Classics and Emerging Classics

Culture has been described as a key to establishing high reliability organizations. The National Quality Forum's Safe Practices for Healthcare and the Leapfrog Group both mandate hospitals to regularly assess their safety culture. This AHRQ Web site provides validated safety culture survey tools (Hospital, Medical Office, Nursing Home, Community Pharmacy, Ambulatory Surgery Center), user guides health care organizations can use to implement the surveys and a bibliography of articles discussing the use of SOPS in the field. Organizations can also use the AHRQ database to compare their Surveys on Patient Safety Culture™ (SOPS®) results. In addition, reports are available that summarize the benchmarking data across cohorts nationwide. An AHRQ WebM&M perspective discussed how to establish a safety culture. The 2024 Nursing Home Survey data collection period is open October 1-21, 2024.

The AHRQ Patient Safety Indicators (PSIs) represent quality measures that make use of a hospital's available administrative data. The PSIs reflect the quality of inpatient care but also focus on preventable complications and iatrogenic events. Investigators have found PSIs to be a useful tool for understanding adverse events and identifying possible areas of improvement within health care delivery systems. Although relying on administrative data has clear limitations, select PSIs have been shown to accurately identify certain accidental inpatient injuries. The AHRQ Web site offers publicly available comparative data, along with resources and tools. Patient safety measurement methods are discussed in an AHRQ WebM&M perspective. Originally released in 2005, the PSI were most recently updated in August 2023.

Drawing on information gathered from the ISMP Medication Errors Reporting Program, this fact sheet provides a comprehensive list of commonly confused medication names, including look-alike and sound-alike name pairs. Drug name confusion can easily lead to medication errors, and the ISMP has recommended interventions such as the use of tall man lettering in order to prevent such errors. An error due to sound-alike medications is discussed in this AHRQ WebM&M commentary.

Effective teamwork plays an essential role in providing safe patient care. The Team Strategies and Tools to Enhance Performance and Patient Safety (TeamSTEPPS) program was developed inititally in collaboration by the United States Department of Defense and AHRQ in order to support effective communication and teamwork in health care. The 3.0 version of the widely implemented program is organized around 5 key strategies: patient focus, integrated platform, modular course design, active adult learning and emergent team challenges and opportunities. It provides new tools to measure its impact, supports increased emphasis on the role of patients in teams, and includes a new pocket guide. A PSNet WebM&M commentary discussed how improved teamwork and shared decision-making might have prevented a missed diagnosis of sepsis that lead to the death of a patient.

Health literacy is a social determinant of health and can affect the ways people understand and interact with the health system. This review describes categories of health literacy, how it functions as a social determinant of health, and interventions to improve health literacy at system, community, and individual levels.

Unconscious assumptions and implicit biases can compromise both clinician decision making and patient outcomes. This article describes two experimental studies exploring the impact of gender biases on pain estimation and treatment recommendations. After controlling for self-reported pain, the first study found that female patients’ pain was under-estimated compared to male patients. The second study replicated these findings and also found that pain-related gender stereotypes (specifically about typical willingness to express pain between females vs males) predicted pain estimation biases and that female patients were judged to benefit more from psychotherapy, whereas male patients were judged to benefit more from pain medicine.

Anticoagulants are common medications that carry the potential for serious harm if administered incorrectly. This retrospective review of 8,576 patients with atrial fibrillation who received direct oral anticoagulants identified inappropriate dosing in nearly 15% of cases, with most patients receiving an inappropriately low dose. Over one year of follow-up, the authors did not identify any significant difference in the incidence of stroke, embolism, bleeding, or ischemic attacks between patients who were inappropriately, versus appropriately, dosed.

Artificial intelligence (AI) has the potential to enhance safety and improve diagnosis, but its use is not without risks and challenges. This article discusses the conceptual, technical, and humanistic challenges with AI in health care and how AI developers, validators, and operational staff can help overcome these challenges.

