Dataset Information

Clinical subphenotypes in COVID-19: derivation, validation, prediction, temporal patterns, and interaction with social determinants of health.

ABSTRACT: The coronavirus disease 2019 (COVID-19) is heterogeneous and our understanding of the biological mechanisms of host response to the viral infection remains limited. Identification of meaningful clinical subphenotypes may benefit pathophysiological study, clinical practice, and clinical trials. Here, our aim was to derive and validate COVID-19 subphenotypes using machine learning and routinely collected clinical data, assess temporal patterns of these subphenotypes during the pandemic course, and examine their interaction with social determinants of health (SDoH). We retrospectively analyzed 14418 COVID-19 patients in five major medical centers in New York City (NYC), between March 1 and June 12, 2020. Using clustering analysis, 4 biologically distinct subphenotypes were derived in the development cohort (N = 8199). Importantly, the identified subphenotypes were highly predictive of clinical outcomes (especially 60-day mortality). Sensitivity analyses in the development cohort, and rederivation and prediction in the internal (N = 3519) and external (N = 3519) validation cohorts confirmed the reproducibility and usability of the subphenotypes. Further analyses showed varying subphenotype prevalence across the peak of the outbreak in NYC. We also found that SDoH specifically influenced mortality outcome in Subphenotype IV, which is associated with older age, worse clinical manifestation, and high comorbidity burden. Our findings may lead to a better understanding of how COVID-19 causes disease in different populations and potentially benefit clinical trial development. The temporal patterns and SDoH implications of the subphenotypes may add insights to health policy to reduce social disparity in the pandemic.

SUBMITTER: Su C

PROVIDER: S-EPMC8280198 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundAccurate assessment of COVID-19 severity in the community is essential for patient care and requires COVID-19-specific risk prediction scores adequately validated in a community setting. Following a qualitative phase to identify signs, symptoms, and risk factors, we aimed to develop and validate two COVID-19-specific risk prediction scores. Remote COVID-19 Assessment in Primary Care-General Practice score (RECAP-GP; without peripheral oxygen saturation [SpO2]) and RECAP-oxygen saturation score (RECAP-O2; with SpO2).MethodsRECAP was a prospective cohort study that used multivariable logistic regression. Data on signs and symptoms (predictors) of disease were collected from community-based patients with suspected COVID-19 via primary care electronic health records and linked with secondary data on hospital admission (outcome) within 28 days of symptom onset. Data sources for RECAP-GP were Oxford-Royal College of General Practitioners Research and Surveillance Centre (RCGP-RSC) primary care practices (development set), northwest London primary care practices (validation set), and the NHS COVID-19 Clinical Assessment Service (CCAS; validation set). The data source for RECAP-O2 was the Doctaly Assist platform (development set and validation set in subsequent sample). The two probabilistic risk prediction models were built by backwards elimination using the development sets and validated by application to the validation datasets. Estimated sample size per model, including the development and validation sets was 2880 people.FindingsData were available from 8311 individuals. Observations, such as SpO2, were mostly missing in the northwest London, RCGP-RSC, and CCAS data; however, SpO2 was available for 1364 (70·0%) of 1948 patients who used Doctaly. In the final predictive models, RECAP-GP (n=1863) included sex (male and female), age (years), degree of breathlessness (three point scale), temperature symptoms (two point scale), and presence of hypertension (yes or no); the area under the curve was 0·80 (95% CI 0·76-0·85) and on validation the negative predictive value of a low risk designation was 99% (95% CI 98·1-99·2; 1435 of 1453). RECAP-O2 included age (years), degree of breathlessness (two point scale), fatigue (two point scale), and SpO2 at rest (as a percentage); the area under the curve was 0·84 (0·78-0·90) and on validation the negative predictive value of low risk designation was 99% (95% CI 98·9-99·7; 1176 of 1183).InterpretationBoth RECAP models are valid tools to assess COVID-19 patients in the community. RECAP-GP can be used initially, without need for observations, to identify patients who require monitoring. If the patient is monitored and SpO2 is available, RECAP-O2 is useful to assess the need for treatment escalation.FundingCommunity Jameel and the Imperial College President's Excellence Fund, the Economic and Social Research Council, UK Research and Innovation, and Health Data Research UK.

Project description:BackgroundTypically observed at 2 y after surgical resection, late recurrence is a major challenge in the management of hepatocellular carcinoma (HCC). We aimed to develop a genomic predictor that can identify patients at high risk for late recurrence and assess its clinical implications.Methods and findingsSystematic analysis of gene expression data from human liver undergoing hepatic injury and regeneration revealed a 233-gene signature that was significantly associated with late recurrence of HCC. Using this signature, we developed a prognostic predictor that can identify patients at high risk of late recurrence, and tested and validated the robustness of the predictor in patients (n = 396) who underwent surgery between 1990 and 2011 at four centers (210 recurrences during a median of 3.7 y of follow-up). In multivariate analysis, this signature was the strongest risk factor for late recurrence (hazard ratio, 2.2; 95% confidence interval, 1.3-3.7; p = 0.002). In contrast, our previously developed tumor-derived 65-gene risk score was significantly associated with early recurrence (p = 0.005) but not with late recurrence (p = 0.7). In multivariate analysis, the 65-gene risk score was the strongest risk factor for very early recurrence (<1 y after surgical resection) (hazard ratio, 1.7; 95% confidence interval, 1.1-2.6; p = 0.01). The potential significance of STAT3 activation in late recurrence was predicted by gene network analysis and validated later. We also developed and validated 4- and 20-gene predictors from the full 233-gene predictor. The main limitation of the study is that most of the patients in our study were hepatitis B virus-positive. Further investigations are needed to test our prediction models in patients with different etiologies of HCC, such as hepatitis C virus.ConclusionsTwo independently developed predictors reflected well the differences between early and late recurrence of HCC at the molecular level and provided new biomarkers for risk stratification. Please see later in the article for the Editors' Summary.

Project description:BackgroundInterstitial lung disease (ILD) results in high morbidity and health-care utilization. Diagnostic delays remain common and often occur in nonpulmonology settings. Screening for ILD in these settings has the potential to reduce diagnostic delays and improve patient outcomes.Research questionThis study sought to determine whether a pulmonary function test (PFT)-derived diagnostic prediction tool (ILD-Screen) could accurately identify incident ILD cases in patients undergoing PFT in nonpulmonology settings.Study design and methodsClinical and physiologic PFT variables predictive of ILD were identified by using iterative multivariable logistic regression models. ILD status was determined by using a multi-reader approach. An ILD-Screen score was generated by using final regression model coefficients, with a score ≥ 8 considered positive. ILD-Screen test performance was validated in an independent external cohort and applied prospectively to PFTs over 1 year to identify incident ILD cases at our institution.ResultsVariables comprising the ILD-Screen were age, height, total lung capacity, FEV1, diffusion capacity, and PFT indication. The ILD-Screen showed consistent test performance across cohorts, with a sensitivity of 0.79 and a specificity of 0.83 when applied prospectively. A positive ILD-Screen strongly predicted ILD (OR, 18.6; 95% CI, 9.4-36.9) and outperformed common ILD clinical features, including cough, dyspnea, lung crackles, and restrictive lung physiology. Prospective ILD-Screen application resulted in a higher proportion of patients undergoing chest CT imaging compared with a historical control cohort (74% vs 56%, respectively; P = .003), with a significantly shorter median time to chest CT imaging (5.6 vs 21.1 months; P < .001).InterpretationThe ILD-Screen showed good test performance in predicting ILD across diverse geographic settings and when applied prospectively. Systematic ILD-Screen application has the potential to reduce diagnostic delays and facilitate earlier intervention in patients with ILD.

Dataset Information

Clinical subphenotypes in COVID-19: derivation, validation, prediction, temporal patterns, and interaction with social determinants of health.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets