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.

BACKGROUND: Hospital-acquired infections (HAIs) increase morbidity, mortality, and health care costs. Effective hand hygiene (HH) is crucial for prevention, but achieving high compliance remains challenge. This study explores using machine learning to integrate an electronic HH auditing system with electronic health records to predict HAIs. METHODS: A retrospective cohort study was conducted at a Brazilian hospital during 2017-2020. HH compliance was recorded electronically, and patient data were collected from electronic health records. The primary outcomes were HAIs per CDC/National Healthcare Safety Network surveillance definitions. Machine learning algorithms, balanced with Random Over Sampling Examples (ROSE), were utilized for predictive modeling, including generalized linear models (GLM); generalized additive models for location, scale, and shape (GAMLSS); random forest; support vector machine; and extreme gradient boosting (XGboost). RESULTS: 125 of 6,253 patients (2%) developed HAIs and 920,489 HH opportunities (49.3% compliance) were analyzed. A direct correlation between HH compliance and HAIs was observed. The GLM algorithm with ROSE demonstrated superior performance, with 84.2% sensitivity, 82.9% specificity, and a 93% AUC. CONCLUSIONS: Integrating electronic HH auditing systems with electronic health records and using machine learning models can enhance infection control surveillance and predict patient outcomes. Further research is needed to validate these findings and integrate them into clinical practice.

Surveillance programs for detection and characterization of emergent pathogens and antimicrobial resistance: results from the Division of Infectious Diseases, UNIFESP.

Several epidemiological changes have occurred in the pattern of nosocomial and community acquired infectious diseases during the past 25 years. Social and demographic changes possibly related to this phenomenon include a rapid population growth, the increase in urban migration and movement across international borders by tourists and immigrants, alterations in the habitats of animals and arthropods that transmit disease, as well as the raise of patients with impaired host defense abilities. Continuous surveillance programs of emergent pathogens and antimicrobial resistance are warranted for detecting in real time new pathogens, as well as to characterize molecular mechanisms of resistance. In order to become more effective, surveillance programs of emergent pathogens should be organized as a multicenter laboratory network connected to the main public and private infection control centers. Microbiological data should be integrated to guide therapy, adapting therapy to local ecology and resistance patterns. This paper presents an overview of data generated by the Division of Infectious Diseases, Federal University of São Paulo, along with its participation in different surveillance programs of nosocomial and community acquired infectious diseases.

Nosocomial bloodstream infections caused by Klebsiella pneumoniae: impact of extended-spectrum beta-lactamase (ESBL) production on clinical outcome in a hospital with high ESBL prevalence.

BACKGROUND: The frequency of ESBL producing Klebsiella pneumoniae bloodstream infections (BSI) is high in Brazilian hospitals, however little is known regarding what role, if any, resistance plays in the expected outcome in hospitals with a high prevalence of these pathogens. METHODS: From 1996 to 2001, hospital acquired K. pneumoniae BSI were evaluated retrospectively. Each patient was included only once at the time of BSI. ESBL producing strains were identified using the E-test method. The association of variables with the mortality related to bacteremia was included in a stepwise logistic regression model. RESULTS: One hundred and eight hospital acquired K. pneumoniae BSI met criteria for inclusion. Fifty two percent were due to ESBL producing strains. The overall in-hospital mortality was 40.8%. Variables independently predicting death by multivariate analysis were the following: mechanical ventilation (p = 0.001), number of comorbidities (p = 0.003), antimicrobials prescribed before bacteremia (p = 0.01) and fatal underlying disease (p = 0.025). CONCLUSION: Bacteremia due to ESBL producing K. pneumoniae strains was not an independent predictor for death in patients with BSI. An increased mortality in hospital-acquired BSI by K. pneumoniae was related to the requirement for mechanical ventilation, more than two comorbidities, the previous use of two or more antibiotics, and the presence of a rapidly fatal disease.

Compliance with handwashing at two intensive care units in São Paulo.

Handwashing compliance was compared at two medical- surgical intensive care units (ICUs) of a teaching and a non-teaching hospital. The mean compliance was 22.2% and 42.6%, respectively. Respiratory therapists at the non-teaching hospital had the best handwashing compliance (52.6%). Nursing assistants at the teaching hospital had the worst compliance (11.5%). Nursing assistant was the only health-care worker category with a significant difference between the two ICUs (odds ratio = 6.0; 95% confidence interval = 3.83-9.43; p< 0.001).

Implementation of an antibiotic prophylaxis protocol in an intensive care unit.

