Programs seeking to implement advances in health care epidemiology must critically evaluate their impact on infection prevention processes, patient safety, and cost prior to wholesale adoption.
Infection prevention (IP) practices must evolve with advancing technology. To follow the dictum “first, do no harm,” hospital leaders must remain vigilant in their prioritization of patient safety and seek new technologies in order to align with the Hippocratic oath. Some recent developments in IP technologies include electronic hand hygiene monitoring systems, antimicrobial textiles, ultraviolet C (UV-C) devices, and the optimization of electronic medical records (EMRs) to include decision-support tools and predictive analytics using machine learning to prevent health care–associated infections (HAIs).
Electronic hand hygiene monitoring systems (EHHMSs) offer hospitals large data sets of hand hygiene (HH) compliance among health care workers (HCWs) who wear their badges. Though the World Health Organization recommends direct observation of hand hygiene practices,1 and though that remains the gold standard,2 it is labor intensive, can be biased by the Hawthorne effect, and yields a much smaller denominator than EHHMSs provide. Limitations are inherent in all HH monitoring programs; however, the volume of data that EHHMs offers is of value to many IP programs. In addition to giving IP programs larger data sets of HH compliance, studies show that EHHMSs can increase HH compliance.3,4
Despite the benefits of EHHMSs, barriers exist to widespread adoptability, including negative perceptions of the systems among HCWs,5,6 variability among the different types of systems,7 and an absence of studies demonstrating improvement in HAI outcomes with EHHMSs.8 Though HCWs routinely recognize the importance of HH and are generally positive about interventions to help improve compliance, Kelly et al found many HCWs worry about the accuracy of the data6 and may feel anxiety about the systems.5 In 2021 Knudsen et al found a statistically significant decrease in health care–associated bloodstream infections with improved HH in their small study,7 and Kovacs-Litman et al reported that hospital outbreaks occurred more often in units with lower HH compliance.9 These studies are encouraging and may lead to greater adoptability, particularly if future studies can reproduce similar results.
Despite mixed results in published studies, antimicrobial textiles (ATs) are
a promising technology for infection prevention. ATs uniquely decrease surface bioburden.10 Although little debate exists on the antimicrobial properties of copper and silver, the HAI prevention impact of impregnating fabrics or surfaces with these metals remains poorly defined.
Fan et al conducted a systematic review and meta-analysis in 2020 of 6 studies and did not find conclusive evidence that copper-impregnated linens decreased HAIs.11 Albarqouni et al conducted an additional systematic review and meta-analysis of 7 studies in 2020 and found the same lack of evidence.12 On a more promising note, Hesselvig et al demonstrated that iodine-impregnated incision drapes did reduce the contamination rate during primary knee arthroplasty compared with using no drape.13 This study does not answer the question of whether intraoperative contamination leads to a subsequent infection and, to date, no studies show a correlation. Madden et al found no decrease in HAIs following the introduction of copper-infused linens.10 However, Butler detected a statistically significant decrease in hospital-onset Clostridioides difficile infections in 2018 among patients in 6 different hospitals.14 The evidence is undeniably mixed.
