Intensive Care Med https://doi.org/10.1007/s00134-018-5153-5 LETTER Effect of a 24-h extended visiting policy on delirium in critically ill patients Glauco Adrieno Westphal 1,2* , Mayara Schirmer Moerschberger 1 , Dailany D’Aroz Vollmann 1 , Amanda Caroline Inácio 1 , Míriam C. Machado 1 , Geonice Sperotto 1 , Alexandre Biasi Cavalcanti 2,3 and Álvaro Koenig 1 © 2018 Springer-Verlag GmbH Germany, part of Springer Nature and ESICM Dear Editor, Absence of relatives has been suggested as a potentially modifiable risk factor for delirium and anxiety in the ICU [1, 2]. us, flexible visitation in the ICU may contribute to prevent delirium. However, this practice is not widely adopted [2–4]. We aimed to evaluate the impact of a 24-h extended visiting policy on the incidence of delirium in the ICU. Methods is was a before–after study conducted in a south Brazil- ian hospital. e records of all consecutive inpatients who remained in the ICU for more than 48 h were evaluated. Reasons for exclusion of patients from the analysis are pre- sented in the study flowchart (Supplementary Fig. 1). In phase I (March 2015–February 2016), relatives could opt for staying with patients for up to 6 h instead of the standard regimen (four half-hour visits). In phase II (March 2016–February 2017), relatives chose between an extended visitation regimen that allowed the presence of relatives at any time for up to 24 h/day or the standard regimen. Adherence was accessed by bar code check of the visitor’s identification tag. Patients were accommo- dated in individual rooms and assessed daily for delirium using the Intensive Care Delirium Screening Check- list (Supplementary Table 1). e primary outcome was the occurrence of delirium. We used Student’s t test or Mann–Whitney test to compare continuous variables, Chi-squared test for categorical variables, and Poisson regression for incidence densities. Multivariate logistic regression was used for the identification of predictors of delirium and mortality. Results We enrolled 248 and 268 patients in phases I and II, respectively. Of all patients, 260 (47.6%) were male, the mean age was 58.3 ± 19.0 years ,and the mean SAPS three score was 45.3 ± 14. Baseline characteristics and out- comes are shown in Supplementary Table 2. e adherence to the 6-h visit regimen was 44.9% in phase I, while 68.6% adhered to the 24-h visit policy in phase II (p < 0.001). Figure 1 shows that the adherence of relatives in the morning, afternoon, and night shifts increased in phase II (p < 0.001), while the incidence of delirium reduced from 12.1% to 6.7% (OR 0.52; 95% CI 0.28–0.96; p = 0.03) and the incidence density of delir- ium reduced from 29.4 to 15.9 per 1000 patient-days (p < 0.001). e extended visit was the only protective factor for the occurrence of delirium (OR 0.36; 95% CI 0.17–0.74; p = 0.005) after logistic regression adjustment for confounders. e association between visiting policy and delirium incidence was not changed across several subgroups (Supplementary Table 3). Healthcare-associ- ated infection and hospital length of stay were similar in both phases. ere was a trend towards higher mortality in phase II, although a multivariate analysis showed that mortality was associated with SAPS 3 and the need for mechanical ventilation, but not with the visiting regimen. Discussion e 24-h extended visiting policy was associated with a reduction in the incidence of delirium. e findings are consistent with previous reports which suggested that the absence of visits was associated with delirium [1], *Correspondence: glauco.ww@gmail.com 1 Centro Hospitalar Unimed de Joinville, Joinville, Santa Catarina, Brazil Full author information is available at the end of the article