The portfolio effect: an opportunity for improving handoffs quality in ICU.

Background: Handoffs are a major determinant of patient's safety but their implementation remains heterogeneous and non-standardized. Organizational factors, including the order in which individual cases are handled within the handoff, may play a role in their quality. We aimed to confirm the existence of the portfolio effect (e.g. a decrease in duration allocated to individual cases as the global handoff progresses) in ICU's morning medical handoffs.

Methods: Two research assistants observed the morning handoffs in two ICUs (ICU-1, a 20-bed trauma and surgical ICU and ICU-2 a 10-bed medical and surgical ICU) within a university hospital, over a 6-month period. They were trained to measure the duration of each case (i.e., the handoff of a single patient). Patients' socio-demographic and clinical data were extracted from electronic medical records. The effect of the case position within the global handoff on its duration was determined using a linear regression after log transformation of duration. The case position was categorized as either before or after the median position (first and second halves). Covariates clinically associated with handoff duration were included in the model (age, sex, Charlson comorbidities index, SAPS II score, number of organ supports, center (ICU-1 or ICU-2) and reason for admission).

Results: 2485 individual cases nested in 169 morning handoffs and related to 494 patients' stays were observed. The mean (± SD) duration of the morning handoff was 60 minutes (± 12.5) in ICU-1 and 35.2 minutes (± 10.6) in ICU-2 with a mean number of cases presented of 18.9 (± 1.3) and 9.3 (± 1.0) respectively. The mean (± SD) duration of a case was 175 seconds (± 108). Trauma stays, patients severity and comorbidities, and the number of organ supports were associated with longer case handoffs. Asjusting for these covariates, cases in the second half were shorter compared to cases in the first half (RR 0.65, 95%CI (0.51 - 0.80)).

Conclusions: We confirmed the existence of a portfolio effect within ICU handoffs, emphasizing that interventions targeting handoffs' improvement should focus on the content and the setting. We suggest avoiding the presentation of a same patient systematically at the end of the round.