An important consideration in the adoption of antimicrobial linens is wastewater management. These linens are more expensive than traditional hospital linens, so investigators must continue to study whether the savings hospitals may realize through reduction of HAIs through antimicrobial linens is worth the cost of purchasing them.15
Much of the recent literature regarding UV-C robots discusses UV-C disinfection and SARS-CoV-2. In January 2022, Viana Martins et al reviewed 60 recent studies and found that UV-C devices inactivated airborne SARS-CoV-2, an important finding in the current fight against the COVID-19 pandemic.16 Although current rates of SARS-CoV-2 infections remain elevated as of this writing, focusing on preventing COVID-19 infections is crucial. Nonetheless, UV-C disinfection remains an important part of preventing many other infections in health care settings. The seminal study, the Benefits of Enhanced Terminal Room (BETR) Disinfection Study—a large, randomized 2016 control trial—found that “patients admitted to rooms previously occupied by patients harboring a multidrug-resistant organism or C difficile were [10% to 30%] less likely to acquire the same organism if the room was terminally disinfected using an enhanced strategy,” with the largest risk reduction occurring when a UV-C device was a part of the disinfection strategy.17 This study demonstrated that UV-C light could be used as part of the standard of care for disinfection in rooms where discharged patients have multidrug-resistant organism infections or C difficile infections, leading to a change in many health care facilities’ disinfection practices. In 2021, Rock et al found that UV-C disinfection did not decrease vancomycin-resistant enterococcus or C difficile infections in immunocompromised patients.18 Although this does not disprove the BETR study’s findings, Rock’s findings suggest that more research about the efficacy of UV-C disinfection in special populations is warranted.
Although UV-C devices can play a critical role in infection prevention programs, for them to provide optimum patient outcomes, health care facilities need a clear strategy for when the devices should be deployed. Like that of EHHMSs, the upfront cost of UV-C devices is significant, so a clear deployment plan helps to ensure the devices are used with consistency.
As EMRs continue to become tools not just for clinician documentation, but also as data collection tools, harnessing the power of EMRs to prevent HAIs becomes increasingly important. Two innovative ways EMRs are being used in this capacity are decision-support tools, helping guide clinicians to make decisions to prevent HAIs; and predictive modeling via machine learning, aiding clinicians and infection programs in targeting patients at greatest risk of developing an HAI.
Decision-support tools are electronic nudges or hard stops that either direct clinicians away from choosing an intervention that may increase the risk of an HAI or disallow a clinician to make that choice at all (typically with an override from a hospital epidemiologist or other hospital leader). Recent studies continue to affirm that electronic nudges improve patient safety by decreasing inappropriate C difficile orders,19 increasing appropriate C difficile prescriptions,20 and decreasing inappropriate antibiotic prescribing.21 However, decision-support tools are most likely to be effective when they feel easy to use, useful for decision-making, and reliable.22
Preventing infections in hospitalized patients via predictive modeling using machine learning (ML) is still a novel field, but teams of investigators, clinicians, and information technologists are working to create tools that can be broadly adapted to help identify patients who are at greatest risk of developing an HAI. Multiple recent studies demonstrate promising results of a variety of predictive tools to assess who is at greatest risk of a central line–associated bloodstream infection,23 ventilator-associated pneumonia (VAP),24,25 C difficile infections,26 and catheter-associated urinary tract infections.27,28 Two of these studies found that the models they created outperformed more traditional regression models.26,28 However, Frondelius et al found fewer positive results in their systematic review of the literature looking at predictive modeling tools and their ability to prevent VAP.29 A limitation to these tools being used regularly and as part of standard infection prevention practice is their newness. More testing and validation must be done before ML can be used to predict at-risk patients consistently and reliably, especially on a large scale. What works at 1 facility must be further validated for other facilities to use the same tool. Extracting correct data from the EMR remains an ongoing challenge at many health care facilities.
Advances in infection prevention generally increase patient safety and decrease hospital costs. The role of infection prevention technologies is evolving. Future studies are required to explore the integration of infection prevention technologies into everyday practice. These investigations should focus on the validation of current technologies, particularly focusing on patient-centered outcomes such as HAI risk reduction. In addition, cost-benefit analyses are required for a better understanding of the potential impact of these technologies on the patient safety mission.
Rachel Pryor, RN, MPH, is a nurse specializing in health care data analytics and research for the Health Care Infection Prevention Program at Virginia Commonwealth University Medical Center.
Gonzalo Bearman, MD, MPH, is an epidemiologist, chief of infectious diseases at Virginia Commonwealth University, and editor in chief of Antimicrobial Stewardship & Healthcare Epidemiology (ASHE) from the Society for Healthcare Epidemiology of America.