1
|
Cho Y, Lee HK, Kim J, Yoo KB, Choi J, Lee Y, Choi M. Prediction of hospital-acquired influenza using machine learning algorithms: a comparative study. BMC Infect Dis 2024; 24:466. [PMID: 38698304 PMCID: PMC11067145 DOI: 10.1186/s12879-024-09358-1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/26/2023] [Accepted: 04/26/2024] [Indexed: 05/05/2024] Open
Abstract
BACKGROUND Hospital-acquired influenza (HAI) is under-recognized despite its high morbidity and poor health outcomes. The early detection of HAI is crucial for curbing its transmission in hospital settings. AIM This study aimed to investigate factors related to HAI, develop predictive models, and subsequently compare them to identify the best performing machine learning algorithm for predicting the occurrence of HAI. METHODS This retrospective observational study was conducted in 2022 and included 111 HAI and 73,748 non-HAI patients from the 2011-2012 and 2019-2020 influenza seasons. General characteristics, comorbidities, vital signs, laboratory and chest X-ray results, and room information within the electronic medical record were analysed. Logistic Regression (LR), Random Forest (RF), Extreme Gradient Boosting (XGB), and Artificial Neural Network (ANN) techniques were used to construct the predictive models. Employing randomized allocation, 80% of the dataset constituted the training set, and the remaining 20% comprised the test set. The performance of the developed models was assessed using metrics such as the area under the receiver operating characteristic curve (AUC), the count of false negatives (FN), and the determination of feature importance. RESULTS Patients with HAI demonstrated notable differences in general characteristics, comorbidities, vital signs, laboratory findings, chest X-ray result, and room status compared to non-HAI patients. Among the developed models, the RF model demonstrated the best performance taking into account both the AUC (83.3%) and the occurrence of FN (four). The most influential factors for prediction were staying in double rooms, followed by vital signs and laboratory results. CONCLUSION This study revealed the characteristics of patients with HAI and emphasized the role of ventilation in reducing influenza incidence. These findings can aid hospitals in devising infection prevention strategies, and the application of machine learning-based predictive models especially RF can enable early intervention to mitigate the spread of influenza in healthcare settings.
Collapse
Affiliation(s)
- Younghee Cho
- College of Nursing, Yonsei University, Seoul, Republic of Korea
- Department of Digital Health, Samsung SDS, Seoul, Republic of Korea
| | - Hyang Kyu Lee
- College of Nursing, Yonsei University, Seoul, Republic of Korea
| | - Joungyoun Kim
- College of Engineering, University of Seoul, Seoul, Republic of Korea
| | - Ki-Bong Yoo
- Division of Health Administration, Yonsei University, Wonju, Republic of Korea
| | - Jongrim Choi
- College of Nursing, Keimyung University, Daegu, Republic of Korea
| | - Yongseok Lee
- Department of Digital Health, Samsung SDS, Seoul, Republic of Korea
| | - Mona Choi
- College of Nursing, Yonsei University, Seoul, Republic of Korea.
- Mo-Im Kim Nursing Research Institute, College of Nursing, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul, 03722, Republic of Korea.
