Healthcare-acquired pressure injuries (HAPI) can result in increased lengths of stay, hospital readmissions, and lower quality of life. This article describes the experience of one hospital which, after it discovered it had higher-than-average HAPI rates, conducted a root cause analysis to determine contributing factors and identify potential solutions. Dedicated nursing staff were hired and trained, and an electronic health record form was developed to document and track HAPI. A root cause analysis was completed for each HAPI to identify trends and implement improvements.

Insufficient or incorrect transfer of patient information, whether caused by human or organizational factors, can result in adverse events during transitions of care. This study used four years of incident reports to identify the types, causes, and consequences of health information exchange- (HIE) related patient safety incidents in emergency care, (ED) emergency medical services (EMS), or home care. The two main kinds of HIE-related incidents were (1) inadequate documentation and inadequate use of information (e.g., deficiencies in content), and (2) causes related to the health professional or organization; consequences were adverse events or additional actions to prevent, avoid, and correct adverse events.

Verbal or physical violence towards healthcare workers can result in harm of both staff and patients. Based on semi-structured interviews with 25 registered nurses working in the emergency department (ED) at one large academic health system, the authors explored nursing perspectives on how EHR-based behavioral flags – used to identify incidents of workplace violence – can promote clinician safety. Participants identified benefits of the flags as well as concerns (e.g., introduction of bias, potential damage to the patient-clinician relationship), highlighted necessary system improvements, and how related challenges in the ED (e.g., unmet mental health needs of patients, COVID-19-related burnout) can contribute to workplace violence.

Reconciling medication lists at transitions of care is a widely recognized safety strategy; however, other parts of the electronic health record (EHR) - allergies and problem lists - also need reconciliation. This article describes an academic medical system's quality improvement project to increase rates of complete reconciliation of problems, medications, and allergies in the EHR. Twenty-six cycles of Plan-Do-Study-Act increased completion rates from 20% to 80%.

Surgical never events are rare but devastating for patients. Using machine learning, this study identified 24 contributing factors to two types of surgical never events - wrong site surgery and retained items. Communication, the number and type of staff present, and the type and length of surgery were identified contributing factors.

Patients and healthcare providers rely on devices that use artificial intelligence or machine learning in diagnostics, treatment, and monitoring. This study utilizes adverse event reports submitted to the FDA's Manufacturer and Use Facility Device Experience (MAUDE) database for machine learning-enabled devices. Mammography was implicated in 69% of reports, and the majority were near-miss events.

Symptom–Disease Pair Analysis of Diagnostic Error (SPADE) is a framework to measure diagnostic errors using existing databases, such as electronic health records or administrative claims. The original developers of the SPADE framework provide additional guidance on types of comparator groups, how to select the appropriate group, and what inferences can be drawn from the analysis.

The majority of healthcare providers use electronic health record (EHR) systems but these systems are not infallible. This analysis used closed malpractice claims from the CRICO malpractice insurance database to identify whether the EHR contributes to diagnostic error, the types of errors, and where in the diagnostic process errors occur. EHR contributed to diagnostic error in 61% of claims, the majority in outpatient care, and 92% at the testing stage.

Medication administration is a complex process with many opportunities for error. Using the Event Analysis of Systemic Teamwork (EAST) model, researchers identified opportunities to improve medication administration system performance and promote patient safety. The authors discuss the networks involved in medication administration (e.g., task network, social network, information network) and how the complexities involved in each network contribute to medication administration errors.

High-reliability organizations are able to achieve safety despite organizational changes or other hazardous conditions. This article describes the implementation of a new electronic health record (EHR) system at one academic health system in Canada and provides examples of how high-reliability principles informed activities to prevent patient harm during this organizational change.

Artificial intelligence (AI) is increasingly used to identify and prevent patient safety threats. This review focuses on AI used to reduce medication errors in the primary care setting. Most studies targeted the prescribing stage, and the main category of AI was computerized decision support system.

To combat serious adverse events (SAE) and suicide among veterans with opioid use disorder (OUD), the Veterans Health Administration (VHA) implemented the Stratification Tool for Opioid Risk Mitigation (STORM) in all VHA facilities. Patients identified as high-risk for SAE by STORM received a mandatory case review. This study focuses on high-risk patients with a new OUD diagnosis. Mandatory case review increased the odds of all-cause mortality, but not SAE. Patients whose opioids were discontinued after case review showed even higher odds of mortality.