Improving prediction of COVID-19 mortality using machine learning in the Spanish SEMI-COVID-19 registry.

COVID-19 is responsible for high mortality, but robust machine learning-based predictors of mortality are lacking. To generate a model for predicting mortality in patients hospitalized with COVID-19 using Gradient Boosting Decision Trees (GBDT). The Spanish SEMI-COVID-19 registry includes 24,514 pseudo-anonymized cases of patients hospitalized with COVID-19 from 1 February 2020 to 5 December 2021. This registry was used as a GBDT machine learning model, employing the CatBoost and BorutaShap classifier to select the most relevant indicators and generate a mortality prediction model by risk level, ranging from 0 to 1. The model was validated by separating patients according to admission date, using the period 1 February to 31 December 2020 (first and second waves, pre-vaccination period) for training, and 1 January to 30 November 2021 (vaccination period) for the test group. An ensemble of ten models with different random seeds was constructed, separating 80% of the patients for training and 20% from the end of the training period for cross-validation. The area under the receiver operating characteristics curve (AUC) was used as a performance metric. Clinical and laboratory data from 23,983 patients were analyzed. CatBoost mortality prediction models achieved an AUC performance of 84.76 (standard deviation 0.45) for patients in the test group (potentially vaccinated patients not included in model training) using 16 features. The performance of the 16-parameter GBDT model for predicting COVID-19 hospital mortality, although requiring a relatively large number of predictors, shows a high predictive capacity.

Clinical Features and Risk Factors for Mortality Among Long-term Care Facility Residents Hospitalized Due to COVID-19 in Spain.

BACKGROUND: COVID-19 severely impacted older adults and long-term care facility (LTCF) residents. Our primary aim was to describe differences in clinical and epidemiological variables, in-hospital management, and outcomes between LTCF residents and community-dwelling older adults hospitalized with COVID-19. The secondary aim was to identify risk factors for mortality due to COVID-19 in hospitalized LTCF residents. METHODS: This is a cross-sectional analysis within a retrospective cohort of hospitalized patients ≥75 years with confirmed COVID-19 admitted to 160 Spanish hospitals. Differences between groups and factors associated with mortality among LTCF residents were assessed through comparisons and logistic regression analysis. RESULTS: Of 6 189 patients ≥75 years, 1 185 (19.1%) were LTCF residents and 4 548 (73.5%) were community-dwelling. LTCF residents were older (median: 87.4 vs 82.1 years), mostly female (61.6% vs 43.2%), had more severe functional dependence (47.0% vs 7.8%), more comorbidities (Charlson Comorbidity Index: 6 vs 5), had dementia more often (59.1% vs 14.4%), and had shorter duration of symptoms (median: 3 vs 6 days) than community-dwelling patients (all, p < .001). Mortality risk factors in LTCF residents were severe functional dependence (adjusted odds ratios [aOR]: 1.79; 95% confidence interval [CI]: 1.13-2.83; p = .012), dyspnea (1.66; 1.16-2.39; p = .004), SatO2 < 94% (1.73; 1.27-2.37; p = .001), temperature ≥ 37.8°C (1.62; 1.11-2.38; p = .013); qSOFA index ≥ 2 (1.62; 1.11-2.38; p = .013), bilateral infiltrates (1.98; 1.24-2.98; p < .001), and high C-reactive protein (1.005; 1.003-1.007; p < .001). In-hospital mortality was initially higher among LTCF residents (43.3% vs 39.7%), but lower after adjusting for sex, age, functional dependence, and comorbidities (aOR: 0.74, 95%CI: 0.62-0.87; p < .001). CONCLUSION: Basal functional status and COVID-19 severity are risk factors of mortality in LTCF residents. The lower adjusted mortality rate in LTCF residents may be explained by earlier identification, treatment, and hospitalization for COVID-19.