| |
Collapse
|
2
|
Suh W, Han SB. Nosocomial influenza in a pediatric general ward: Effects of isolation and cohort placement of children with influenza. Infect Control Hosp Epidemiol 2023; 44:1637-1642. [PMID: 36924052 DOI: 10.1017/ice.2023.14] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 03/18/2023]
Abstract
OBJECTIVE Many studies have described nosocomial outbreaks of influenza in specialized wards. We evaluated nosocomial transmission of influenza in a pediatric general ward. DESIGN Retrospective observational study. SETTING Single secondary hospital. PATIENTS The study included 814 hospitalized children with influenza between September 2015 and August 2020. METHODS The medical records of the included children were retrospectively reviewed, and clinical characteristics of children with community-acquired (CA) influenza and hospital-acquired (HA) influenza were determined. The room of each included child during hospitalization was traced to identify the children exposed to them. RESULTS CA influenza and HA influenza were diagnosed in 789 (96.9%) and 25 (3.1%) children, respectively. Among children with CA influenza, 691 (87.6%) were isolated or place in a cohort on admission. In total, 98 children (12.4%) admitted to multibed rooms exposed 307 children with noninfluenza diseases to influenza during 772 patient days; 3 exposed children (1.0%) were diagnosed with HA influenza. Including these 3 children, 25 children (19 without definite in-hospital exposure to influenza and 3 exposed to other children with HA influenza) were diagnosed with HA influenza, and 11 (44.0%) exposed 31 children with noninfluenza diseases to influenza for 85 patient days. Also, 3 exposed children (9.7%) were diagnosed with HA influenza, a significantly higher rate than that for CA influenza (P = .005). The clinical characteristics were comparable between children with HA influenza and those with CA influenza. CONCLUSIONS Cohort placement of children with influenza in a pediatric general ward can be effective in controlling nosocomial transmission of influenza. However, control measures for children with HA influenza should be emphasized.
Collapse
Affiliation(s)
- Woosuck Suh
- Department of Pediatrics, Daejeon St. Mary's Hospital, The Catholic University of Korea, Daejeon, Republic of Korea
| | - Seung Beom Han
- Department of Pediatrics, Hallym University Hangang Sacred Heart Hospital, Seoul, Republic of Korea
| |
Collapse
|
3
|
Saadatian-Elahi M, Henaff L, Elias C, Nunes MC, Hot A, Martin-Gaujard G, Escuret V, Amour S, Vanhems P. Patient influenza vaccination reduces the risk of hospital-acquired influenza: An incident test negative-case control study in Lyon university hospital, France (2004-2020). Vaccine 2023; 41:4341-4346. [PMID: 37321894 DOI: 10.1016/j.vaccine.2023.05.060] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/27/2023] [Revised: 05/02/2023] [Accepted: 05/26/2023] [Indexed: 06/17/2023]
Abstract
BACKGROUND Literature is limited on the impact of patient vaccination on the risk of hospital-acquired influenza (HAI). This test negative case-control study nested in a surveillance program aimed at evaluating the effectiveness of influenza vaccination in reducing the risk of HAI in hospitalized patients during 15 influenza seasons (2004-05 to 2019-20). METHODS HAI cases were those who developed influenza like illness (ILI) symptoms at least 72 h after hospitalization and had a positive reverse transcriptase-polymerase chain reaction (RT-PCR). Controls were those with ILI symptoms and a negative RT-PCR test. A nasal swab as well as socio-demographic, clinical data and information on influenza vaccination were collected. RESULTS Of the 296 patients included, 67 were confirmed HAI cases. Influenza vaccine coverage was significantly higher among controls compared to HAI cases (p = 0.002). The risk of HAI was reduced by almost 60 % in vaccinated patients. CONCLUSIONS A better control of HAI can be achieved by vaccinating hospitalized patients.
