How Infection Prevention and Control Uses Artificial Intelligence

Over the past several years, discussion of artificial intelligence (AI) has intensified, particularly regarding its potential to influence daily work across industries. Although the current enthusiasm around AI mirrors the excitement of the early dot-com era, the circumstances are fundamentally different. Unlike past technological hype cycles, AI is already embedded within core systems that drive health care operations and even our daily lives. Autocorrect on our smartphones or predictive text in email are examples of early AI. Rather than speculating about whether AI will arrive, the health care sector is now focused on integrating AI responsibly and effectively into existing workflows.

Experts continue to debate the long-term implications of AI, including whether it will fundamentally replace specific job roles. What most experts agree on, however, is that AI will primarily function as an augmentation tool, enhancing human performance rather than eliminating the need for human expertise and touch.¹ This distinction is crucial in infection prevention, a discipline that is heavily data-driven and dependent on contextual clinical judgment and critical thinking.

Infection preventionists (IPs) routinely manage large volumes of data, including those concerning laboratory results, clinical documentation, device utilization, antimicrobial medication use, and patient movement. Historically, much of this work has required manual chart review and interpretation. Over the past decade, data mining and surveillance tools such as TheraDoc, BD Insights, Epic Bugsy, and Cerner have been integrated as embedded modules within electronic medical record (EMR) systems to streamline this process. Although these systems significantly improved efficiency, they primarily relied on predefined rules and static logic.²

AI is transforming hand hygiene adherence from a largely manual, observational process into an automated, data-driven one. AI-enabled systems often use computer vision and machine learning to continuously monitor hand hygiene stations, detect handwashing or sanitizer use with high accuracy, and generate real-time insights that surpass those obtained through direct observation. For instance, depth sensors paired with AI algorithms have been shown to detect dispenser use with sensitivity and specificity comparable to that of human auditors, while also enabling continuous monitoring across an entire health care unit. These technologies not only enhance adherence tracking but also provide actionable analytics to target areas of suboptimal performance, ultimately strengthening infection control practices in clinical settings.³

Advances in AI agents and machine learning have transformed surveillance capabilities. Unlike conventional rule-based systems, AI-enabled tools can process data from multiple sources simultaneously, uncover subtle relationships, and continuously improve as they receive user feedback. Emerging evidence suggests that these systems can accurately identify health care-associated infections (HAIs), such as central line-associated bloodstream infections, catheter-associated urinary tract infections, and surgical site infections, using National Healthcare Safety Network–aligned criteria, often in a fraction of the time required for manual review.⁴

Data from recent studies demonstrate that generative AI and machine learning models can effectively sift through EMRs to flag potential HAIs, significantly reducing the surveillance burden on IPs. In some proof-of-concept evaluations, AI systems replicated infection determinations with high concordance with expert reviewers, highlighting their potential as decision-support tools rather than autonomous decision makers. These technologies allow AI agents to work continuously in the background, monitoring patient data in real time and prompting IPs only when meaningful signals are detected.

Looking ahead, AI integration within EMRs is expected to become more sophisticated. Future systems may not only identify infections retrospectively, but also predict infection risk before clinical deterioration occurs. Predictive AI models have already been explored for anticipating bloodstream infections, multidrug-resistant organism acquisition, and sepsis. Importantly, newer approaches emphasize explainable AI, ensuring that IPs can understand why a model generated a particular alert, which is essential for trust, validation, and regulatory compliance.⁵

Beyond surveillance and prediction, AI is also being applied to improve adherence to infection prevention practices. Computer vision technologies have been studied for monitoring hand hygiene and adherence to personal protective equipment. At the same time, large language models (LLMs) are being evaluated for their ability to support infection control inquiries and education.⁶ Early comparisons suggest that, when used cautiously, LLMs may assist with rapid access to evidence-based guidance, though human oversight remains critical.⁷

Despite these advancements, AI is not positioned to replace IPs. Regulatory requirements, including the Centers for Medicare & Medicaid Services Conditions of Participation, continue to mandate the presence of qualified IPs within health care facilities.⁸ More importantly, many core IP functions, such as outbreak investigation, interdisciplinary communication, staff education, leadership rounding, and reporting to public health authorities, require clinical insight, situational awareness, and professional judgment that AI cannot replicate.

