- Design
- Target-trial emulation in an ICU database
- Population
- 12,922 adult ICU stays of at least 48 hours, delirium-free at entry
- Primary outcome
- Incident delirium
- Effect
- High vs low night-time contact rate: aHR 1.43 (95% CI 1.27–1.60)
Using records from 12,922 ICU stays in the MIMIC-IV database, investigators emulated a target trial. Patients were included if they had been in intensive care at least 48 hours and screened negative for delirium at that point. Exposure was the rate of vital-sign charting, blood draws and medication administrations between 22:00 and 06:00 in the first 48 hours.
Patients with above-median night-time contact rates had a higher hazard of later delirium (adjusted HR 1.43, 95% CI 1.27–1.60), with a gradient across tertiles. Charting and phlebotomy carried more of the association than medicine rounds.
The authors are clear about the limits. Sicker patients are checked more often, and early, unrecognised delirium may itself prompt more checks — reverse causation that weighting cannot fully remove. Still, the finding fits what is known about sleep disruption and delirium, and clustering non-urgent tasks costs little.
- Review whether overnight vital signs and blood tests are clinically needed for each stable patient.
- Move routine morning blood draws away from the small hours where the laboratory allows.
- Cluster unavoidable night-time tasks so the patient is woken once rather than repeatedly.
- Screen for delirium every shift; this study cannot tell which came first, disruption or confusion.
Why it matters
Some of the delirium risk on wards and in intensive care may sit in routines rather than in the illness.
Don't overread it
This was observational — it cannot show that reducing night-time contacts prevents delirium.
The statistics, in plain English
A hazard ratio of 1.43 means delirium developed about 43% faster in the high-contact group after weighting. A target-trial emulation organises observational data like a trial, but it cannot randomise, so confounding by illness severity remains possible.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for internal medicine, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free