Prediction of hospital-acquired influenza using machine learning algorithms: a comparative study.

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.

COVID-19 vaccination intention among nursing, medical, and dental students: A systematic review and meta-regression analysis.

The introduction of the COVID-19 vaccine amidst the pandemic has heralded a paradigm shift. Healthcare students in nursing, medicine, and dentistry must have positive attitudes owing to their future role in vaccine recommendations to the public and patients. This meta-regression analysis assessed the differences in COVID-19 vaccination intention (VI) of nursing, medical, and dental students. Medline/PubMed, EMBASE, CINAHL, Cochrane Library, and Korean MBASE were searched for eligible studies. Quality was assessed by the Joanna Briggs Institute's quality appraisal. Forty-one studies were included in the final analysis, and the estimation of pooled prevalence (68%) and relevant factors were assessed. Dental students were found to have the lowest VI (57%), which significantly (p = .018) differed from that of nursing students (64%). Countries in South-East Asia and those with high income levels were found to have high VI. Therefore, global vaccine programs for healthcare students should be prepared considering their geographical and economic status.

Incidence and risk factors of postoperative nausea and vomiting in patients with fentanyl-based intravenous patient-controlled analgesia and single antiemetic prophylaxis.

PURPOSE: We evaluated the incidence and risk factors of postoperative nausea and vomiting (PONV) in patients with fentanyl-based intravenous patient-controlled analgesia (IV-PCA) and single antiemetic prophylaxis of 5-hydroxytryptamine type 3 (5 HT3)-receptor antagonist after the general anesthesia. MATERIALS AND METHODS: In this retrospective study, incidence and risk factors for PONV were evaluated with fentanyl IV-PCA during postoperative 48 hours after various surgeries. RESULTS: Four hundred-forty patients (23%) of 1878 had showed PONV. PCA was discontinued temporarily in 268 patients (14%), mostly due to PONV (88% of 268 patients). In multivariate analysis, female, non-smoker, history of motion sickness or PONV, long duration of anesthesia (>180 min), use of desflurane and intraoperative remifentanil infusion were independent risk factors for PONV. If one, two, three, four, five, or six of these risk factors were present, the incidences of PONV were 18%, 19%, 22%, 31%, 42%, or 50%. Laparoscopic surgery and higher dose of fentanyl were not risk factors for PONV. CONCLUSION: Despite antiemetic prophylaxis with 5 HT3-receptor antagonist, 23% of patients with fentanyl-based IV-PCA after general anesthesia showed PONV. Long duration of anesthesia and use of desflurane were identified as risk factors, in addition to risk factors of Apfel's score (female, non-smoker, history of motion sickness or PONV). Also, intraoperative remifentanil infusion was risk factor independent of postoperative opioid use. As the incidence of PONV was up to 50% according to the number of risk factors, risk-adapted, multimodal or combination therapy should be applied.

Incidence and risk factors of acute kidney injury after thoracic aortic surgery for acute dissection.

BACKGROUND: Previous studies have reported a high incidence of acute kidney injury (AKI) after thoracic aortic surgery in heterogeneous patient cohorts, including various aortic diseases and the use of deep hypothermic circulatory arrest. Moderate hypothermia with cerebral perfusion makes deep hypothermia nonessential, but can make end organs susceptible to ischemia during circulatory arrest. We investigated the incidence and risk factors of AKI after thoracic aortic surgery with and without moderate hypothermic circulatory arrest for acute dissection. METHODS: We reviewed the medical records of 98 patients undergoing graft replacement of the thoracic aorta for acute dissection between 2008 and 2011 at a university hospital. Acute kidney injury was defined by RIFLE criteria, which is based on serum creatinine or glomerular filtration rate. RESULTS: The mean age was 55±15 years. The surgical procedures, 96% of which were emergencies, involved the ascending aorta (67%), aortic arch (41%), descending aorta (41%), and aortic valve (5%). Moderate hypothermic circulatory arrest was performed in 75%. The overall incidence of AKI was 54%, and 11% of 98 patients required renal replacement therapy. Thirty-day mortality increased with AKI severity (p=0.002). Independent risk factors for AKI were long cardiopulmonary bypass duration (>180 minutes; odds ratio, 7.50; p=0.008) and preoperative serum creatinine level (odds ratio, 8.43; p=0.016). CONCLUSIONS: Acute kidney injury was common after thoracic aortic surgery for acute dissection with or without moderate hypothermic circulatory arrest and worsened 30-day mortality. Prolonged cardiopulmonary bypass and increased preoperative serum creatinine were independent risk factors for AKI, but moderate hypothermic circulatory arrest was not.