NYSDA Publications

AHRQ Issues Regulatory Update

Jan 20, 2026

January 20, 2026 | Issue 988

In This Week's Issue: AI-driven insights into dementia-related hospital safety; virtual reality–based evaluation of clinical decision support in pediatric intensive care; automated tool to enhance medication safety; interprofessional training to improve nurse–intern collaboration.

AI Sheds Light on Dementia-Related Hospital Safety Risks

An AHRQ-funded study published in BMJ Open Quality found that machine learning tools can provide insight into the causes of patient safety events for hospitalized people living with dementia.  Researchers evaluated 1,387 dementia-related patient safety event records from a diverse 10-hospital health system from January 2018 to July 2023.  After categorizing the free-text reports by contributing factors, the team developed two machine learning models to identify whether the events were caused by situational factors, such as patient-related care challenges, or active failures, such as those caused by staff.  The models were able to identify situational factors—such as agitation, wandering, or mobility challenges—that drive most dementia-related safety events with over 70 percent accuracy, but detected just over 8 percent of errors caused by staff.  These findings suggest that AI has strong potential to inform targeted training, medication management, and systemwide interventions to enhance safety for hospitalized patients with dementia.  Read the full article.

Study Uses Virtual Reality To Assess Clinical Decision Support in Pediatric ICU

Using a novel systems approach and virtual reality observation method, an AHRQ-funded Patient Safety Learning Lab study has pinpointed key challenges in the use of clinical decision support (CDS) in the pediatric intensive care unit.  Common issues included poor communication, workflow inefficiencies, limited real-time access to clinical data, and consistent clinician disregard for existing in-room CDS tools.  Researchers mapped unit workflows and conducted qualitative interviews, then tested clinician behavior in an immersive virtual reality "digital twin" of the unit.  A subsequent failure modes and effects analysis revealed failures related to information flow, diagnostic accuracy, and situation awareness.  These findings indicate opportunities for predictive, intuitive, and seamlessly integrated CDS design, supported by systems safety and human factors principles.  Read the article in the Journal of Patient Safety.

Automated Tool Shows Potential for Enhancing Medication Safety

A new AHRQ-funded study published in BMJ Health & Care Informatics evaluated an automated tool that shows potential for detecting discrepancies between prescribed and dispensed medications.  Using the tool, the System Approach to Verifying Electronic Prescriptions (SAV E-Rx), researchers analyzed over 1.2 million e-prescription records from 14 pharmacies across 9 states.  Of the 662 cases reviewed by pharmacists, 89 percent were intended substitutions (such as dosage or form changes), while 11 percent represented unintended mismatches, primarily due to human factors and labeling issues.  Though dispensing errors were rare, unintended mismatches could pose clinical risks.  Most pharmacists favored receiving future alerts for these errors.  The findings suggest that integrating automated verification tools like SAV E-Rx can strengthen patient safety by catching medication discrepancies that may otherwise go unnoticed, highlighting opportunities for improved labeling standards, workflow optimization, and pharmacy staff training to prevent dispensing errors.  Access the study.

Shadowing Program Strengthens Nurse-Intern Collaboration and Communication

A reciprocal shadowing program between pediatric interns and nurses improved understanding, teamwork, and communication between the two professions, according to an AHRQ-funded study.  Published in the Journal of Hospital Medicine, the study paired 60 pediatric interns and 52 acute care pediatric nurses in a 4-hour shadowing experience where each observed the other’s daily responsibilities from January 2020 to June 2023.  Using pre-and post-shadowing surveys, researchers found that both groups gained clearer insight into each other's workflows and roles.  Interns more accurately described nurses' work schedules after the program, and nurses significantly improved their understanding of interns' shift routines.  More than 85 percent of interns and 90 percent of nurses reported that the experience positively influenced their teamwork and communication in daily practice.  Researchers recommended incorporating such programs early in medical and nursing training to promote lasting interprofessional understanding.  Access the study.

AHRQ Stats: High Blood Pressure Treatment by Insurance and Access

In 2021-22, 47.0 percent of nonelderly adults with diagnosed or treated high blood pressure who lacked a usual source of care had no medical treatment for their high blood pressure.  Among those who were uninsured for the entire year, 48.7 percent went without treatment.  (Source: AHRQ Medical Expenditure Panel Survey Statistical Brief #563, Recent Changes in Treatment Patterns for Diagnosed or Treated Hypertension, 2018-19 to 2021-22.)

Register for Upcoming Webinar

AHRQ in the Professional Literature

Impact of COVID-19 pandemic on quality of life, anxiety, connections to friends, and access to resources among people with HIV: using the social ecological model.  Dawson-Rose CS, Horvat Davey C, Huang E, et al. AIDS Behav. 2025 Oct 27. [Epub ahead of print.]  Access the abstract on PubMed®.

'In Texas, everybody wants antibiotics': reducing inappropriate antibiotic expectations and use with a provider-patient communication tool in primary care.  Collazo A, Wermuth P, Luna Rodriguez J, et al. Fam Med Community Health. 2025 Nov 11;13(4).  Access the abstract on PubMed®.

The association of sleep quality, sociodemographic, and disease-related factors in youth living with sickle cell disease.  Tucker TD, Padmanabhan DS, Alishlash AS, et al. Pediatr Blood Cancer. 2026 Jan;73(1):e32115. Epub 2025 Nov 2.  Access the abstract on PubMed®.

Human performance evaluation of a pediatric artificial intelligence sepsis model.  Kandaswamy S, Muthu N, Braykov N, et al. J Am Med Inform Assoc. 2025 Oct;32(10):1552-61.  Access the abstract on PubMed®.

Expanded child tax credit payments during pregnancy were associated with decreased odds of adverse birth outcomes.  Vasan A, Wood JI, Luan X, et al. Health Affairs. 2025 Oct;44(10):1298-1306.  Access the abstract on PubMed®.

Practice-level spending variation for radiation treatment episodes among older adults with cancer.  Lam MB, Landrum MB, McWilliams JM, et al. JAMA Health Forum. 2025 Jul 3;6(7):e251952.  Access the abstract on PubMed®.

An adaptive simulation intervention decreases emergency physician physiologic stress while caring for patients during COVID-19: a randomized clinical trial.  Evans LV, Bonz JW, Buck S, et al. PLoS One. 2025 Sep 3;20(9):e0331488.  Access the abstract on PubMed®.

A patient-centered outcomes framework for evaluating behavioral health crisis care.  Burns A, Meanwell E, Blackburn J, et al. Psychiatr Serv. 2025 Sep;76(9):834-45. Epub 2025 Jun 19.  Access the abstract on PubMed®.