Collapse
Affiliation(s)
- Mitra Saadatian-Elahi
- Service Hygiène, Epidémiologie et Prévention, Centre Hospitalier Hôpital Eduard Herriot, Hospices Civils de Lyon, 69437 Lyon Cedex, France; CIRI, Centre International de Recherche en Infectiologie, (Team Public Health, Epidemiology and Evolutionary Ecology of Infectious Diseases (PHE3ID)), Univ Lyon, Inserm, U1111, Université Claude Bernard Lyon 1, CNRS, UMR5308, ENS de Lyon, F-69007 Lyon, France
| | - Laetitia Henaff
- Service Hygiène, Epidémiologie et Prévention, Centre Hospitalier Hôpital Eduard Herriot, Hospices Civils de Lyon, 69437 Lyon Cedex, France; CIRI, Centre International de Recherche en Infectiologie, (Team Public Health, Epidemiology and Evolutionary Ecology of Infectious Diseases (PHE3ID)), Univ Lyon, Inserm, U1111, Université Claude Bernard Lyon 1, CNRS, UMR5308, ENS de Lyon, F-69007 Lyon, France
| | - Christelle Elias
- Service Hygiène, Epidémiologie et Prévention, Centre Hospitalier Hôpital Eduard Herriot, Hospices Civils de Lyon, 69437 Lyon Cedex, France; CIRI, Centre International de Recherche en Infectiologie, (Team Public Health, Epidemiology and Evolutionary Ecology of Infectious Diseases (PHE3ID)), Univ Lyon, Inserm, U1111, Université Claude Bernard Lyon 1, CNRS, UMR5308, ENS de Lyon, F-69007 Lyon, France
| | - Marta C Nunes
- CIRI, Centre International de Recherche en Infectiologie, (Team Public Health, Epidemiology and Evolutionary Ecology of Infectious Diseases (PHE3ID)), Univ Lyon, Inserm, U1111, Université Claude Bernard Lyon 1, CNRS, UMR5308, ENS de Lyon, F-69007 Lyon, France; Centre for Excellence in Respiratory Pathogens, Hospices Civils de Lyon, Lyon, France; South African Medical Research Council, Vaccines & Infectious Diseases Analytics Research Unit, and Department of Science and Technology/National Research Foundation, South African Research Chair Initiative in Vaccine Preventable Diseases, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
| | - Arnaud Hot
- Internal Medicine, University Hospital Edouard Herriot, Hospices Civils de Lyon, France
| | | | - Vanessa Escuret
- Laboratoire de Virologie, Institut des Agents Infectieux, Hôpital de la Croix-Rousse, Hospices Civils de Lyon, Lyon, France; Virpath - Grippe, de l'émergence au contrôle, Centre International de Recherche en Infectiologie (CIRI), Inserm U111, CNRS 5308, ENS, UCBL1, Faculté de Médecine RTH Laënnec, Lyon, France
| | - Selilah Amour
- Service Hygiène, Epidémiologie et Prévention, Centre Hospitalier Hôpital Eduard Herriot, Hospices Civils de Lyon, 69437 Lyon Cedex, France; CIRI, Centre International de Recherche en Infectiologie, (Team Public Health, Epidemiology and Evolutionary Ecology of Infectious Diseases (PHE3ID)), Univ Lyon, Inserm, U1111, Université Claude Bernard Lyon 1, CNRS, UMR5308, ENS de Lyon, F-69007 Lyon, France
| | - Philippe Vanhems
- Service Hygiène, Epidémiologie et Prévention, Centre Hospitalier Hôpital Eduard Herriot, Hospices Civils de Lyon, 69437 Lyon Cedex, France; CIRI, Centre International de Recherche en Infectiologie, (Team Public Health, Epidemiology and Evolutionary Ecology of Infectious Diseases (PHE3ID)), Univ Lyon, Inserm, U1111, Université Claude Bernard Lyon 1, CNRS, UMR5308, ENS de Lyon, F-69007 Lyon, France.
