Objectives: The ICU built environment-including the presence of windows-has long been thought to play a role in delirium. This study investigated the association between the presence or absence of windows in patient rooms and ICU delirium.

Design: Retrospective single institution cohort study. Delirium was assessed with the Confusion Assessment Method for the ICU.

Setting and patients: ICU patients between January 1, 2020, and September 1, 2023, were categorized into windowed or nonwindowed groups based on their ICU room design. The primary outcome was the presence or absence of delirium at any time during the patient’s ICU stay. Secondary outcomes included the presence of delirium during the first 7 days of the ICU stay, hospital length of stay, ICU length of stay, in-hospital mortality, pain scores, and Richmond Agitation-Sedation Scale scores.

Interventions: None.

Measurements and main results: A total of 3527 patient encounters were included in the final analysis, of which 1292 distinct patient encounters were admitted to a room without windows (37%). Delirium was observed in 21% of patients (460/2235) in windowed rooms and 16% of patients (206/1292) in nonwindowed rooms. In adjusted analyses, patients in windowed rooms were associated with an increase in the odds of the presence of delirium (odds ratio, 1.29; 95% CI, 1.07-1.56; p = 0.008). Patients in windowed rooms were found to have longer hospital (adjusted hazard ratio [aHR], 0.94; 95% CI, 0.87-1.00) and ICU length of stay (aHR, 0.93; 95% CI, 0.87-1.00) compared with patients in the nonwindowed rooms, although this was not statistically significant in adjusted analyses (p = 0.06 and 0.05, respectively). No statistically significant difference was observed in other secondary outcomes.

Conclusions: The current study provides insightful information regarding associations between a component of the ICU built environment, specifically the presence or absence of windows, and the frequency of delirium.