A machine learning approach to predict healthcare-associated infections at intensive care unit admission: findings from the SPIN-UTI project

Artificial intelligence and prescription of antibiotic therapy: present and future

INTRODUCTION

In the past few years, the use of artificial intelligence in healthcare has grown exponentially. Prescription of antibiotics is not exempt from its rapid diffusion, and various machine learning (ML) techniques, from logistic regression to deep neural networks and large language models, have been explored in the literature to support decisions regarding antibiotic prescription.

AREAS COVERED

In this narrative review, we discuss promises and challenges of the application of ML-based clinical decision support systems (ML-CDSSs) for antibiotic prescription. A search was conducted in PubMed up to April 2024.

EXPERT OPINION

Prescribing antibiotics is a complex process involving various dynamic phases. In each of these phases, the support of ML-CDSSs has shown the potential, and also the actual ability in some studies, to favorably impacting relevant clinical outcomes. Nonetheless, before widely exploiting this massive potential, there are still crucial challenges ahead that are being intensively investigated, pertaining to the transparency of training data, the definition of the sufficient degree of prediction explanations when predictions are obtained through black box models, and the legal and ethical framework for decision responsibility whenever an antibiotic prescription is supported by ML-CDSSs.

Assisting the infection preventionist: Use of artificial intelligence for health care-associated infection surveillance

BACKGROUND

Health care-associated infection (HAI) surveillance is vital for safety in health care settings. It helps identify infection risk factors, enhancing patient safety and quality improvement. However, HAI surveillance is complex, demanding specialized knowledge and resources. This study investigates the use of artificial intelligence (AI), particularly generative large language models, to improve HAI surveillance.

METHODS

We assessed 2 AI agents, OpenAI's chatGPT plus (GPT-4) and a Mixtral 8×7b-based local model, for their ability to identify Central Line-Associated Bloodstream Infection (CLABSI) and Catheter-Associated Urinary Tract Infection (CAUTI) from 6 National Health Care Safety Network training scenarios. The complexity of these scenarios was analyzed, and responses were matched against expert opinions.

RESULTS

Both AI models accurately identified CLABSI and CAUTI in all scenarios when given clear prompts. Challenges appeared with ambiguous prompts including Arabic numeral dates, abbreviations, and special characters, causing occasional inaccuracies in repeated tests.

DISCUSSION

The study demonstrates AI's potential in accurately identifying HAIs like CLABSI and CAUTI. Clear, specific prompts are crucial for reliable AI responses, highlighting the need for human oversight in AI-assisted HAI surveillance.

CONCLUSIONS

AI shows promise in enhancing HAI surveillance, potentially streamlining tasks, and freeing health care staff for patient-focused activities. Effective AI use requires user education and ongoing AI model refinement.

Public and Research Interest in Telemedicine From 2017 to 2022: Infodemiology Study of Google Trends Data and Bibliometric Analysis of Scientific Literature

BACKGROUND

Telemedicine offers a multitude of potential advantages, such as enhanced health care accessibility, cost reduction, and improved patient outcomes. The significance of telemedicine has been underscored by the COVID-19 pandemic, as it plays a crucial role in maintaining uninterrupted care while minimizing the risk of viral exposure. However, the adoption and implementation of telemedicine have been relatively sluggish in certain areas. Assessing the level of interest in telemedicine can provide valuable insights into areas that require enhancement.

OBJECTIVE

The aim of this study is to provide a comprehensive analysis of the level of public and research interest in telemedicine from 2017 to 2022 and also consider any potential impact of the COVID-19 pandemic.

METHODS

Google Trends data were retrieved using the search topics "telemedicine" or "e-health" to assess public interest, geographic distribution, and trends through a joinpoint regression analysis. Bibliographic data from Scopus were used to chart publications referencing the terms "telemedicine" or "eHealth" (in the title, abstract, and keywords) in terms of scientific production, key countries, and prominent keywords, as well as collaboration and co-occurrence networks.

RESULTS

Worldwide, telemedicine generated higher mean public interest (relative search volume=26.3%) compared to eHealth (relative search volume=17.6%). Interest in telemedicine remained stable until January 2020, experienced a sudden surge (monthly percent change=95.7%) peaking in April 2020, followed by a decline (monthly percent change=-22.7%) until August 2020, and then returned to stability. A similar trend was noted in the public interest regarding eHealth. Chile, Australia, Canada, and the United States had the greatest public interest in telemedicine. In these countries, moderate to strong correlations were evident between Google Trends and COVID-19 data (ie, new cases, new deaths, and hospitalized patients). Examining 19,539 original medical articles in the Scopus database unveiled a substantial rise in telemedicine-related publications, showing a total increase of 201.5% from 2017 to 2022 and an average annual growth rate of 24.7%. The most significant surge occurred between 2019 and 2020. Notably, the majority of the publications originated from a single country, with 20.8% involving international coauthorships. As the most productive country, the United States led a cluster that included Canada and Australia as well. European, Asian, and Latin American countries made up the remaining 3 clusters. The co-occurrence network categorized prevalent keywords into 2 clusters, the first cluster primarily focused on applying eHealth, mobile health (mHealth), or digital health to noncommunicable or chronic diseases; the second cluster was centered around the application of telemedicine and telehealth within the context of the COVID-19 pandemic.

CONCLUSIONS

Our analysis of search and bibliographic data over time and across regions allows us to gauge the interest in this topic, offer evidence regarding potential applications, and pinpoint areas for additional research and awareness-raising initiatives.

Innovative Techniques for Infection Control and Surveillance in Hospital Settings and Long-Term Care Facilities: A Scoping Review

Healthcare-associated infections (HAIs) pose significant challenges in healthcare systems, with preventable surveillance playing a crucial role. Traditional surveillance, although effective, is resource-intensive. The development of new technologies, such as artificial intelligence (AI), can support traditional surveillance in analysing an increasing amount of health data or meeting patient needs. We conducted a scoping review, following the PRISMA-ScR guideline, searching for studies of new digital technologies applied to the surveillance, control, and prevention of HAIs in hospitals and LTCFs published from 2018 to 4 November 2023. The literature search yielded 1292 articles. After title/abstract screening and full-text screening, 43 articles were included. The mean study duration was 43.7 months. Surgical site infections (SSIs) were the most-investigated HAI and machine learning was the most-applied technology. Three main themes emerged from the thematic analysis: patient empowerment, workload reduction and cost reduction, and improved sensitivity and personalization. Comparative analysis between new technologies and traditional methods showed different population types, with machine learning methods examining larger populations for AI algorithm training. While digital tools show promise in HAI surveillance, especially for SSIs, challenges persist in resource distribution and interdisciplinary integration in healthcare settings, highlighting the need for ongoing development and implementation strategies.

Machine learning models to detect and predict patient safety events using electronic health records: A systematic review

INTRODUCTION

Identifying patient safety events using electronic health records (EHRs) and automated machine learning-based detection methods can help improve the efficiency and quality of healthcare service provision.

OBJECTIVE

This study aimed to systematically review machine learning-based methods and techniques, as well as their results for patient safety event management using EHRs.

METHODS

We reviewed the studies that focused on machine learning techniques, including automatic prediction and detection of patient safety events and medical errors through EHR analysis to manage patient safety events. The data were collected by searching Scopus, PubMed (Medline), Web of Science, EMBASE, and IEEE Xplore databases.

RESULTS

After screening, 41 papers were reviewed. Support vector machine (SVM), random forest, conditional random field (CRF), and bidirectional long short-term memory with conditional random field (BiLSTM-CRF) algorithms were mostly applied to predict, identify, and classify patient safety events using EHRs; however, they had different performances. BiLSTM-CRF was employed in most of the studies to extract and identify concepts, e.g., adverse drug events (ADEs) and adverse drug reactions (ADRs), as well as relationships between drug and severity, drug and ADEs, drug and ADRs. Recurrent neural networks (RNN) and BiLSTM-CRF had the best results in detecting ADEs compared to other patient safety events. Linear classifiers and Naive Bayes (NB) had the highest performance for ADR detection. Logistic regression had the best results in detecting surgical site infections. According to the findings, the quality of articles has non-significantly improved in recent years, but they had low average scores.

CONCLUSIONS

Machine learning can be useful in automatic detection and prediction of patient safety events. However, most of these algorithms have not yet been externally validated or prospectively tested. Therefore, further studies are required to improve the performance of these automated systems.

Association between quality of governance, antibiotic consumption, and antimicrobial resistance: an analysis of Italian regions

BACKGROUND

Emerging research has provided evidence suggesting the potential influence of governance on the development and spread of antimicrobial resistance (AMR), accounting for significant disparities observed both between and within countries. In our study, we conducted an ecological analysis to investigate the relationship between governance quality, antibiotic consumption, and AMR across Italian regions.

METHODS

By leveraging data from three distinct sources at the regional level, we compiled a comprehensive dataset comprising: AMR proportions for three specific pathogen-antibiotic combinations in the year 2021, antibiotic consumption data for systemic use in the year 2020, and the 2021 European Quality of Government Index (EQI) and its corresponding pillars. Employing mediation analysis, we investigated the potential mediating role of antibiotic consumption in the association between the EQI and an average measure of AMR.