| |
Collapse
|
4
|
Snell LB, Vink JP, Verlander NQ, Miah S, Lackenby A, Williams D, Mitchell H, Beviz C, Kabia M, Cunningham E, Batra R, Edgeworth JD, Zambon M, Nebbia G. Nosocomial acquisition of influenza is associated with significant morbidity and mortality: Results of a prospective observational study. J Infect Public Health 2022; 15:1118-1123. [PMID: 36137361 PMCID: PMC10166711 DOI: 10.1016/j.jiph.2022.08.021] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/07/2022] [Revised: 08/04/2022] [Accepted: 08/31/2022] [Indexed: 10/14/2022] Open
Abstract
BACKGROUND Nosocomial acquisition of influenza is known to occur but the risk after exposure to a known case and the outcomes after acquisition are poorly defined. METHODS Prospective observational study of patients exposed to influenza from another patient in a multi-site healthcare organisation, with follow-up of 7 days or until discharge, and PCR-confirmation of symptomatic disease. Multivariable analysis was used to investigate association of influenza acquisition with high dependency unit/intensive care unit (HDU/ITU) admission and in-hospital mortality. RESULTS 23/298 (7.7%) contacts of 11 cases were subsequently symptomatic and tested influenza-positive during follow-up. HDU/ITU admission was significantly higher in these secondary cases (6/23, 26%) compared to flu-negative contacts (20/275, 7.2%; p = 0.002). In-hospital mortality was significantly higher in secondary cases (5/23, 21.7%) compared to flu-negative contacts (11/275, 4%; p < 0.001). In multivariable analysis, age (OR 1.25 95% CI: 1.01-1.54, p = 0.02) and being a secondary case (OR 4.77, 95% CI: 1.63-13.9, p = 0.008) were significantly associated with HDU/ITU admission in contacts. Age (OR 1.00, 95% CI: 0.93-1.00, p = 0.02), being a secondary case after exposure to influenza (OR 3.81, 95% CI 1.09-13.3, p = 0.049) and co-morbidity (OR 1.29 per unit increment in the Charlson score, 95% CI 1.02-1.61, p = 0.03) were significantly associated with in-hospital mortality in contacts. CONCLUSIONS Nosocomial acquisition of influenza was significantly associated with increased risk of HDU/ITU admission and in-hospital mortality.
Collapse
Affiliation(s)
- L B Snell
- Centre for Clinical Infection and Diagnostics Research, Guy's and St. Thomas' NHS Foundation Trust, London, UK; Department of Infection, Guy's and St Thomas' NHS Foundation Trust, UK.
| | - J P Vink
- Centre for Clinical Infection and Diagnostics Research, Guy's and St. Thomas' NHS Foundation Trust, London, UK
| | | | - S Miah
- UK Health Security Agency, Colindale, UK
| | - A Lackenby
- UK Health Security Agency, Colindale, UK
| | - D Williams
- UK Health Security Agency, Colindale, UK
| | - H Mitchell
- Department of Infection, Guy's and St Thomas' NHS Foundation Trust, UK
| | - C Beviz
- Department of Infection, Guy's and St Thomas' NHS Foundation Trust, UK
| | - M Kabia
- Department of Infection, Guy's and St Thomas' NHS Foundation Trust, UK
| | | | - R Batra
- Centre for Clinical Infection and Diagnostics Research, Guy's and St. Thomas' NHS Foundation Trust, London, UK
| | - J D Edgeworth
- Centre for Clinical Infection and Diagnostics Research, Guy's and St. Thomas' NHS Foundation Trust, London, UK; Department of Infection, Guy's and St Thomas' NHS Foundation Trust, UK
| | - M Zambon
- UK Health Security Agency, Colindale, UK
| | - G Nebbia
- Centre for Clinical Infection and Diagnostics Research, Guy's and St. Thomas' NHS Foundation Trust, London, UK; Department of Infection, Guy's and St Thomas' NHS Foundation Trust, UK
| |
Collapse
|
5
|
Mangas-Moro A, Zamarrón-de-Lucas E, Carpio-Segura CJ, Álvarez-Sala-Walther R, Arribas-López JR, Prados-Sánchez C. Impact and characteristics of hospital-acquired influenza over 10 seasons in a third-level university hospital. ENFERMEDADES INFECCIOSAS Y MICROBIOLOGIA CLINICA (ENGLISH ED.) 2022:S2529-993X(22)00173-3. [PMID: 36064786 DOI: 10.1016/j.eimce.2022.08.005] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 09/14/2021] [Revised: 10/27/2021] [Accepted: 11/03/2021] [Indexed: 06/15/2023]
Abstract