BACKGROUND: When properly employed, the prophylactic use of antimicrobials is associated with a reduction in surgical site infections (SSIs). We found that the appropriate use of antimicrobial prophylaxis was only 50.5% (53/105) among patients undergoing surgery in the adult intensive care unit of our hospital. In 2001, a protocol was designed to improve compliance with recommended practice. METHODS: We used a prospective interventional study and a case control study carried out between 2001 and 2007, including follow-up and daily intervention to improve compliance with antimicrobial prophylaxis guidelines and to monitor antimicrobial consumption and SSI rates. Cases of noncompliance to the prophylaxis protocol (group I) were matched to controls (group II) with appropriate prophylaxis and compared with regards to type of surgery, operative duration, intraoperative antimicrobial use, type of antimicrobial used, length of hospital stay, severity of illness, comorbidities, invasive devices, possible adverse reactions, and death. RESULTS: Compliance with antimicrobial prophylaxis metrics reached 85%; however, we were unable to detect a change in SSI rate or consumption and cost of antimicrobials. Inappropriate use was not associated with higher likelihood of death. There were no other significant differences between the 2 groups. CONCLUSION: Our intervention increased compliance with appropriate antimicrobial surgical prophylaxis with no negative impact on patient safety.

Imipenem-resistant Pseudomonas aeruginosa infection at a medical-surgical intensive care unit: risk factors and mortality.

OBJECTIVES: The aim of this study was to evaluate the risk factors and attributable mortality associated with imipenem-resistant Pseudomonas aeruginosa (IRPA) infections in a medical-surgical intensive care unit (ICU). METHODS: A retrospective case-control study was carried out at a 16-bed medical-surgical ICU in a 780-bed, university-affiliated hospital. All patients admitted from January 1, 2003, to December 31, 2004, who had nosocomial infection caused by IRPA, were included in the study. RESULTS: Imipenem-resistant P. aeruginosa was recovered from 63 patients during the study period. One hundred eighty-two controls were matched with cases by period of admission, age, and time at risk. Urinary tract (34.9%) and respiratory tract (22.2%) were the main sources of IRPA isolation. In multivariate analysis, a previous stay in the ICU (odds ratio, 3.54; 95% confidence interval [CI], 1.29-9.73; P = .03) was the only independent risk factor for IRPA infection. The in-hospital mortality rate among case patients was 49% (31 of 63) compared with 33% (61 of 182) for control patients (odds ratio, 1.92; 95% CI, 1.07-3.44; P = .02). Thus, we had an attributable mortality of 16% (95% CI, 9.74%-22.3%; P = .03). CONCLUSIONS: Our study suggests that IRPA infections are strongly related to previous ICU stay, and that IRPA infections significantly increase mortality in those critical patients.

Risk factors for surgical site infection in kidney transplant recipients.

We analyzed the epidemiologic characteristics and risk factors for surgical site infection (SSI) in kidney transplant recipients. From among 1,939 kidney transplant recipients, 120 with corresponding control subjects were evaluated in this study (1:1 ratio). Reoperation, chronic glomerulonephritis, acute graft rejection, delayed graft function, diabetes, and high body mass index were identified in the analysis as risk factors for SSI.

Bloodstream infections with metallo-beta-lactamase-producing Pseudomonas aeruginosa: epidemiology, microbiology, and clinical outcomes.

Pseudomonas aeruginosa strains that produce metallo-beta-lactamases (MBLs) are becoming increasingly prevalent. We evaluated the epidemiological and microbiological characteristics of monomicrobial bloodstream infections caused by MBL-producing P. aeruginosa isolates, as well as the clinical outcomes in patients with these infections.

Surveillance programs for detection and characterization of emergent pathogens and antimicrobial resistance: results from the Division of Infectious Diseases, UNIFESP

Several epidemiological changes have occurred in the pattern of nosocomial and community acquired infectious diseases during the past 25 years. Social and demographic changes possibly related to this phenomenon include a rapid population growth, the increase in urban migration and movement across international borders by tourists and immigrants, alterations in the habitats of animals and arthropods that transmit disease, as well as the raise of patients with impaired host defense abilities. Continuous surveillance programs of emergent pathogens and antimicrobial resistance are warranted for detecting in real time new pathogens, as well as to characterize molecular mechanisms of resistance. In order to become more effective, surveillance programs of emergent pathogens should be organized as a multicenter laboratory network connected to the main public and private infection control centers. Microbiological data should be integrated to guide therapy, adapting therapy to local ecology and resistance patterns. This paper presents an overview of data generated by the Division of Infectious Diseases, Federal University of São Paulo, along with its participation in different surveillance programs of nosocomial and community acquired infectious diseases.

