RESUMO
ObjectiveFor multi-center heterogeneous Real-World Data (RWD) with time-to-event outcomes and high-dimensional features, we propose the SurvMaximin algorithm to estimate Cox model feature coefficients for a target population by borrowing summary information from a set of health care centers without sharing patient-level information. Materials and MethodsFor each of the centers from which we want to borrow information to improve the prediction performance for the target population, a penalized Cox model is fitted to estimate feature coefficients for the center. Using estimated feature coefficients and the covariance matrix of the target population, we then obtain a SurvMaximin estimated set of feature coefficients for the target population. The target population can be an entire cohort comprised of all centers, corresponding to federated learning, or can be a single center, corresponding to transfer learning. ResultsSimulation studies and a real-world international electronic health records application study, with 15 participating health care centers across three countries (France, Germany, and the U.S.), show that the proposed SurvMaximin algorithm achieves comparable or higher accuracy compared with the estimator using only the information of the target site and other existing methods. The SurvMaximin estimator is robust to variations in sample sizes and estimated feature coefficients between centers, which amounts to significantly improved estimates for target sites with fewer observations. ConclusionsThe SurvMaximin method is well suited for both federated and transfer learning in the high-dimensional survival analysis setting. SurvMaximin only requires a one-time summary information exchange from participating centers. Estimated regression vectors can be very heterogeneous. SurvMaximin provides robust Cox feature coefficient estimates without outcome information in the target population and is privacy-preserving.
RESUMO
ObjectivesTo perform an international comparison of the trajectory of laboratory values among hospitalized patients with COVID-19 who develop severe disease and identify optimal timing of laboratory value collection to predict severity across hospitals and regions. DesignRetrospective cohort study. SettingThe Consortium for Clinical Characterization of COVID-19 by EHR (4CE), an international multi-site data-sharing collaborative of 342 hospitals in the US and in Europe. ParticipantsPatients hospitalized with COVID-19, admitted before or after PCR-confirmed result for SARS-CoV-2. Primary and secondary outcome measuresPatients were categorized as "ever-severe" or "never-severe" using the validated 4CE severity criteria. Eighteen laboratory tests associated with poor COVID-19-related outcomes were evaluated for predictive accuracy by area under the curve (AUC), compared between the severity categories. Subgroup analysis was performed to validate a subset of laboratory values as predictive of severity against a published algorithm. A subset of laboratory values (CRP, albumin, LDH, neutrophil count, D-dimer, and procalcitonin) was compared between North American and European sites for severity prediction. ResultsOf 36,447 patients with COVID-19, 19,953 (43.7%) were categorized as ever-severe. Most patients (78.7%) were 50 years of age or older and male (60.5%). Longitudinal trajectories of CRP, albumin, LDH, neutrophil count, D-dimer, and procalcitonin showed association with disease severity. Significant differences of laboratory values at admission were found between the two groups. With the exception of D-dimer, predictive discrimination of laboratory values did not improve after admission. Sub-group analysis using age, D-dimer, CRP, and lymphocyte count as predictive of severity at admission showed similar discrimination to a published algorithm (AUC=0.88 and 0.91, respectively). Both models deteriorated in predictive accuracy as the disease progressed. On average, no difference in severity prediction was found between North American and European sites. ConclusionsLaboratory test values at admission can be used to predict severity in patients with COVID-19. Prediction models show consistency across international sites highlighting the potential generalizability of these models.
RESUMO
ISARIC (International Severe Acute Respiratory and emerging Infections Consortium) partnerships and outbreak preparedness initiatives enabled the rapid launch of standardised clinical data collection on COVID-19 in Jan 2020. Extensive global participation has resulted in a large, standardised collection of comprehensive clinical data from hundreds of sites across dozens of countries. Data are analysed regularly and reported publicly to inform patient care and public health response. This report, our 17th report, is a part of a series published over the past 2 years. Data have been entered for 800,459 individuals from 1701 partner institutions and networks across 60 countries. The comprehensive analyses detailed in this report includes hospitalised individuals of all ages for whom data collection occurred between 30 January 2020 and up to and including 5 January 2022, AND who have laboratory-confirmed SARS-COV-2 infection or clinically diagnosed COVID-19. For the 699,014 cases who meet eligibility criteria for this report, selected findings include: O_LImedian age of 58 years, with an approximately equal (50/50) male:female sex distribution C_LIO_LI29% of the cohort are at least 70 years of age, whereas 4% are 0-19 years of age C_LIO_LIthe most common symptom combination in this hospitalised cohort is shortness of breath, cough, and history of fever, which has remained constant over time C_LIO_LIthe five most common symptoms at admission were shortness of breath, cough, history of fever, fatigue/malaise, and altered consciousness/confusion, which is unchanged from the previous reports C_LIO_LIage-associated differences in symptoms are evident, including the frequency of altered consciousness increasing with age, and fever, respiratory and constitutional symptoms being present mostly in those 40 years and above C_LIO_LI16% of patients with relevant data available were admitted at some point during their illness into an intensive care unit (ICU), which is slightly lower than previously reported (19%) C_LIO_LIantibiotic agents were used in 35% of patients for whom relevant data are available (669,630), a significant reduction from our previous reports (80%) which reflects a shifting proportion of data contributed by different institutions; in ICU/HDU admitted patients with data available (50,560), 91% received antibiotics C_LIO_LIuse of corticosteroids was reported in 24% of all patients for whom data were available (677,012); in ICU/HDU admitted patients with data available (50,646), 69% received corticosteroids C_LIO_LIoutcomes are known for 632,518 patients and the overall estimated case fatality ratio (CFR) is 23.9% (95%CI 23.8-24.1), rising to 37.1% (95%CI 36.8-37.4) for patients who were admitted to ICU/HDU, demonstrating worse outcomes in those with the most severe disease C_LI To access previous versions of ISARIC COVID-19 Clinical Data Report please use the link below: https://isaric.org/research/covid-19-clinical-research-resources/evidence-reports/
