Comparative Study of Vaccinated and Unvaccinated Hospitalised Patients: A Retrospective Population Study of 500 Hospitalised Patients with SARS-CoV-2 Infection in a Spanish Population of 220,000 Inhabitants.

OBJECTIVES: This study aimed to compare the characteristics of fully and partially vaccinated or unvaccinated coronavirus disease 2019 (COVID-19) patients who were hospitalised in a population of 220,000 habitants. METHODS: Retrospective, observational, and population studies were conducted on patients who were hospitalised due to COVID-19 from March to October 2021. We assessed the impact of vaccination and other risk factors through Cox multivariate analysis. RESULTS: A total of 500 patients were hospitalised, among whom 77 (15.4%) were fully vaccinated, 86 (17.2%) were partially vaccinated, and 337 (67.4%) were unvaccinated. Fully vaccinated (FV) patients were older and had a higher Charlson index than those of partially vaccinated and unvaccinated patients (NFV). Bilateral pneumonia was more frequent among NFV (259/376 (68.9%)) than among FV patients (32/75 (42.7%)). The former had more intensive care unit admissions (63/423) than the latter (4/77); OR: 2.80; CI (1.07-9.47). Increasing age HZ: 1.1 (1.06-1.14)) and haematological disease at admission HZ: 2.99 (1.26-7.11)) were independent risk factors for higher mortality during the first 30 days of hospitalisation. The probability of an earlier discharge in the subgroup of 440 patients who did not die during the first 30 days of hospitalisation was related to age (older to younger: HZ: 0.98 (0.97-0.99)) and vaccination status. CONCLUSIONS: Among the patients hospitalised because of COVID-19, complete vaccination was associated with less severe forms of COVID-19, with an earlier discharge date. Age and haematological disease were related to a higher mortality rate during the first 30 days of hospitalisation.

Prognostic factors and combined use of tocilizumab and corticosteroids in a Spanish cohort of elderly COVID-19 patients.

Coronavirus disease 2019 (COVID-19) infection in elderly patients is more aggressive and treatments have shown limited efficacy. Our objective is to describe the clinical course and to analyze the prognostic factors associated with a higher risk of mortality of a cohort of patients older than 80 years. In addition, we assess the efficacy of immunosuppressive treatments in this population. We analyzed the data from 163 patients older than 80 years admitted to our institution for COVID-19, during March and April 2020. A Lasso regression model and subsequent multivariate Cox regression were performed to select variables predictive of death. We evaluated the efficacy of immunomodulatory therapy in three cohorts using adjusted survival analysis. The mortality rate was 43%. The mean age was 85.2 years. The disease was considered severe in 76.1% of the cases. Lasso regression and multivariate Cox regression indicated that factors correlated with hospital mortality were: age (hazard ratio [HR] 1.12, 95% confidence interval [CI]: 1.03-1.22), alcohol consumption (HR 3.15, 95% CI: 1.27-7.84), CRP > 10 mg/dL (HR 2.67, 95% CI: 1.36-5.24), and oxygen support with Venturi Mask (HR 6.37, 95% CI: 2.18-18.62) or reservoir (HR 7.87, 95% CI: 3.37-18.38). Previous treatment with antiplatelets was the only protective factor (HR 0.47, 95% CI: 0.23-0.96). In the adjusted treatment efficacy analysis, we found benefit in the combined use of tocilizumab (TCZ) and corticosteroids (CS) (HR 0.09, 95% CI: 0.01-0.74) compared to standard treatment, with no benefit of CS alone (HR 0.95, 95% CI: 0.53-1.71). Hospitalized elderly patients suffer from a severe and often fatal form of COVID-19 disease. In this regard, several parameters might identify high-risk patients upon admission. Combined use of TCZ and CS could improve survival.

Diagnosing hospital bacteraemia in the framework of predictive, preventive and personalised medicine using electronic health records and machine learning classifiers.

Background: The bacteraemia prediction is relevant because sepsis is one of the most important causes of morbidity and mortality. Bacteraemia prognosis primarily depends on a rapid diagnosis. The bacteraemia prediction would shorten up to 6 days the diagnosis, and, in conjunction with individual patient variables, should be considered to start the early administration of personalised antibiotic treatment and medical services, the election of specific diagnostic techniques and the determination of additional treatments, such as surgery, that would prevent subsequent complications. Machine learning techniques could help physicians make these informed decisions by predicting bacteraemia using the data already available in electronic hospital records. Objective: This study presents the application of machine learning techniques to these records to predict the blood culture's outcome, which would reduce the lag in starting a personalised antibiotic treatment and the medical costs associated with erroneous treatments due to conservative assumptions about blood culture outcomes. Methods: Six supervised classifiers were created using three machine learning techniques, Support Vector Machine, Random Forest and K-Nearest Neighbours, on the electronic health records of hospital patients. The best approach to handle missing data was chosen and, for each machine learning technique, two classification models were created: the first uses the features known at the time of blood extraction, whereas the second uses four extra features revealed during the blood culture. Results: The six classifiers were trained and tested using a dataset of 4357 patients with 117 features per patient. The models obtain predictions that, for the best case, are up to a state-of-the-art accuracy of 85.9%, a sensitivity of 87.4% and an AUC of 0.93. Conclusions: Our results provide cutting-edge metrics of interest in predictive medical models with values that exceed the medical practice threshold and previous results in the literature using classical modelling techniques in specific types of bacteraemia. Additionally, the consistency of results is reasserted because the three classifiers' importance ranking shows similar features that coincide with those that physicians use in their manual heuristics. Therefore, the efficacy of these machine learning techniques confirms their viability to assist in the aims of predictive and personalised medicine once the disease presents bacteraemia-compatible symptoms and to assist in improving the healthcare economy.