OBJECTIVES To analyze the characteristics of patients with nosocomial flu, to compare them with patients with community-acquired influenza to study possible differences and to identify possible risk factors associated with this type of flu. PATIENTS AND METHODS Observational, cross-sectional and retrospective study of hospitalized patients with a microbiological confirmation of influenza in a third-level university hospital over 10 seasons, from 2009 to 2019. Nosocomial influenza was defined as that infection whose symptoms began 72h after hospital admission, and its incidence, characteristics and consequences were further analyzed. RESULTS A total of 1260 hospitalized patients with a microbiological diagnosis of influenza were included, which 110 (8.7%) were nosocomial. Patients with hospital-acquired influenza were younger (71.74±16.03 years, P=0.044), had a longer hospital stay (24.25±20.25 days, P<0.001), had more frequently a history of chronic pulmonary pathologies (P=0.010), immunodeficiency (P<0.001), and were associated with greater development of bacterial superinfection (P<0.001), respiratory distress (P=0.003), and admission to the intensive care unit (ICU) (P<0.001). In the multivariate logistic regression analysis, the following characteristics were identified as independent risk factors: immunodeficiency (ORa=2.33; 95% CI: 1.47-3.60); ICU admission (ORa=4.29; 95% CI: 2.23-10.91); bacterial superinfection (ORa=1.64; 95% CI: 1.06-2.53) and respiratory distress (ORa=3.88; 95% CI: 1.23-12.23). CONCLUSIONS Nosocomial influenza is more common in patients with a history of immunodeficiency. In addition, patients with hospital-acquired influenza had an increased risk of bacterial superinfection, admission to the ICU, and development of respiratory distress.
Collapse
Affiliation(s)
- Alberto Mangas-Moro
- Servicio de Neumología, Hospital Universitario La Paz, Universidad Autónoma de Madrid, IdiPAZ, Madrid, Spain.
| | - Ester Zamarrón-de-Lucas
- Servicio de Neumología, Hospital Universitario La Paz, Universidad Autónoma de Madrid, IdiPAZ, Madrid, Spain
| | - Carlos Javier Carpio-Segura
- Servicio de Neumología, Hospital Universitario La Paz, Universidad Autónoma de Madrid, IdiPAZ, Madrid, Spain
| | | | - José Ramón Arribas-López
- Servicio de Medicina Interna, Unidad de Enfermedades Infecciosas, Universidad Autónoma de Madrid, IdiPAZ, Hospital Universitario La Paz, Madrid, Spain
| | - Concepción Prados-Sánchez
- Servicio de Neumología, Hospital Universitario La Paz, Universidad Autónoma de Madrid, IdiPAZ, Madrid, Spain
| |
Collapse
|
6
|
Mangas-Moro A, Zamarrón-de-Lucas E, Carpio-Segura CJ, Álvarez-Sala-Walther R, Arribas-López JR, Prados-Sánchez C. Impacto y características de la gripe nosocomial a lo largo de 10 temporadas en un hospital universitario de tercer nivel. Enferm Infecc Microbiol Clin 2021. [DOI: 10.1016/j.eimc.2021.11.005] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
|
7
|
Zhang B, Wu XL, Li R. A Meta-Analysis on Evaluation of Nosocomial Infections Amongst Patients in a Tertiary Care Hospital. JOURNAL OF HEALTHCARE ENGINEERING 2021; 2021:4386423. [PMID: 34630986 PMCID: PMC8500752 DOI: 10.1155/2021/4386423] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 07/26/2021] [Revised: 08/22/2021] [Accepted: 09/13/2021] [Indexed: 11/18/2022]
Abstract
Background Hospital-acquired infections, also known as nosocomial infections, are one of the many severe outcomes amongst patients in tertiary care hospitals. Hospital-acquired influenza is amongst the most common infection which has affected huge population. Objective We have performed a meta-analysis in order to summarize the effects of epidemiology and clinical characteristics in HAI. Methods We performed literature review with help of PubMed, Cochrane Library, Embase, Scopus, Web of Science, China National Knowledge Infrastructure (CNKI), The Global Index Medicus (GIM), and other clinical databases till 2021. Many random models were used in order to obtain pooled proportions, mean difference, odds ratio, and CI. Results A total of six studies were analyzed, where a total of 491 nosocomial and 4030 nonnosocomial infection cases were reported. The odds ratio of mortality was 0.02 with 95% CI and the risk ration for males was 1.08 with 95% CI. Conclusion The proportion of nosocomial infections in cases of influenza was higher in patients admitted in tertiary care hospitals. Thus, a surveillance system for vaccination for all the high-risk patients must be made mandatory.