RESULTS

Our analysis revealed substantial variation in the percentages of AMR across different regions in Italy, demonstrating a discernible North-to-South gradient concerning both antibiotic usage and governance quality. The EQI exhibited a statistically significant negative correlation with both antibiotic consumption and AMR percentages, encompassing both specific combinations and their average value. Regions characterized by higher levels of governance quality consistently displayed lower values of antibiotic consumption and AMR, while regions with lower governance quality tended to exhibit higher levels of antibiotic use and AMR. Furthermore, we observed a significant total effect of the EQI on average AMR (β = - 0.97; CI - 1.51; - 0.43). Notably, this effect was found to be mediated by antibiotic consumption, as evidenced by a significant indirect effect (β = - 0.89; CI - 1.45; - 0.32).

CONCLUSIONS

These findings draw attention to the regional disparities observed in AMR levels, antibiotic consumption patterns, and governance quality in Italy. Our study also highlights the mediating role of antibiotic consumption in the relationship between governance quality and AMR. This underscores the significance of implementing focused interventions and policies aimed at improving governance quality and promoting responsible antibiotic use.

The intertwining of healthcare-associated infections and COVID-19 in Italian intensive care units: an analysis of the SPIN-UTI project from 2006 to 2021

BACKGROUND

Although healthcare-associated infections (HAIs) pose an extraordinary burden on public health, the impact of coronavirus disease 2019 (COVID-19) is still a matter of debate.

AIM

To describe trends of HAIs in Italian intensive care units (ICUs) from 2006 to 2021, and to compare characteristics and outcomes of patients with or without COVID-19.

METHODS

We evaluated patients participating in the 'Italian Nosocomial Infections Surveillance in Intensive Care Units' (SPIN-UTI) project, who were admitted to ICUs for more than 48 h. Data regarding diagnosis, clinical conditions, therapies, treatments and outcomes of COVID-19 patients were also collected.

FINDINGS

From a total of 21,523 patients from 2006 to 2021, 3485 (16.2%) presented at least one HAI. We observed an increasing trend for both the incidence of patients with HAI and the incidence density of HAIs (P-trend <0.001). Compared with the pre-pandemic period, the incidence density of HAIs increased by about 15% in 2020-2021, with pneumoniae being the greatest contributors to this increase (P-trend <0.001). Moreover, incidence of HAIs was higher in ICUs dedicated to COVID-19 patients (P<0.001), who showed a greater risk of HAIs and death than patients without COVID-19 (P-values <0.001). Accordingly, the mortality in ICUs increased over the years and doubled during the pandemic (P-trend <0.001). Notably, co-infected patients had higher mortality (75.2%) than those with COVID-19 (66.2%) or HAI (39.9%) alone, and those without any infection (23.2%).

CONCLUSIONS

Our analysis provides useful insight into whether and how the COVID-19 pandemic influenced HAI incidence and death in Italian ICUs, highlighting the need for evaluation of the long-term effects of the pandemic.

An observational study on factors associated with ICU mortality in Covid-19 patients and critical review of the literature

The novel pandemic caused by SARS-CoV-2 has been associated with increased burden on healthcare system. Recognizing the variables that independently predict death in COVID-19 is of great importance. The study was carried out prospectively in a single ICU in northern Greece. It was based on the collection of data during clinical practice in 375 adult patients who were tested positive for SARS-CoV-2 between April 2020 and February 2022. All patients were intubated due to acute respiratory insufficiency and received Invasive Mechanical Ventilation. The primary outcome was ICU mortality. Secondary outcomes were 28-day mortality and independent predictors of mortality at 28 days and during ICU hospitalization. For continuous variables with normal distribution, t-test was used for means comparison between two groups and one-way ANOVA for multiple comparisons. When the distribution was not normal, comparisons were performed using the Mann-Whitney test. Comparisons between discrete variables were made using the x² test, whereas the binary logistic regression was employed for the definition of factors affecting survival inside the ICU and after 28 days. Of the total number of patients intubated due to COVID-19 during the study period, 239 (63.7%) were male. Overall, the ICU survival was 49.6%, whereas the 28-day survival reached 46.9%. The survival rates inside the ICU for the four main viral variants were 54.9%, 50.3%, 39.7% and 50% for the Alpha, Beta, Delta and Omicron variants, respectively. Logistic regressions for outcome revealed that the following parameters were independently associated with ICU survival: wave, SOFA @day1, Remdesivir use, AKI, Sepsis, Enteral Insufficiency, Duration of ICU stay and WBC. Similarly, the parameters affecting the 28-days survival were: duration of stay in ICU, SOFA @day1, WBC, Wave, AKI and Enteral Insufficiency. In this observational cohort study of critically ill COVID-19 patients we report an association between mortality and the wave sequence, SOFA score on admission, the use of Remdesivir, presence of AKI, presence of gastrointestinal failure, sepsis and WBC levels. Strengths of this study are the large number of critically ill COVID-19 patients included, and the comparison of the adjusted mortality rates between pandemic waves within a two year-study period.

