All Content

While artificial intelligence (AI) in healthcare may potentially improve some areas of patient care, its overall safety depends, in part, on the algorithms used to train it. One hospital developed four AI models at the start of the COVID-19 pandemic to predict risks such as hospitalization or ICU admission. Researchers found inconsistent instances of model-level bias and recommend a holistic approach to search for unrecognized bias in health AI.

The COVID-19 pandemic led to a rapid transition of healthcare from in-person to remote and virtual care. This review assessed the safety and effectiveness of pulse oximetry in remote patient monitoring (RPM) of patients at home with COVID-19. Results show RPM was safe for patients in identifying risk of deterioration. However, it was not evident whether remote pulse oximetry was more effective than other virtual methods, such as virtual visits, monitoring consultations, or online or paper diaries.

The COVID-19 pandemic dramatically increased the use of telehealth, including remote primary care triage and assessment. This qualitative study explored patients’ and clinicians’ experiences managing suspected COVID-19 during remote triage and assessment. Findings highlight the need to strengthen organizational resilience, develop and implement triage decision support tools, and address vulnerable and unsupported groups (e.g., the elderly, vulnerable and homebound patients).

The COVID pandemic has increased demand and acceptance of remote care modalities. This commentary suggests that home monitoring is a promising telehealth approach and that its application could improve value while enhancing safety for hospital-at-home and other levels of home-based care patients.

COVID-19 restrictions and patient safety concerns have greatly expanded the use of telehealth and virtual visits. Through patient satisfaction surveys and patient complaints, this US hospital evaluated the quality and safety of virtual visits in March and April of 2020. Five patient complaints were submitted during this timeframe and overall patient satisfaction remained high. Safety and quality risks were identified (e.g., diagnostic error) and best practices were established.

In response to concerns about COVID-19 transmission, many ambulatory care visits have transitioned to telehealth visits. This commentary describes the impact of telehealth on diagnostic errors and medication safety in ambulatory settings. Recommendations to further understand the impact of telemedicine on patient safety include: systematically measuring patient safety outcomes and increasing reporting of safety incidents; identifying the patients and clinical scenarios with the greatest risk of unsafe telehealth care; identifying and supporting best practices to ensure equal access to safe telehealth.

When hospitals began to fill up with COVID-19 patients, new strategies had to be developed and implemented quickly to reduce the spread of the virus. This article describes one strategy implemented by a New York hospital: relocating smart intravenous (iv) infusion pumps outside of patient rooms. Challenges, facilitators, and lessons learned are discussed.

This survey of individuals living in the United Kingdom, Sweden, Italy, and Germany identified an increased use of virtual primary care services – such as telephone or video consultation, remote triage, and secure messaging systems – since the onset of the COVID-19 pandemic. Respondents reported that virtual technologies positively impacted multiple dimensions of care quality, including timeliness, safety, patient-centeredness, and equity.

The COVID-19 pandemic has exacerbated existing challenges associated with diagnostic error. This study used natural language processing to identify and categorize diagnostic errors occurring during the pandemic. The study compared a review of all patient safety reports explicitly mentioning COVID-19, and using natural language processing, identified additional safety reports involving COVID-19 diagnostic errors and delays. This innovative approach may be useful for organizations wanting to identify emerging risks, including safety concerns related to COVID-19.

Telemedicine has become a more accepted care mode due to the COVID pandemic and general rural care access issues. This commentary suggests a 5-part framework for examining patient, physician, technological, clinical and health system influences on care management decisions that affect the safety of telediagnosis in primary care.

The COVID-19 pandemic has led to dramatic changes in healthcare delivery. Based on semi-structured interviews and direct observations, researchers evaluated the impact of a new model of remote inpatient care using telemedicine technologies in response to the pandemic. Intensive care and internal medicine units were divided into contaminated and clean zones and an integrated control room with audio-visual technologies allowed for remote supervision, communication, and support. The authors conclude that this model can increase flexibility in staffing via remote consultations and allow staff to supervise and monitor more patients without compromising patient and staff safety.

The COVID-19 pandemic led to an increase in the use of telehealth in order to limit patient exposure to the virus. Findings from this study highlight the value of telehealth visits for patients with suspected respiratory infections to prevent further transmission. Researchers found that patients exposed to influenza-like illness in primary care office settings were more likely than nonexposed patients to return with a similar illness within two weeks.

This commentary outlines six benefits of the new requirement that health systems offer online patient access to their medical records, including clinician notes. Benefits include strengthened patient-clinician relationships, patient engagement, and adherence to mental health care plans. While online access may help narrow the digital divide, the authors also point out that lack of access to technology is still a barrier.

Artificial intelligence (AI) systems can support diagnostic decision-making. This study evaluates diagnostic “shortcuts” learned by AI systems in detecting COVID-19 in chest radiographs. Results reveal a need for better training data, improved choice in the prediction task, and external validation of the AI system prior to dissemination and implementations in different hospitals.