Collapse
Affiliation(s)
- Baozhi Zhang
- The Second Affiliated Hospital Of Guangdong Medical University, Zhanjiang, Guangdong, China
| | - Xiao Ling Wu
- The Second Affiliated Hospital Of Guangdong Medical University, Zhanjiang, Guangdong, China
| | - Ruiping Li
- The Second Affiliated Hospital Of Guangdong Medical University, Zhanjiang, Guangdong, China
| |
Collapse
|
8
|
Li Y, Wang LL, Xie LL, Hou WL, Liu XY, Yin S. The epidemiological and clinical characteristics of the hospital-acquired influenza infections: A systematic review and meta-analysis. Medicine (Baltimore) 2021; 100:e25142. [PMID: 33725996 PMCID: PMC7982188 DOI: 10.1097/md.0000000000025142] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/08/2020] [Accepted: 02/21/2021] [Indexed: 01/05/2023] Open
Abstract
BACKGROUND The hospital-acquired influenza (HAI) were usually contributed to severe outcomes among the inpatients. Here, we performed a meta-analysis to summarize and quantify the epidemiological and clinical characteristics of HAI. METHODS We performed a literature search thorough PubMed, Web of Science, Cochrane Library, Embase, Scopus and China National Knowledge Infrastructure (CNKI), and Wanfang databases for observational studies. Random/fix-effects models were used to obtain pooled proportion, odds ratio (OR), and weighted mean difference (WMD). RESULTS A total of 14 studies involving 1483 HAI and 71849 non-hospital-acquired influenza infections (NHAI) cases were included.The proportion of the HAI among the influenza cases was 11.38% (95% confidence interval [CI]: 5.19%-19.55%) and it was increased after 2012 (6.15% vs 12.72%). The HAI cases were significantly older (WMD = 9.51, 95% CI: 0.04-18.98) and the patients with chronic medical diseases were at increased risk of HAI (OR = 1.85, 95% CI: 1.57-2.19). Among them, metabolic disorders (OR = 8.10, 95% CI: 2.46-26.64) ranked the highest danger, followed by malignancy (OR = 3.18, 95% CI: 2.12-4.76), any chronic diseases (OR = 2.81, 95% CI: 1.08-9.31), immunosuppression (OR = 2.13, 95% CI: 1.25-3.64), renal diseases (OR = 1.72, 95% CI:1.40-2.10), heart diseases (OR = 1.52, 95% CI: 1.03-1.44), and diabetes (OR = 1.22, 95% CI: 1.03-1.44). The HAI cases were more likely to experience longer hospital stay (WMD = 10.23, 95% CI: 4.60-15.85) and longer intensive care unit (ICU) stay (WMD = 2.99, 95% CI: 1.50-4.48). In the outcomes within 30 days, those population was still more likely to receive hospitalization (OR = 6.55, 95% CI: 5.19-8.27), death in hospital (OR = 1.99, 95% CI: 1.65-2.40) but less likely to discharged (OR = 0.20, 95% CI: 0.16-0.24). CONCLUSION The proportion of the HAI among the influenza cases was relatively high. Reinforcement of the surveillance systems and vaccination of the high-risk patients and their contacts are necessary for the HAI control.
Collapse
Affiliation(s)
- Yi Li
- Division of Life Sciences and Medicine, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, PR China
| | | | | | | | | | | |
Collapse
|