Instead, the IP’s role is likely to evolve. As AI systems assume more of the routine surveillance and data-processing workload, IPs will be able to dedicate greater time to prevention strategies, system redesign, education, and high-impact interventions. The relationship between AI and infection prevention should be viewed as collaborative, with AI serving as a force multiplier rather than a replacement.

Although AI agents and machine-learning tools are still in relatively early adoption within IP, their trajectory is clear. With continued refinement, user feedback, and appropriate governance, these systems will become more accurate, more reliable, and more deeply embedded in infection prevention programs. Despite these advances, adoption of AI in infection prevention and control has been slower than in other areas of health care. Historically, health care change and technological adoption have lagged behind those in other industries, and AI implementation is no exception. When AI tools are introduced in hospitals, priority is often given to departments with immediate financial or operational impact, such as revenue cycle management, clinical documentation improvement, and coding and billing. As a result, infection prevention programs are often among the last to adopt emerging AI technologies.⁹

This delayed adoption does not reflect a lack of value in IP, but rather the reality of organizational prioritization and resource allocation. Meaningful integration of AI into infection prevention workflows will likely occur gradually, as systems mature, regulatory expectations evolve, and health care organizations gain confidence in these tools. Over time, as AI becomes more embedded in EMRs and operational platforms, infection prevention programs will be better positioned to leverage these technologies sustainably and impactfully.

Looking ahead, AI is unlikely to replace IPs, but it is poised to reshape their roles. As AI systems assume a greater share of routine surveillance and data-processing tasks, IPs will be able to focus more intently on prevention strategies, system redesign, staff education, and high-impact interventions. The future of infection prevention should be viewed as a collaborative model in which AI serves as a force multiplier, enhancing efficiency and insight. At the same time, clinical expertise, judgment, and leadership remain firmly human-driven. This is an exciting and transformative time for the field, offering the opportunity to enhance patient safety, reduce preventable harm, and elevate the strategic role of IPs in health care.

References

1. Adler-Milstein J, Aggarwal N, Ahmed M, et al. Meeting the moment: addressing barriers and facilitating clinical adoption of artificial intelligence in medical diagnosis. NAM Perspect. 2022;2022:10.31478/202209c. doi:10.31478/202209c

2. Wiemken TL, Carrico RM. Assisting the infection preventionist: use of artificial intelligence for health care–associated infection surveillance. Am J Infect Control, 2024;52(6):625-629. doi:10.1016/j.ajic.2024.02.007

3. Singh A, Haque A, Alahi A, et al. Automatic detection of hand hygiene using computer vision technology. J Am Med Inform Assoc. 2020;27(8):1316-1320. doi:10.1093/jamia/ocaa115

4. Saad AA, Hassan A, Alali A, Alkhatib F, Tolba MF, Simsekler MCE. The role of artificial intelligence in managing central line-associated bloodstream infection (CLABSI) for patient safety and quality of care. Risk Manag Healthc Policy. 2025;18:2887-2898. doi:10.2147/RMHP.S520035

5. Gastaldi S, Tartari E, Satta G, Allegranzi B. Advancing infection prevention and control through artificial intelligence: a scoping review of applications, barriers, and a decision-support checklist. Antimicrob Steward Healthc Epidemiol. 2025;5(1):e317. doi:10.1017/ash.2025.10191

6. Yuan M, Bao P, Yuan J, et al. Large language models illuminate a progressive pathway to artificial intelligence healthcare assistant. Med Plus. 2024;1(2):100030. doi:10.1016/j.medp.2024.100030

7. Cotia ALF, Scorsato AP, da Silva Victor E, et al. Integration of an electronic hand hygiene auditing system with electronic health records using machine learning to predict hospital-acquired infection in a health care setting. Am J Infect Control. 2025;53(1):58-64. doi:10.1016/j.ajic.2024.09.012

8. Condition of Participation: Infection Prevention and Control and Antibiotic Stewardship Programs, 42 USC §482.42 (2026). Accessed February 12, 2026. https://www.ecfr.gov/current/title-42/chapter-IV/subchapter-G/part-482/subpart-C/section-482.42

9. Abdelwanis M, Simsekler MCE, Gabor AF, Sleptchenko A, Omar M. Artificial intelligence adoption challenges from healthcare providers’ perspectives: a comprehensive review and future directions. Safety Science. 2026;193:107028. doi:10.1016/j.ssci.2025.107028