OR-to-ICU Handoff-Specific Contributors to Diagnostic Errors and Uncertainty
The transition from the OR to the ICU involves coordinating teams from multiple disciplines in the movement and management of critically ill patients and complex equipment. This transition is prone to technical and diagnostic error due to not only the high patient acuity and time pressure common across many care transitions, but also the competing prioritization of information among team members from different disciplines, including anesthesia, surgery, and critical care.72 Communication failures during perioperative care transitions are well recognized to contribute to medical error,73,74 including loss of important patient information, exposure to unnecessary interventions, and preventable harm.62
During the OR-to-ICU handoff, time pressure inherent in a critical care setting can shift individual clinician decision-making processes toward naturalistic decision making over deliberative thought.75,76 This type of thinking is more prone to cognitive bias and increases the likelihood of diagnostic error.
Studies examining the incidence of cognitive bias among surgeons10,77 and anesthesiologists,78 key participants in the OR to ICU handoff, have identified several common biases, including:
- Anchoring,
- Confirmation,
- Omission,
- Commission,
- Premature closure, and
- Framing effect (whether the option is presented as a loss or a gain).
These studies also aimed to quantify the number of cases affected by common cognitive biases, ranging from 32.7 percent of all cases77 to more than half of emergencies.78 These cognitive biases contribute to diagnostic error, which in turn has real consequences for patients, including death. One study by Antonacci, et al., found that death after surgery related to communication, system, diagnostic, and judgment error—all errors in which these cognitive biases are implicated—was significantly higher than death related to technical error.79
Strategies To Mitigate Diagnostic Errors and Uncertainty at OR-to-ICU Transitions
A large body of evidence shows the benefits of standardizing the perioperative handoff process for patient outcomes, information exchange, and medical error.73,75,80,81 In a systematic review and meta-analysis, Abraham, et al., identified 32 studies since 2007, the majority of which were published after 2010, that examined the impact of OR-to-ICU handoff interventions.80 Most interventions studied were process-based protocols and communication checklists to support information processing and social interactions. These interventions resulted in fewer information omissions and errors, but the statistical quality of many of the studies was low.
Systematic reviews examining the OR-to-ICU handoff intervention literature among subsets of surgical populations, such as cardiothoracic surgery,73 have reported similar findings. Most studies have been small, single center, and focused on the effectiveness of a handoff intervention rather than sustainability.
Lane-Fall, et al., have done extensive work examining both the effectiveness of a standardized OR-to-ICU handoff as well as strategies for implementation. In their initial study, Handoffs and Transitions in Critical Care (HATRICC),82 they examined the implementation of a handoff protocol using an information template in two mixed ICUs, with the primary outcome of information omissions. Standardization of the handoff practice was associated with decreased information omissions, but no change was found for ICU mortality or length of stay. The study did not explicitly look at medical error.
Subsequent work by Lane-Fall, et al., expands on these findings in an ongoing larger hybrid effectiveness-implementation trial in 12 ICUs across 5 academic medical centers (HATRICC-US).75 This is the only large multicenter prospective study to date examining the implementation and effectiveness of a standardized OR-to-ICU handoff process. The primary implementation outcome of this study will be fidelity to the handoff protocol and the primary effectiveness outcome will be a composite measure of new-onset multiorgan failure. Secondary outcomes will include information omissions and adverse events.
The ongoing HATRICC-US study aims to address many of the gaps in existing literature regarding the OR-to-ICU handoff process, although several remain. Within this growing body of literature, established consensus is limited regarding the most effective way to standardize the OR-to-ICU information exchange and the content that is most important to include in a structured handoff tool.83 Ascertaining which elements to include in a handoff tool is vital to preventing cognitive overload, reducing the risk of communication failure,83 and mitigating diagnostic error.
Many of the proposed tools prioritize specific operative details over anticipatory guidance and tend to focus on important details such as blood products given, allergy list, and neuromuscular blockade given. However, they sometimes have less information synthesizing the patient's clinical course to date, proposed diagnosis, or anticipated clinical trajectory.
Most tools were developed by experts and adapted to suit the institution in which the handoff process was studied. No studies used a design method that explicitly focused on the mitigation of cognitive bias. Further work should focus on the development and standardization of handoff tools designed to address the specific cognitive biases that lead to diagnostic error.
Another area in need of more study is how to sustain implementation of an OR-to-ICU handoff tool. Critical care settings are underrepresented in implementation science literature,84 due in part to time pressure inherent in caring for critically ill patients.75 Yet it is that time pressure that makes a sustainable and standardized handoff tool even more necessary.
Zjwadecz, et al., found that only 23 percent of reviewed articles mentioned sustainability of an OR-to-ICU handoff intervention and none mentioned such interventions in the context of outcome measures.85 In practice, uptake of these handoff strategies is limited and no studies besides the HATRICC-US trial have examined how to sustain implementation on a large scale. Hopefully, the ongoing HATRICC-US trial will provide some insight into effective protocols for and barriers to implementation and will establish the groundwork for future implementation efforts in different settings (community, specialty ICU).
It is challenging to explicitly evaluate diagnostic error as an outcome in the postoperative period: for example, was the patient's postoperative shock hemorrhagic due to high estimated blood loss, or was it actually cardiogenic due to aggressive fluid and blood product resuscitation? Although medical error and information omission are common outcomes in many studies, none isolate the effect of handoff standardization on diagnostic error in the postoperative period.
To prevent a patient from deteriorating after surgery, one must have accurate and complete information to reduce the risk of diagnostic error during rapid clinical change. The transmission of this information occurs during the OR-to-ICU handoff. Thus, an evaluation of diagnostic safety is inherent within the composite outcome measure of new-onset multiorgan failure proposed by HATRICC-US, although it is difficult to isolate.
Further research could be directed at understanding the best way to evaluate diagnostic safety during the OR-to-ICU transition. This research could more explicitly explore how the HATRICC bundle could be used to reduce diagnostic errors. In establishing a standardized outcome measure, the evaluation and comparison of the effect of different handoff tools and protocols on diagnostic error will become more meaningful.
In conclusion, the OR-to-ICU transition is inherently high risk, and the growing body of evidence on this topic highlights the need for sustained implementation of standardized handoff tools to mitigate the impact of cognitive biases and reduce the risk of diagnostic error in the OR-to-ICU transition.