Predictive Analysis of Healthcare-Associated Blood Stream Infections in the Neonatal Intensive Care Unit Using Artificial Intelligence: A Single Center Study

Background: Neonatal infections represent one of the six main types of healthcare-associated infections and have resulted in increasing mortality rates in recent years due to preterm births or problems arising from childbirth. Although advances in obstetrics and technologies have minimized the number of deaths related to birth, different challenges have emerged in identifying the main factors affecting mortality and morbidity. Dataset characterization: We investigated healthcare-associated infections in a cohort of 1203 patients at the level III Neonatal Intensive Care Unit (ICU) of the “Federico II” University Hospital in Naples from 2016 to 2020 (60 months). Methods: The present paper used statistical analyses and logistic regression to identify an association between healthcare-associated blood stream infection (HABSIs) and the available risk factors in neonates and prevent their spread. We designed a supervised approach to predict whether a patient suffered from HABSI using seven different artificial intelligence models. Results: We analyzed a cohort of 1203 patients and found that birthweight and central line catheterization days were the most important predictors of suffering from HABSI. Conclusions: Our statistical analyses showed that birthweight and central line catheterization days were significant predictors of suffering from HABSI. Patients suffering from HABSI had lower gestational age and birthweight, which led to longer hospitalization and umbilical and central line catheterization days than non-HABSI neonates. The predictive analysis achieved the highest Area Under Curve (AUC), accuracy and F1-macro score in the prediction of HABSIs using Logistic Regression (LR) and Multi-layer Perceptron (MLP) models, which better resolved the imbalanced dataset (65 infected and 1038 healthy).

Burden of Healthcare-Associated Infections in Sicily, Italy: Estimates from the Regional Point Prevalence Surveys 2016-2018

An assessment of the burden of healthcare-associated infections (HAIs) in terms of disability-adjusted life years (DALYs) is useful for comparing and ranking HAIs and to support infection prevention and control strategies. We estimated the burden of healthcare-associated pneumoniae (HAP), bloodstream infection (HA BSI), urinary tract infection (HA UTI), and surgical site infection (SSI) in Sicily, Italy. We used data from 15,642 patients aged 45 years and above, identified during three repeated point prevalence surveys (PPSs) conducted from 2016 to 2018 according to the European Centre for Disease Prevention and Control protocol. The methodology of the Burden of Communicable Diseases in Europe project was employed. The selected HAIs accounted for 8424 DALYs (95% uncertainty interval (UI): 7394-9605) annually in Sicily, corresponding to 344 DALYs per 100,000 inhabitants aged 45 years and above (95% UI: 302-392). Notably, more than 60% of the burden was attributable to HAP, followed by HA BSI, SSI, and HA UTI. The latter had the lowest burden despite a relatively high incidence, whereas HA BSI generated a high burden even through a relatively low incidence. Differences between our estimates and those of European and Italian PPSs encourage the estimation of the burden of HAIs region by region.

Disentangling the Association of Hydroxychloroquine Treatment with Mortality in Covid-19 Hospitalized Patients through Hierarchical Clustering

The efficacy of hydroxychloroquine (HCQ) in treating SARS-CoV-2 infection is harshly debated, with observational and experimental studies reporting contrasting results. To clarify the role of HCQ in Covid-19 patients, we carried out a retrospective observational study of 4,396 unselected patients hospitalized for Covid-19 in Italy (February–May 2020). Patients' characteristics were collected at entry, including age, sex, obesity, smoking status, blood parameters, history of diabetes, cancer, cardiovascular and chronic pulmonary diseases, and medications in use. These were used to identify subtypes of patients with similar characteristics through hierarchical clustering based on Gower distance. Using multivariable Cox regressions, these clusters were then tested for association with mortality and modification of effect by treatment with HCQ. We identified two clusters, one of 3,913 younger patients with lower circulating inflammation levels and better renal function, and one of 483 generally older and more comorbid subjects, more prevalently men and smokers. The latter group was at increased death risk adjusted by HCQ (HR[CI95%] = 3.80[3.08-4.67]), while HCQ showed an independent inverse association (0.51[0.43-0.61]), as well as a significant influence of cluster∗HCQ interaction (p < 0.001). This was driven by a differential association of HCQ with mortality between the high (0.89[0.65-1.22]) and the low risk cluster (0.46[0.39-0.54]). These effects survived adjustments for additional medications in use and were concordant with associations with disease severity and outcome. These findings suggest a particularly beneficial effect of HCQ within low risk Covid-19 patients and may contribute to clarifying the current controversy on HCQ efficacy in Covid-19 treatment.