Hospira Research Grant

Patient Handoffs: The Impact of a Fresh Perspective on Patient Mortality in Critical Care Settings

Emily Patterson, PhD, The Ohio State University Medical Center

Patient handoffs have been repeatedly identified as critically important in ensuring patient safety. When limits to resident physician duty hours were implemented to reduce fatigue, many were concerned that the increased frequency of patient handoffs would increase adverse events due to poor information transfer to the incoming caregiver. Similarly, employing different physicians during the week than the weekend (to make it easier to recruit and retain physicians) was anticipated to cause patient harm, particularly in critical care settings. Despite the conventional wisdom, in both situations, improving the quality of work life for physicians was not associated with increased risk-adjusted patient mortality in multi-site trials.

The most likely explanation for these non-intuitive findings is that positive benefits of conducting additional patient handoffs compensate for the negative effects of transferring responsibility for care. Specifically, the incoming provider brings a fresh perspective that enables catching erroneous assumptions, thereby improving critical thinking. Although there is a statement in nearly every publication on patient handoffs that they can be a point of resilience as well as a point of vulnerability, surprisingly little research on the strategies used to improve patient safety during handoffs has been conducted. The investigators’ recent findings about the benefits of providing coverage with a different intensivist physician during the week and on the weekend suggest that there is an exciting opportunity to discover patient handoff strategies that may reduce patient mortality. This is a necessary first step in order to disseminate best practices that mitigate patient safety risks during transfers of responsibility for care.

Specifically, the project will explore three research questions using targeted ethnographic observations in two ICUs:

1. Do diagnoses change following a patient handoff by attending physicians, resident physicians, nurse practitioners, or nurses in critical care settings?
2. What strategies are employed by critical care providers before, during, and after patient handoffs to increase patient safety?
3. Does change of leadership affect the willingness for an interdisciplinary team member to bring up a patient safety concern during rounds?

A deliverable at the completion of this two-year study is training materials on patient handoffs. This training will include collaborative cross-checking strategies that may improve patient safety and measurement approaches to assess their effectiveness.

The significance of this project also lies in bringing a human factors/cognitive system engineering perspective to the multi-faceted and pervasive challenge of designing, training, and regulating handoffs to reliably protect patients from harm.

This grant is funded by Hospira, the global specialty pharmaceutical and medication delivery company that provides solutions to help improve the safety, cost, and productivity of patient care.

NPSF Board Research Grant

Rapid Response Team Activations as a Burst-like Phenomena: Understanding the Role of Care Team Structure and Designing Solutions

James Gray, MD, Beth Israel Deaconess Medical Center

Rapid response teams (RRT) can improve the care of hospitalized patients by ensuring that appropriate providers, resources, and triaging are immediately available to patients experiencing a deterioration. The investigators’ work with rapid response systems has shown that rapid response team activations are not distributed across time in a manner that would be predicted by usual statistical models. Instead they exhibit a significant degree of clustering or “burstiness”: The presence of a RRT event in one patient appears to influence or predict the risk of future deterioration in other patients. This means that, to some degree, the risk of decompensation is “transmissible” between patients. The investigators will use network analysis to understand this field effect phenomenon by examining the patient, care-team and resource-based correlates of this clustering.

Based on previous studies of rapid response team activations at their institution, the investigators hypothesize that an individual patient’s risk of clinical deterioration and subsequent need for RRT activation is a function not only of patient level characteristics, but also the complex set of connections that exist between patients, their care and the teams that provide this care. The investigators will use network analytic approaches to perform a rigorous examination of the role that this interconnectedness plays in the genesis of the observed. This work will enable future improvements in the structuring of inpatient teams, rapid response systems, and their responses to patients’ needs.

The investigators will draw upon an existing dataset of over 13,000 RRT activations to understand how events in one patient influence the risk for subsequent events in others. Aims will be

1. Create a very large dataset (>100,000 admissions) that includes both patient-level and care team characteristics. The investigators have previously built a very large, patient-level RRT activation dataset. They will now apply network analysis methods developed by their group to this existing dataset. These methods use data stored in the electronic health record to ascertain the care and social networks involved in each patient’s care. The investigators will process the resulting network dataset to identify important characteristics that quantify the size, interconnectedness and assembly of relevant patient-patient and patient-provider networks.
2. Quantify the contribution of these networks to patient risk. The investigators will perform multivariate modeling to examine patient risk over time as a function of patient and care team characteristics.
3. Identify opportunities for improvement. The investigators will use resulting models to understand how modifications in rapid response team structure, patient staffing assignments and team based surge capacity can mitigate the risks seen

The NPSF Board Research Grant is supported in part by generous contributions from NPSF Board members.