3.86)
Further pooled results
| Outcome | OR (95% CI) | |
|---|---|---|
| Mortality, early vs late IMV, observational studies restricted to subjects who all received IMV (subgroup) | 0.75 (0.55–1.02) | |
| Mortality, early vs late IMV, studies also including never-intubated subjects (subgroup) | 1.63 (1.30–2.04) | |
| 0.512Odds ratio (log scale) |
Original publication
Observational studies of early versus late salvage therapies in critical care exhibit intrinsic selection bias: two meta-analyses. Crit Care. Published 29 September 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →