COVID-19 throughout pandemic waves and virus variants: a real-life experience in an Italian hospital.

New therapies and vaccines changed the management of COVID-19. The aim of this retrospective study was to describe characteristics, in-hospital mortality and its predictors in patients with moderate/severe COVID-19, considering the 4 different pandemic waves and viral variants' prevalence from February 2020 to January 2022. Among 1135 patients included, 873 (77%) had at least one comorbidity, 177 (16%) were immunocompromised. From waves 1 to 4, patients with severe respiratory failure and ICU admission decreased over time (p < 0.001), like the length of in-hospital stay (p < 0.001). Despite a reduction of in-hospital mortality from 19% to 11%, increased risk of death was related to older age and immunocompromising conditions, especially during the 4th wave (HR = 5.07 and HR = 10.86, p < 0.001 respectively) while remdesivir treatment in the 3rd wave (HR = 0.41, p = 0.010) and positive serology (aHR = 0.66, p = 0.027) were protective for survival. These data support the need for tailoring vaccine campaign for future COVID-19 waves.

Artificial intelligence and prescription of antibiotic therapy: present and future.

INTRODUCTION: In the past few years, the use of artificial intelligence in healthcare has grown exponentially. Prescription of antibiotics is not exempt from its rapid diffusion, and various machine learning (ML) techniques, from logistic regression to deep neural networks and large language models, have been explored in the literature to support decisions regarding antibiotic prescription. AREAS COVERED: In this narrative review, we discuss promises and challenges of the application of ML-based clinical decision support systems (ML-CDSSs) for antibiotic prescription. A search was conducted in PubMed up to April 2024. EXPERT OPINION: Prescribing antibiotics is a complex process involving various dynamic phases. In each of these phases, the support of ML-CDSSs has shown the potential, and also the actual ability in some studies, to favorably impacting relevant clinical outcomes. Nonetheless, before widely exploiting this massive potential, there are still crucial challenges ahead that are being intensively investigated, pertaining to the transparency of training data, the definition of the sufficient degree of prediction explanations when predictions are obtained through black box models, and the legal and ethical framework for decision responsibility whenever an antibiotic prescription is supported by ML-CDSSs.

Transforming Ontology Web Language Elements into Common Terminology Service 2 Terminology Resources.

Communication and cooperation are fundamental for the correct deployment of P5 medicine, and this can be achieved only by correct comprehension of semantics so that it can aspire to medical knowledge sharing. There is a hierarchy in the operations that need to be performed to achieve this goal that brings to the forefront the complete understanding of the real-world business system by domain experts using Domain Ontologies, and only in the last instance acknowledges the specific transformation at the pure information and communication technology level. A specific feature that should be maintained during such types of transformations is versioning that aims to record the evolution of meanings in time as well as the management of their historical evolution. The main tool used to represent ontology in computing environments is the Ontology Web Language (OWL), but it was not created for managing the evolution of meanings in time. Therefore, we tried, in this paper, to find a way to use the specific features of Common Terminology Service-Release 2 (CTS2) to perform consistent and validated transformations of ontologies written in OWL. The specific use case managed in the paper is the Alzheimer's Disease Ontology (ADO). We were able to consider all of the elements of ADO and map them with CTS2 terminological resources, except for a subset of elements such as the equivalent class derived from restrictions on other classes.

Prediction of candidemia with machine learning techniques: state of the art.

In this narrative review, we discuss studies assessing the use of machine learning (ML) models for the early diagnosis of candidemia, focusing on employed models and the related implications. There are currently few studies evaluating ML techniques for the early diagnosis of candidemia as a prediction task based on clinical and laboratory features. The use of ML tools holds promise to provide highly accurate and real-time support to clinicians for relevant therapeutic decisions at the bedside of patients with suspected candidemia. However, further research is needed in terms of sample size, data quality, recognition of biases and interpretation of model outputs by clinicians to better understand if and how these techniques could be safely adopted in daily clinical practice.

Candida is a type of fungus that can cause fatal infections. To confirm the presence of the infection, doctors may search for the fungus in the blood. Here, we discuss if computer systems can help to identify infection more easily and more rapidly.

Towards the automatic calculation of the EQUAL Candida Score: Extraction of CVC-related information from EMRs of critically ill patients with candidemia in Intensive Care Units.

OBJECTIVES: Candidemia is the most frequent invasive fungal disease and the fourth most frequent bloodstream infection in hospitalized patients. Its optimal management is crucial for improving patients' survival. The quality of candidemia management can be assessed with the EQUAL Candida Score. The objective of this work is to support its automatic calculation by extracting central venous catheter-related information from Italian text in clinical notes of electronic medical records. MATERIALS AND METHODS: The sample includes 4,787 clinical notes of 108 patients hospitalized between January 2018 to December 2020 in the Intensive Care Units of the IRCCS San Martino Polyclinic Hospital in Genoa (Italy). The devised pipeline exploits natural language processing (NLP) to produce numerical representations of clinical notes used as input of machine learning (ML) algorithms to identify CVC presence and removal. It compares the performances of (i) rule-based method, (ii) count-based method together with a ML algorithm, and (iii) a transformers-based model. RESULTS: Results, obtained with three different approaches, were evaluated in terms of weighted F1 Score. The random forest classifier showed the higher performance in both tasks reaching 82.35%. CONCLUSION: The present work constitutes a first step towards the automatic calculation of the EQUAL Candida Score from unstructured daily collected data by combining ML and NLP methods. The automatic calculation of the EQUAL Candida Score could provide crucial real-time feedback on the quality of candidemia management, aimed at further improving patients' health.

Perceptions of Animal Welfare on Livestock: Evidence from College Agronomy Students in Costa Rica.

Ethical considerations regarding our treatment of animals have gained strength, leading to legislation and a societal focus across various disciplines. This is a subject of study within curricula related to agri-food sciences. The aim was to determine the perceptions of agronomy university students concerning animal welfare in livestock production systems. A survey was conducted to encompass various aspects, from participants' sociodemographic attributes to their attitudes and behaviors regarding animal welfare and the consumption of animal products. Statistical analysis, performed using R software, delved into the associations between participants' characteristics and their perspectives on the ethical, bioethical, and legal dimensions of animal welfare. Associations between demographic factors and ethical viewpoints among students were identified. Gender differences emerged in animal treatment perceptions, while rural and urban environments impacted perspectives on various animals. Bioethical considerations revealed distinctive disparities based on gender and education in concerns regarding animal welfare, value perceptions, evaluations of animal behaviors, and opinions on animal research. It is crucial to distinguish between animal welfare and the ethical considerations arising from coexisting with sentient beings capable of experiencing suffering. Ethical theories provide a lens through which we perceive our obligations toward animals. The responsibility to ensure animal welfare is firmly rooted in recognizing that animals, like humans, experience pain and physical suffering. Consequently, actions causing unjustified suffering or mistreatment, particularly for entertainment purposes, are considered morally unacceptable.

Time-Dependent Long-Term Effect of Memantine following Repetitive Mild Traumatic Brain Injury.

Repetitive mild traumatic brain injury (rmTBI, e.g., sports concussions) may be associated with both acute and chronic symptoms and neurological changes. Despite the common occurrence of these injuries, therapeutic strategies are limited. One potentially promising approach is N-methyl-D-aspartate receptor (NMDAR) blockade to alleviate the effects of post-injury glutamatergic excitotoxicity. Initial pre-clinical work using the NMDAR antagonist, memantine, suggests that immediate treatment following rmTBI improves a variety of acute outcomes. It remains unclear (1) whether acute memantine treatment has long-term benefits and (2) whether delayed treatment following rmTBI is beneficial, which are both clinically relevant concerns. To test this, animals were subjected to rmTBI via a weight drop model with rotational acceleration (five hits in 5 days) and randomized to memantine treatment immediately, 3 months, or 6 months post-injury, with a treatment duration of one month. Behavioral outcomes were assessed at 1, 4, and 7 months post-injury. Neuropathological outcomes were characterized at 7 months post-injury. We observed chronic changes in behavior (anxiety-like behavior, motor coordination, spatial learning, and memory), as well as neuroinflammation (microglia, astrocytes) and tau phosphorylation (T231). Memantine treatment, either immediately or 6 months post-injury, appears to confer greater rescue of neuroinflammatory changes (microglia) than vehicle or treatment at the 3-month time point. Although memantine is already being prescribed chronically to address persistent symptoms associated with rmTBI, this study represents the first evidence of which we are aware to suggest a small but durable effect of memantine treatment in mild, concussive injuries. This effect suggests that memantine, although potentially beneficial, is insufficient to treat all aspects of rmTBI alone and should be combined with other therapeutic agents in a multi-therapy approach, with attention given to the timing of treatment.

Weight Gain in Overweight and Obese People with HIV-The OBHIV Cohort.

BACKGROUND: HIV and non-HIV-related factors have been related to weight gain (WG); however, their specific impact on people with HIV (PWH) who are overweight or obese remains unclear. METHODS: This is a single-center observational study enrolling PWH with a BMI > 25 kg/m2. A generalized linear model was used to assess variables related to greater WG during 12 years of observation. RESULTS: A total of 321 PWH were enrolled, 67% overweight and 33% obese, who gained an average of 0.2 ± 1.3 and 1.7 ± 1.5 kg/year, respectively (p < 0.0001). Years since HIV infection were the only variable significantly associated with WG (ß -0.048, 95% CI -0.083; -0.013) during the study period, while type of ART did not influence the outcome. Narrowing the observation to the period of the SARS-CoV-2 pandemic, PWH with a longer duration of infection (ß 0.075, 95% CI 0.033; 0.117) and a greater increase in triglycerides (ß 0.005; 95% CI 0.000; 0.011) gained more weight, while higher BMI (ß -0.256, 95% CI -0.352; -0.160), obesity (ß -1.363, 95% CI -2.319; -0.408), diabetes mellitus (ß -1.538, 95% CI -2.797; -0.278), and greater abdominal circumference (ß -0.086, 95% CI -0.142; -0.030) resulted in protection. CONCLUSION: Among overweight and obese PWH, the amount of WG was higher in the first years after diagnosis of HIV and decreased thereafter, despite aging, regardless of the type of ART.

Explainable and Interpretable Machine Learning for Antimicrobial Stewardship: Opportunities and Challenges.

There is growing interest in exploiting the advances in artificial intelligence and machine learning (ML) for improving and monitoring antimicrobial prescriptions in line with antimicrobial stewardship principles. Against this background, the concepts of interpretability and explainability are becoming increasingly essential to understanding how ML algorithms could predict antimicrobial resistance or recommend specific therapeutic agents, to avoid unintended biases related to the "black box" nature of complex models. In this commentary, we review and discuss some relevant topics on the use of ML algorithms for antimicrobial stewardship interventions, highlighting opportunities and challenges, with particular attention paid to interpretability and explainability of employed models. As in other fields of medicine, the exponential growth of artificial intelligence and ML indicates the potential for improving the efficacy of antimicrobial stewardship interventions, at least in part by reducing time-consuming tasks for overwhelmed health care personnel. Improving our knowledge about how complex ML models work could help to achieve crucial advances in promoting the appropriate use of antimicrobials, as well as in preventing antimicrobial resistance selection and dissemination.

NLP-based tools for localization of the epileptogenic zone in patients with drug-resistant focal epilepsy.

Epilepsy surgery is an option for people with focal onset drug-resistant (DR) seizures but a delayed or incorrect diagnosis of epileptogenic zone (EZ) location limits its efficacy. Seizure semiological manifestations and their chronological appearance contain valuable information on the putative EZ location but their interpretation relies on extensive experience. The aim of our work is to support the localization of EZ in DR patients automatically analyzing the semiological description of seizures contained in video-EEG reports. Our sample is composed of 536 descriptions of seizures extracted from Electronic Medical Records of 122 patients. We devised numerical representations of anamnestic records and seizures descriptions, exploiting Natural Language Processing (NLP) techniques, and used them to feed Machine Learning (ML) models. We performed three binary classification tasks: localizing the EZ in the right or left hemisphere, temporal or extra-temporal, and frontal or posterior regions. Our computational pipeline reached performances above 70% in all tasks. These results show that NLP-based numerical representation combined with ML-based classification models may help in localizing the origin of the seizures relying only on seizures-related semiological text data alone. Accurate early recognition of EZ could enable a more appropriate patient management and a faster access to epilepsy surgery to potential candidates.

Early diagnosis of candidemia with explainable machine learning on automatically extracted laboratory and microbiological data: results of the AUTO-CAND project.

BACKGROUND: Candidemia is associated with a heavy burden of morbidity and mortality in hospitalized patients. The availability of blood culture results could require up to 48-72 h after blood draw; thus, early treatment decisions are made in the absence of a definite diagnosis. METHODS: In this retrospective study, we assessed the performance of different supervised machine learning algorithms for the early differential diagnosis of candidemia and bacteremia in adult patients on a large dataset automatically extracted within the AUTO-CAND project. RESULTS: Overall, 12,483 episodes of candidemia (1275; 10%) or bacteremia (11,208; 90%) were included in the analysis. A random forest classifier achieved the best diagnostic performance for candidemia, with sensitivity 0.98 and specificity 0.65 on the training set (true skill statistic [TSS] = 0.63) and sensitivity 0.74 and specificity 0.57 on the test set (TSS = 0.31). Then, the random classifier was trained in the subgroup of patients with available serum ß-D-glucan (BDG) and procalcitonin (PCT) values by exploiting the feature ranking learned in the entire dataset. Although no statistically significant differences were observed from the performance measures obtained by employing BDG and PCT alone, the performance measures of the classifier that included the features selected in the entire dataset, plus BDG and PCT, were the highest in most cases. CONCLUSIONS: Random forest classifiers trained on large datasets of automatically extracted data have the potential to improve current diagnostic algorithms for candidemia. However, further development through implementation of automatically extracted clinical features may be necessary to achieve crucial improvements.

Regular Humoral and Cellular Immune Responses in Individuals with Chronic Myeloid Leukemia Who Received a Full Vaccination Schedule against COVID-19.

Individuals with chronic myeloid leukemia (CML) constitute a unique group within individuals with oncohematological disease (OHD). They receive treatment with tyrosine kinase inhibitors (TKIs) that present immunomodulatory properties, and they may eventually be candidates for treatment discontinuation under certain conditions despite the chronic nature of the disease. In addition, these individuals present a lower risk of infection than other immunocompromised patients. For this study, we recruited a cohort of 29 individuals with CML in deep molecular response who were on treatment with TKIs (n = 23) or were on treatment-free remission (TFR) (n = 6), and compared both humoral and cellular immune responses with 20 healthy donors after receiving the complete vaccination schedule against SARS-CoV-2. All participants were followed up for 17 months to record the development of COVID-19 due to breakthrough infections. All CML individuals developed an increased humoral response, with similar seroconversion rates and neutralizing titers to healthy donors, despite the presence of high levels of immature B cells. On the whole, the cellular immune response was also comparable to that of healthy donors, although the antibody dependent cytotoxic activity (ADCC) was significantly reduced. Similar rates of mild breakthrough infections were observed between groups, although the proportion was higher in the CML individuals on TFR, most likely due to the immunomodulatory effect of these drugs. In conclusion, as with the healthy donors, the vaccination did not impede breakthrough infections completely in individuals with CML, although it prevented the development of severe or critical illness in this special population of individuals with OHD.

Defining a Preprocessing Pipeline for the MULTI-SITA Project and General Medical Italian Natural Language Data.

The application of Natural Language Processing (NLP) to medical data has revolutionized different aspects of health care. The benefits obtained from the implementation of this technique spill over into several areas, including in the implementation of chatbots, which can provide medical assistance remotely. Every possible application of NLP depends on one first main step: the pre-processing of the corpus retrieved. The raw data must be prepared with the aim to be used efficiently for further analysis. Considerable progress has been made in this direction for the English language but for other languages, such as Italian, the state of the art is not equivalently advanced, especially for texts containing technical medical terms. The aim of this work is to identify and develop a preprocessing pipeline suitable for medical data written in Italian. The pipeline has been developed in Python environment, employing Enchant, ntlk modules and Hugging Face's BERT and BART-based models. Then, it has been tested on real conversations typed between patients and physicians regarding medical questions. The algorithm has been developed within the MULTI-SITA project of the Italian Society of Anti-Infective Therapy (SITA), but shows a flexible structure that can adapt to a large variety of data.

Barriers to HCV micro-elimination in a cohort of people living with HIV (PLWH).

To achieve the World Health Organization goal of hepatitis C virus (HCV) eradication, barriers to treatment should be investigated and overcome. The aim of this study was to identify those barriers and describe the strategies adopted to achieve HCV micro-elimination in a cohort of coinfected people living with HIV (PLWH-HCV). Adult PLWH-HCV followed at our hospital with detectable serum HCV-RNA in 2018 were enrolled. After a three-year follow-up, barriers to HCV treatment were investigated and strategies to overcome them were described. Of 492 PLWH-HCV seen in 2018, 29 (5.9%) had detectable serum HCV-RNA. Eight out of 29 (27.6%) were excluded because they were already under treatment, while 2 others were excluded because they moved to other outpatient clinics. Among the remaining 19 study participants, the most common barriers to treatment were poor adherence to therapies and follow-up visits (n=9, 47%), recent HCV diagnosis awaiting proper staging (n=3, 16%) and treatment hesitancy (n=2, 10%). During the following three years, direct-acting antivirals (DAAs) treatment was completed in 11/19 (58%) cases, with achievement of sustained virological response in 100% of cases. For the remaining cases, 2/19 (10.5%) were lost to follow-up, 2/19 (10.5%) died before treatment initiation and 4/19 (21.0%) are still awaiting treatment. Despite 3 years of effort, HCV micro-elimination has not been achieved at our center. We observed that poor adherence and treatment hesitancy were the main barriers to treatment. Strategies addressing these issues need to be implemented.

Effect of SARS-CoV-2 vaccination in a vulnerable COVID-19 cohort: a real-life experience in an Italian Hospital.

Clinical trials demonstrated that SARS-CoV-2 vaccines reduce COVID-19-related mortality and morbidity. We describe the effect of vaccination on COVID-19-patients admitted at our hospital. Retrospective, single-center study conducted in Genoa, Italy, including patients ≥18years hospitalized for COVID-19 from May to December 2021. Demographical and clinical data were collected, vaccinated (group-A) and not-vaccinated (group-B) patients were compared. Impact of vaccination on mortality, ICU admission, and oxygen need was studied using Cox proportional hazards and logistic regression models after adjusting for propensity scores. Overall, 395 patients SARS-CoV-2 infected were included, of which 150 (38%) were vaccinated and 245 (62%) were not vaccinated. Patients in group-A were older, more disable, and with higher morbidity. Overall, 64 patients (16%) died within 30 days from admission, 34 in Group A (23%), and 30 in group B (12%). However, no statistically significant differences were observed (group-A versus group-B: HR 0.83, 95% CI 0.49-1.40, p = 0.483). On the other hand, vaccination was protective in terms of ICU admission (OR = 0.23, p = 0.046) and oxygen need (OR = 0.33, p = 0.008). Our study confirms that SARS-CoV-2 vaccination reduces morbidity among patients hospitalized for COVID-19. The still high mortality in our cohort of vaccinated individuals could be partially due to vulnerable conditions of our patients.

Trend over time of HIV-1 drug resistance to nonnucleoside reverse transcriptase inhibitors (NNRTIs) and their drivers: A cohort study from Antiviral Response Cohort Analysis (ARCA).

The rise of HIV-1 drug resistance to nonnucleoside reverse transcriptase inhibitors (NNRTIs) threatens the long-term success of NNRTI-based therapies. Our study aims to describe the circulation of major resistance-associated mutations (RAMs) for NNRTIs in people living with HIV (PLWH) in Italy from 2000 to 2020. We included 5982 naïves and 28 505 genotypes from 9387 treatment-experienced PLWH from the Antiviral Response Cohort Analysis (ARCA) cohort. Transmitted drug resistance (TDR) was found in 12.5% and declined from 17.3% in 2000-2003 to 10.9% in 2016-2020 (p = 0.003). Predictors of TDR were viral subtype B [vs. non-B, adjusted odds ratio (aOR) = 1.94, p < 0.001], zenith viral load (VL) (per 1 log10 higher, aOR = 0.86, p = 0.013), nadir CD4 cell count (per 100 cells/µL increase aOR = 0.95, p = 0.013). At least one RAM for NNRTIs among treatment experienced PLWH was detected in 33.2% and pre-treatment drug resistance (PDR) declined from 43.4% in 2000-2003 to 20.9% in 2016-2020 (p < 0.001). Predictors of PDR were sexual transmission route (vs. others, aOR = 0.78, p < 0.001), time since HIV diagnosis (per 1 month longer, aOR = 1.002, p < 0.001), viral subtype B (vs. non B, aOR = 1.37, p < 0.001), VL (per 1 log10 higher, aOR = 1.12, p < 0.001), nadir CD4 count (per 100 cells/µL increase, aOR = 0.91, p < 0.001), previous exposure to any NNRTI (aOR = 2.31, p < 0.001) and a more recent calendar year sequence (any time span > 2008 vs. 2000-2003, any aOR <1, p < 0.001). Circulation of RAMs to NNRTIs declined during the last 20 years in Italy. NNRTIs remain pivotal drugs for the management of HIV-1 due to safety concerns and long-acting options.

