The coronavirus disease (COVID-19) pandemic has increased the necessity of immediate clinical decisions and effective usage of healthcare resources. Currently, the most validated diagnosis test for COVID-19 (RT-PCR) is in shortage in most developing countries, which may increase infection rates and delay important preventive measures. The objective of this study was to predict the risk of positive COVID-19 diagnosis with machine learning, using as predictors only results from emergency care admission exams. We collected data from 235 adult patients from the Hospital Israelita Albert Einstein in São Paulo, Brazil, from 17 to 30 of March, 2020, of which 102 (43%) received a positive diagnosis of COVID-19 from RT-PCR tests. Five machine learning algorithms (neural networks, random forests, gradient boosting trees, logistic regression and support vector machines) were trained on a random sample of 70% of the patients, and performance was tested on new unseen data (30%). The best predictive performance was obtained by the support vector machines algorithm (AUC: 0.85; Sensitivity: 0.68; Specificity: 0.85; Brier Score: 0.16). The three most important variables for the predictive performance of the algorithm were the number of lymphocytes, leukocytes and eosinophils, respectively. In conclusion, we found that targeted decisions for receiving COVID-19 tests using only routinely-collected data is a promising new area with the use of machine learning algorithms.

Objective.: To assess the immunogenicity and safety of non-adjuvanted influenza A H1N1/2009 vaccine in patients with juvenile autoimmune rheumatic disease (ARD) and healthy controls, because data are limited to the adult rheumatologic population.Methods.: A total of 237 patients with juvenile ARD [juvenile systemic lupus erythematosus (JSLE), juvenile idiopathic arthritis (JIA), juvenile dermatomyositis (JDM), juvenile scleroderma, and vasculitis] and 91 healthy controls were vaccinated. Serology for anti-H1N1 was performed by hemagglutination inhibition assay. Seroprotection rate, seroconversion rate, and factor-increase in geometric mean titer (GMT) were calculated. Adverse events were evaluated.Results.: Age was comparable in patients and controls (14.8 ± 3.0 vs 14.6 ± 3.7 years, respectively; p = 0.47). Three weeks after immunization, seroprotection rate (81.4% vs 95.6%; p = 0.0007), seroconversion rate (74.3 vs 95.6%; p < 0.0001), and the factor-increase in GMT (12.9 vs 20.3; p = 0.012) were significantly lower in patients with juvenile ARD versus controls. Subgroup analysis revealed reduced seroconversion rates in JSLE (p < 0.0001), JIA (p = 0.008), JDM (p = 0.025), and vasculitis (p = 0.017). Seroprotection (p < 0.0001) and GMT (p < 0.0001) were decreased only in JSLE. Glucocorticoid use and lymphopenia were associated with lower seroconversion rates (60.4 vs 82.9%; p = 0.0001; and 55.6 vs 77.2%; p = 0.012). Multivariate logistic regression including diseases, lymphopenia, glucocorticoid, and immunosuppressants demonstrated that only glucocorticoid use (p = 0.012) remained significant.Conclusion.: This is the largest study to demonstrate a reduced but adequate immune response to H1N1 vaccine in patients with juvenile ARD. It identified current glucocorticoid use as the major factor for decreased antibody production. The short-term safety results support its routine recommendation for patients with juvenile ARD. ClinicalTrials.gov; NCT01151644.

Objective: To determine risk factors for the development of long coronavirus disease 2019 (COVID-19) in healthcare personnel (HCP).Methods: We conducted a case–control study among HCP who had confirmed symptomatic COVID-19 working in a Brazilian healthcare system between March 1, 2020, and July 15, 2022. Cases were defined as those having long COVID according to the Centers for Disease Control and Prevention definition. Controls were defined as HCP who had documented COVID-19 but did not develop long COVID. Multiple logistic regression was used to assess the association between exposure variables and long COVID during 180 days of follow-up.Results: Of 7,051 HCP diagnosed with COVID-19, 1,933 (27.4%) who developed long COVID were compared to 5,118 (72.6%) who did not. The majority of those with long COVID (51.8%) had 3 or more symptoms. Factors associated with the development of long COVID were female sex (OR, 1.21; 95% CI, 1.05–1.39), age (OR, 1.01; 95% CI, 1.00–1.02), and 2 or more SARS-CoV-2 infections (OR, 1.27; 95% CI, 1.07–1.50). Those infected with the SARS-CoV-2 δ (delta) variant (OR, 0.30; 95% CI, 0.17–0.50) or the SARS-CoV-2 o (omicron) variant (OR, 0.49; 95% CI, 0.30–0.78), and those receiving 4 COVID-19 vaccine doses prior to infection (OR, 0.05; 95% CI, 0.01–0.19) were significantly less likely to develop long COVID.Conclusions: Long COVID can be prevalent among HCP. Acquiring >1 SARS-CoV-2 infection was a major risk factor for long COVID, while maintenance of immunity via vaccination was highly protective.