Clinical decision support (CDS) systems are intended to support diagnosis and therapeutic processes of care. This position paper defines adaptive CDS as “systems that can learn and change performance over time, incorporate new clinical evidence, data types, data sources, and methods for interpreting data.” Recommendations for the effective management and monitoring of adaptive CDS are outlined.

Provider implicit bias can impact patient safety through clinical misdiagnosis, pain management, and poor patient outcomes. This literature review sought to define implicit bias and identify the impact on clinical practice and research. The authors found that no effective debiasing strategies seem to currently exist. A December 2020 WebM&M commentary discusses how implicit bias can contribute to poor communication between healthcare teams.

Patients taking oral anti-cancer drugs may experience severe side effects and medication errors. In this randomized controlled study, patients taking oral chemotherapy drugs were randomized to receive usual care (control) or additional intensive pharmacological/pharmaceutical care (intervention). Patients in the intervention group reported considerably fewer medication errors and side effects and increased treatment satisfaction.

This study explored the role of machine-learning based clinical decision support (CDS) algorithms to support (rather than replace) human decision-making and the impact on diagnostic performance. This systematic review of 37 studies found limited evidence that the use of machine learning-based CDS systems contributes to improved diagnostic performance among clinicians. Interobserver agreement, user feedback, and clinician override were the most commonly reported outcomes. The authors emphasize the importance of further evaluation of human-computer interaction.

Racial bias and racism are increasingly seen as a critical patient safety issue. In this article, the authors outline the components of systemic racism (structural/institutional, cultural, and interpersonal), how they manifest and affect the long-term care system, and the detrimental impact of systemic racism on Blacks during the COVID-19 pandemic.

This systematic review sought to determine the impact of automated dispensing cabinets (ADCs), barcode medication administration (BCMA), and closed-loop electronic medication management systems (EMMS) used by hospitals in reducing controlled substance medication errors in hospitals. Overall, only 4 studies (out of 16) focused directly on controlled medications. A variety of types of errors (e.g., log-in, data, entry, override) compromised patient safety. High-quality targeted research is urgently needed to evaluate the risks and benefits of medication-related technology.

Unanticipated adverse events harm not only patients, but also have the potential to cause psychological harm to the healthcare providers involved in the incident. This study investigated how Maryland hospitals currently support “second victims.” Even though all study participants agreed that organizations should offer support programs to second victims, they stated that several barriers exist, including stigma. Future research efforts should involve second victims themselves in order to identify barriers and facilitators, such as safety culture, to the use of organization support programs.

Incident reporting plays an essential role in identifying patient safety threats. This study aggregated findings from qualitative studies to identify barriers to incident reporting among nurses. Fear of negative consequences was the most common barrier; other barriers included inadequate reporting systems, lack of interdisciplinary and interdepartmental cooperation, lack of necessary training, and blame culture.

Artificial Intelligence (AI) is used across healthcare settings to address a variety of patient safety targets. This scoping review evaluated the potential of AI to improve patient safety across eight domains including adverse drug events, decompensation, and diagnostic errors. Both traditional (e.g. EHR) and novel (e.g. wearables) data sources can be used to develop models and interventions to improve patient safety.

High reliability remains an elusive goal for health care organizations. The authors of this study posit that medical teams’ ABCs – attitudes, behaviors, and cognitions – are critical for high-reliability, enhancing team adaptation and increasing patient safety. The article outlines practical tools and educational strategies that can be leveraged by a variety of healthcare organizations to improve team-based care.

High-risk medications have the potential to cause serious patient harm if not administered correctly. In this randomized trial, a pharmacist-directed intervention (including in-home assessment by a clinical pharmacist, communication with the primary care team, and telephone follow-up) did not result in a lower rate of adverse drug events or medication errors involving high-risk drug classes during the posthospitalization period.

A culture of safety is a key component to successful, sustainable patient safety programs. The authors review existing models of safety culture and propose a framework which synthesizes information across fragmented concepts – including organizational culture, social identity, and social learning – to illustrate the dynamic nature and drivers of safety culture.