Selecting survivors changes the population to which results apply.
What the supplied paper shows
A Korean claims analysis included 6,809 ARDS patients who survived at least sixty days. Recorded delirium was associated with several features, including continuous benzodiazepine and propofol infusions. Cases were identified through diagnosis coding.
Limits of interpretation
Early deaths were excluded by design. Coded delirium can miss episodes detected by regular bedside assessment, and medication selection may reflect illness severity. Findings among survivors do not directly describe all admitted patients.
Implications for case review
When discussing the study, disclose the survival condition before interpreting prognosis. Do not use a survivor-only analysis to establish that a medicine does or does not affect early mortality. For an individual, review sedation timing and actual cognitive assessments. The source supports awareness of selection and ascertainment limits rather than a categorical drug-causation conclusion.
Source and related reading
Source: PMC8559136, supplied full-text paper. Findings are paraphrased; the forensic discussion is an interpretation.
Related: Dominant Midazolam Sedation Was Associated With ICU Delirium in a Matched Cohort.
Leave a Reply