Várias alterações epidemiológicas ocorreram no perfil das doenças infecciosas hospitalares e comunitárias nos últimos 25 anos. Mudanças sociais e demográficas possivelmente relacionadas com esse fenômeno incluem o rápido crescimento populacional, o aumento da migração urbana e deslocamento através de fronteiras internacionais por turistas e imigrantes, alterações nos habitats de animais e artrópodes que transmitem doença assim como o aumento no número de pacientes com deficiências nas respostas de defesa. Os programas contínuos de vigilância de patógenos emergentes e resistência antimicrobiana são necessários para a detecção em tempo real de novos patógenos assim como para caracterizar mecanismos moleculares de resistência. Para serem mais efetivos, os programasde vigilância dos patógenos emergentes devem ser organizados em uma rede de laboratórios multicêntricos ligados aos principais centros de controle de infecções, públicos e privados. Os dados microbiológicos devem ser integrados a guias terapêuticos adaptando práticas terapêuticas à ecologia local eaos padrões de resistência. O artigo apresenta uma revisão dos dados gerados pela Disciplina de Infectologia, Universidade Federal de São Paulo, contemplando sua participação nos diferentes programas de vigilância de doenças infecciosas hospitalares e adquiridas na comunidade.

Compliance with handwashing at two intensive care units in São Paulo

Handwashing compliance was compared at two medical- surgical intensive care units (ICUs) of a teaching and a non-teaching hospital. The mean compliance was 22.2 percent and 42.6 percent, respectively. Respiratory therapists at the non-teaching hospital had the best handwashing compliance (52.6 percent). Nursing assistants at the teaching hospital had the worst compliance (11.5 percent). Nursing assistant was the only health-care worker category with a significant difference between the two ICUs (odds ratio = 6.0; 95 percent confidence interval = 3.83-9.43; p< 0.001).

¿Es aplicable el sistema de vigilancia epidemiológica de las infecciones intrahospitalarias que recomienda el CDC-americano (sistema NNIS) en un hospital chileno? / Application of the national nosocomial infections surveillance system recommended by the CDC, in a Chilean hospital

Background: The National Nosocomial Infections Surveillance System (NNIS system) is the method for surveying nosocomial infections used by the Centers for Disease Control (CDC). This strategy allows the comparison of different hospitals, using rate adjustments. In Chile, this system is not used. Aim: To report the application of this system in a tertiary reference hospital in Chile. Materials and methods: We performed a six months prospective cohort study. The active surveillance was carried out by using the intensive care unit and surgery components of the NNIS system. Tabulation and analysis of the data were done according to the NNIS system. In a parallel prevalence study, we determined the NNIS system sensitivity to detect nosocomial infections. Results: A total of 492 patients were followed with a global nosocomial infection rate of 14 per cent, for discharged patients. The calculated sensitivity and specificity of the system was 84.2 and 97 per cent respectively. In the intensive care unit, 45 of 169 patients had nosocomial infections, with an adjusted rate of 2.8 per cent for mean hospitalization time and severity of illness. In the cardiovascular and thoracic surgical units, 216 and 107 procedures were surveyed, respectively. The global rates of nosocomial infections were 7.4 and 7.5 per cent, respectively. The adjusted rates according to risk factors were 0.9 and 2.3 per cent, respectively. Conclusions: These data indicate that the surgical units had surgical site infections rates similar to those reported by the CDC. Nosocomial infections rates in Chile can be compared with rates observed in other countries. The epidemiological data collected can be useful to focus intervention or preventive strategies

O uso da cefepima em pacientes com infecçöes moderadas do trato respiratório inferior e do trato urinário superior

Objetivo: Determinar a eficácia e segurança do tratamento empírico com a cefepima em infecçöes moderadas do trato respiratório inferior e do trato urinário superior. Local: Hospital Universitário de 600 leitos. Desenho do estudo: Estudo clínico aberto, näo-comparativo. Causuística: Foram avaliados 27 pacientes adultos com infecçäo do trato respiratório inferior e do trato urinário superior que requereram hospitalizaçäo e tratamento empírico. Pacientes neutropênicos foram excluídos. O acompanhamento foi realizado até a alta hospitalar. Foram analisadas as respostas ao tratamento do ponto de vista clínico e microbiológico. Resultados: Vinte e sete pacientes foram analisados, no período de 3/95 a 12/96. A idade média foi de 40,5 anos e pacientes do sexo feminino foram os mais prevalentes, 17 (63 por cento). As infecçöes responsáveis pelas hospitalizaçöes foram: pneumonia (14), pielonefrite (11) e bronquite (2). A média do tempo de hospitalizaçöes foi de 14 dias. Dezesete pacientes eram portadores de doença de base no momento da admissäo ao hospital. Bactérias consideradas como agentes causadores da infecçäo foram isoladas em 25 pacientes, todas suscetíveis à cefepima; P.aeruginosa (4), A.baumanii (2), entre outras. A cura microbiológica foi obtida em todos os pacientes, e a cura clínica em 26. Um doente apresentou cura parcial. Nenhum efeito adverso grave foi encontrado. Um paciente apresentou febre persistente que resolveu com a suspensäo do antibiótico e um paciente apresentou leucopenia leve, näo necessitando suspender o tratamento