Nurse union survey: AI patient safety threats prove need for strict guardrails

Editor's Note

Artificial intelligence (AI) threatens patient safety by undermining the judgment of human nurses, according to a survey by National Nurses United (NNU). “The findings, coupled with anecdotal reports from nurses at hospitals nationwide, underscore the urgent need for stricter regulation and greater input from nurses and health care staff on whether and how AI is deployed,” the nation’s largest union of registered nurses states in a May 15 press release.

Conducted from January 18 to March 4, the survey attracted responses from 2,300 registered nurses and members of NNU. Of that total, 60% disagreed with the statement: “I trust my employer will implement AI with patient safety as the first priority.”

As for the extent of implementation, 40% report employers introducing new devices, gadgets, and changes to the electronic health records (EHR) in the past year. “While they’ve been marketed as tools to free clinicians from documentation burdens and support clinicians in decision making, NNU’s survey finds otherwise,” the union reports.

For example, among the half of nurses reporting their employers use algorithms to help determine patient acuity, 69% say computer-generated assessments don’t match their own and don’t account for educational, psycho-social, or emotional needs of patients and families. Also, among the 12% of nurses reporting the use of automated handoffs, 48% say those reports miss critical information. Other AI criticism focused on the difficulty with modifying AI recommendations based on nurses’ judgment.

In related news, NNU recently announced a “Nurses’ and Patients’ Bill of Rights,” a set of guiding principles designed to ensure just and safe application of AI in healthcare settings. Meanwhile, affiliates continue to protest attempts to push AI in hospital settings, including in an April 22 San Francisco gathering of hundreds of California Nurses Association members protesting Kaiser Permanente’s International Integrated Care Experience conference.