In this column, William F. Moroney reports on human factors/ergonomics issues that have appeared in the press. Your contributions are invited. Please send electronic copy with references to wmoroney1@udayton.edu.
Top 10 Patient Safety Concerns 2026, ECRI. https://bit.ly/4tKaO69. This report identifies the major concerns for patient safety. It is “…. informed by evidence-based research, event data, and expert insights from both ECRI and the Institute for Safe Medication Practices. Beyond identifying key concerns, we also offer guidance and resources to help organizations address them, supporting a systems approach to safety that strengthens clinical care and enhances patient outcomes.” This year the list describes the following:
(1) Navigating the AI Diagnostic Dilemma.
(2) Reduced Access to Rural Healthcare Increases Health Risks and Disparities.
(3) Increasing Rates of Preventable Acute Diseases in Communities and Healthcare Settings.
(4) Effects of Federal Funding Cuts on Healthcare Operations and Patient Safety.
(5) Lack of Recognition and Reporting of Harm Events.
(6) Structural and Systemic Barriers Inhibit Equitable Pain Management for Women.
(7) Persistent Workforce Shortages Continue to Burden Staff and Restrict Access to Care.
(8) The Impact on System Improvement When a Culture of Blame Hinders Learning.
(9) Emergency Department Boarding Contributes to Worse Patient Outcomes.
(10) Persistent Gaps in Manufacturer Packaging and Labeling Design Continue to Undermine Medication Safety Efforts.
The report also provides recommendations, resources, and references.
Can AI detect usability problems like researchers? Measuring U. By Jim Lewis, Jeff Sauro, Will Schiavone, and Lucas Plabst. May 5, 2026. https://bit.ly/4v1CfJL. Four experienced usability evaluators and two AI programs (ChatGPT and Gemini) evaluated a video of an individual making a reservation at a restaurant after ascertaining that the restaurant met the specified criteria. Humans identified nine problems while ChatGPT identified five and Gemini identified four. Humans had higher within-group reliability than the LLMs. Any-2 agreement among human evaluators was 66%, ChatGPT had 31% agreement, while Gemini had 57%. Agreement between humans and AI was low. AI identified roughly half the problems humans found. AI generated more new problems than humans did. AI-only problems created a new validation burden. “Someone has to determine which AI-generated problems are real, and that means a human reviewing each one. If AI is being used to save time, the volume of unverified AI-generated problems may offset much of those savings. Whether the tradeoff is worth it likely depends on how many of those problems” are real. Also See: https://bit.ly/3Rl16db.
Does flying only get safer after something almost goes wrong? USA Today. May 6, 2026. By Zach Wichter. https://bit.ly/4tQjMPq. Recent fatal accidents involving commercial aircraft (Air Canada aircraft vs fire truck at La Guardia [LGA], the Army helicopter incident at National Airport, and close calls at Kennedy airport—two Southwest Airlines within 500 vertical feet of each other) increased the visibility of aviation safety. “Aviation experts say rising near-misses highlight both system strengths and risks in a busier airspace.” Mary Schiavo, former Department of Transportation Inspector General, said: “To wait until there are deaths when you see, all these bad statistics is just ridiculous.” Specifically, the fire truck which crossed the active-duty runway at LGA did not have, and was not required to have, a transponder that would have shown its location to the controllers. Once again, technological solutions are available, but their use has not been mandated. On a more positive note, the FAA has been using AI to analyze flight safety data for potential hazards and created an integrated safety management office. Also, the overdue new air traffic control system is expected to anticipate potential hazards. The NTSB continues to encourage the FAA and regulators to incorporate NTSB recommendations.
Systems that let drivers take their hands off the wheel don’t improve safety, NTSB head says, AP. By Josh Funk, March 31, 2026. https://bit.ly/49djnPF. The systems that allow drivers to take their hands off the wheel are convenient but don’t improve safety because people are often too reliant on them and end up paying more attention to their cellphones and infotainment screens than to the road, the head of the National Transportation Safety Board said Tuesday. “These systems function primarily as convenience features rather than safety enhancements.” FYI: Tesla is planning on release a no-wheel-no-pedal Cybercab later this year (https://bit.ly/49sTnjm).
ADDITIONAL LINKS
Makeshift signage, TED https://bit.ly/43m4uai.
$40 billion operator error https://bit.ly/4umX4PT.
Why are most humans right-handed ? https://bit.ly/3RgF638.
NASA’s long term isolation plan: https://bit.ly/4v5ecty.
LLM and health care: Review https://bit.ly/4uzt3Na.
Technology and parasports. Best paper award at CHI. https://bit.ly/4dr4XxP.
Anthropomorphic AI terms create gaps in accountability https://bit.ly/4dZmUnh.
AI hallucinating medical issues https://bit.ly/4nS3AMk.
Using AI to improve design critiques https://bit.ly/4wLlILQ.
Improving Army maintenance by using HF https://bit.ly/4wIayHw.
Pharmacists error rate increases https://bit.ly/4vBUCFJ and https://bit.ly/3PywO6a.
AI turns the impossible into the complex https://bit.ly/49iVbvg.
AI will not replace humans in healthcare https://bit.ly/4nGGjN6.
Understanding trust in AI https://bit.ly/4wFhftX.
Using soft robotic exoskeleton for gait training with Parkinson patients https://bit.ly/4e3t2Kn.
Hybrid collision avoidance: New tech to cut collision frequency https://bit.ly/4tNb2cH.
Acceptance and exoskeleton set up and take down complexity https://bit.ly/4dn7FV6.