Protocol of Validation Mechanisms Within the Design of an Electronic Case Report Form for the MULTI-SITA Project.

With the wide diffusion of web technology, dedicated electronic Case Report Forms (eCRFs) became the main tool for collecting patient data. The focus of this work is to thoroughly consider the data quality in every aspect of the design of the eCRF, with the result of having multiple steps of validation that should produce a diligent and multidisciplinary approach towards every step of data acquisition. This goal affects every aspect of the system design.

Persistent Immunity against SARS-CoV-2 in Individuals with Oncohematological Diseases Who Underwent Autologous or Allogeneic Stem Cell Transplantation after Vaccination.

The high morbimortality due to SARS-CoV-2 infection in oncohematological diseases (OHD) and hematopoietic stem cell transplant (HSCT) recipients in the pre-vaccine era has made vaccination a priority in this group. After HSCT, the immune responses against common vaccines such as tetanus, varicella, rubella, and polio may be lost. However, the loss of immunity developed by COVID-19 vaccination after HSCT has not been completely defined. In this study, both humoral and cellular immunity against SARS-CoV-2 were analyzed in 29 individuals with OHD who were vaccinated before receiving allogeneic (n = 11) or autologous (n = 18) HSCT. All participants had low but protective levels of neutralizing IgGs against SARS-CoV-2 after HSCT despite B-cell lymphopenia and immaturity. Although antibody-dependent cellular cytotoxicity was impaired, direct cellular cytotoxicity was similar to healthy donors in participants with autologous-HSCT, in contrast to individuals with allogeneic-HSCT, which severely deteriorated. No significant changes were observed in the immune response before and after HSCT. During follow-up, all reported post-HSCT SARS-CoV-2 infections were mild. This data emphasizes that COVID-19 vaccination is effective, necessary, and safe for individuals with OHD and also supports the persistence of some degree of immune protection after HSCT, at least in the short term, when patients cannot yet be revaccinated.

External validation of unsupervised COVID-19 clinical phenotypes and their prognostic impact.

BACKGROUND: Hospitalized patients with coronavirus disease 2019 (COVID-19) can be classified into different clinical phenotypes based on their demographic, clinical, radiology, and laboratory features. We aimed to validate in an external cohort of hospitalized COVID-19 patients the prognostic value of a previously described phenotyping system (FEN-COVID-19) and to assess the reproducibility of phenotypes development as a secondary analysis. METHODS: Patients were classified in phenotypes A, B or C according to the severity of oxygenation impairment, inflammatory response, hemodynamic and laboratory tests according to the FEN-COVID-19 method. RESULTS: Overall, 992 patients were included in the study, and 181 (18%), 757 (76%) and 54 (6%) of them were assigned to the FEN-COVID-19 phenotypes A, B, and C, respectively. An association with mortality was observed for phenotype C vs. A (hazard ratio [HR] 3.10, 95% confidence interval [CI] 1.81-5.30, p < 0.001) and for phenotype C vs. B (HR 2.20, 95% CI 1.50-3.23, p < 0.001). A non-statistically significant trend towards higher mortality was also observed for phenotype B vs. A (HR 1.41; 95% CI 0.92-2.15, p = 0.115). By means of cluster analysis, three different phenotypes were also identified in our cohort, with an overall similar gradient in terms of prognostic impact to that observed when patients were assigned to FEN-COVID-19 phenotypes. CONCLUSIONS: The prognostic impact of FEN-COVID-19 phenotypes was confirmed in our external cohort, although with less difference in mortality between phenotypes A and B than in the original study.

Hospitalized patients with COVID-19 can be classified into different clinical phenotypes based on their demographic, clinical, radiology, and laboratory featuresIn this study, we externally confirmed the prognostic impact of clinical phenotypes previously identified by Gutierrez-Gutierrez and colleagues in a Spanish cohort of hospitalized patients with COVID-19, and the usefulness of their simplified probabilistic model for phenotypes assignmentThis could indirectly support the validity of both phenotype's development and their extrapolation to other hospitals and countries for management decisions during other possible future viral pandemics.