Clinical and Cost-Effectiveness of an AI-Synthesized Nursing Care Bundle for Preventing Medical Device-Related Pressure Injuries

The Clinical and Cost-Effectiveness of an Artificial Intelligence-Synthesized Nursing Care Bundle in Preventing Medical Device-Related Pressure Injuries

Medical device-related pressure injuries (MDRPIs) are an important and largely preventable patient safety problem, particularly in intensive care units where the use of medical devices is frequent and prolonged. A recent meta-analysis reported an MDRPI incidence of approximately 19% among adults in critical care settings (Zhang et al., 2024). Although evidence-based recommendations for MDRPI prevention are available, their consistent and sustainable integration into routine nursing care remains challenging. Care bundles may facilitate the translation of evidence into routine clinical practice; however, existing MDRPI prevention bundles vary considerably in their components and implementation strategies, and evidence regarding their clinical and economic effectiveness remains limited (Neill & Martin, 2024).

This study aims to develop an artificial intelligence (AI)-synthesized nursing care bundle for the prevention of MDRPIs and to evaluate its clinical effectiveness and cost-effectiveness compared with routine care. The care bundle will consist of 3-5 evidence-based components identified through a systematic review of the literature and AI-assisted evidence synthesis. The draft bundle will subsequently be evaluated and finalized through a two-round Delphi process involving experts with experience in MDRPI prevention. Before clinical implementation, participating nurses will receive standardized training, and their adherence to the care bundle will be assessed by the research team, with an adherence level of ≥95% required before implementation.

The clinical effectiveness of the finalized care bundle will be evaluated in a randomized controlled trial conducted in an adult intensive care unit. A total of 220 patients will be allocated in a 1:1 ratio, using blocked stratified randomization based on MDRPI risk level, to either the intervention group receiving the AI-synthesized nursing care bundle or the control group receiving routine care. All medical devices in place at enrollment will be recorded, and participants will be followed for up to 14 days; the primary outcome will be the occurrence of an MDRPI related to these devices during the follow-up period. An economic evaluation will be conducted from the hospital perspective based on product utilization, nursing labor time, and hospitalization-related resource use, and the cost-effectiveness of the intervention will be assessed using the incremental cost-effectiveness ratio.

Inclusion Criteria:

• Adults aged 18 years or older
• Presence of at least one medical device placed for diagnostic or therapeutic purposes.
• No existing medical device-related pressure injury (MDRPI) at the time of study enrollment.
• Expected length of stay in the intensive care unit of at least 24 hours.
• Willingness to participate in the study and provision of written informed consent by the participant or their legal representative.

Exclusion Criteria:

• Presence of only a peripheral intravenous (IV) catheter as a medical device. (Peripheral IV catheters are excluded because they are routinely present in ICU patients and their replacement frequency and treatment-related use may vary rapidly and unpredictably)
• Presence of an existing MDRPI at the time of study enrollment.
• Expected intensive care unit stay of less than 24 hours.
• Refusal to participate, failure to provide informed consent, or withdrawal of consent during the study.