Fall prevention remains a patient safety priority. This article describes how fall prevention strategies are being implemented and operationalized across 60 hospitals in the United States. While many hospitals employed recommended strategies identified, implementation was suboptimal at times – for example, interdisciplinary fall committees were common but rarely included physicians.

This review expands upon previous work evaluating implementation strategies for high-reliability organizations. Review findings indicate that health care system adoption of high-reliability principles is associated with improved outcomes, but the level of evidence is low. Future research should include concurrent control groups to minimize bias and focus on whether certain high-reliability frameworks, metrics, or intervention components lead to greater improvements.  

Many factors can contribute to early, unplanned readmissions among critical care patients. In this prospective cohort study, adult patients who were discharged directly home after an ICU admission were followed for 8 weeks post-discharge to explore the predictors of adverse events and unplanned return visits to a health care facility. Among 129 patients, there were 39 unplanned return visits. Researchers identified eight predictors of unplanned return visits including prior substance abuse, hepatitis, discharge diagnosis of sepsis, ICU length of stay exceeding 2 days, nursing workload, and leaving against medical advice.  

This article discusses the tension between evidence generation from observational studies and rigorous randomized controlled trials to treat COVID-19 and important safety considerations, such as use of unproven therapies.

This study found that two early warning scores – the Modified Early Warning Score (MEWS) and the National Early Warning Score (NEWS) – can predict patients at risk of deterioration and who will need to be seen by a physician overnight. The authors conclude that use of such early warning scores may be useful for improving handoffs and resource allocation for overnight care.

This article discusses the action hierarchy, which is a tool for generating corrective actions to improve safety and focuses on those recommendations relying less on human factors and more on systems change. The authors propose a multifaceted definition of ‘systems change’ and a rubric for determining the extent to which a corrective action addresses ‘systems change’ (‘systems change hierarchy’).

This systematic review explored how artificial intelligence (AI) based on machine learning algorithms and natural language processing is used to address and report patient safety outcomes. The review suggests that AI-enabled decision support systems can improve error detection, patient stratification, and drug management, but that additional evidence is needed to understand how well AI can predict safety outcomes.  

As part of a quality improvement initiative to enhance surgical onboarding, the authors used semi-structured interviews with 20 physicians to understand potential areas of risk when a physician begins working in an unfamiliar setting. Qualitative analysis found that three key findings: (1) physicians often receive little to no onboarding when starting to practice in a new setting, which can limit their ability to provide safe care; (2) physicians felt onboarding inadequately fostered strong interpersonal relationships among health care teams, which impedes psychological safety and team cohesion, and; (3) physicians noted an increased risk of patient harm during emergency situations in new settings due to lack of understanding of culture, workflow, roles/responsibilities and available equipment.

Care for older hospitalized patients can be complex due to multimorbidities and polypharmacy. This review found that restraint use and limiting mobility to prevent falls can result in functional losses in geriatric patients. The authors suggest multidisciplinary approaches to shifting the goals of nursing care to focus on preserving and restoring function in older patients.