ABSTRACT
Background: Disparate COVID-19 outcomes have been observed between Hispanic, Non-Hispanic Black, and White patients. The underlying causes for these disparities are not fully understood. Methods: : This was a retrospective study utilizing electronic medical record data from five hospitals within a single academic health system based in New York City. Multivariable logistic regression models were used to identify demographic, clinical, and lab values associated with in-hospital mortality. Results: : 3,086 adult patients with self-reported race/ethnicity information presenting to the emergency department and hospitalized with COVID-19 up to April 13, 2020 were included in this study. While older age (multivariable OR 1.06, 95% CI 1.05-1.07) and baseline hypoxia (multivariable OR 2.71, 95% CI 2.17-3.36) were associated with increased mortality overall and across all races/ethnicities, Non-Hispanic Black (median age 67, IQR 58-76) and Hispanic (median age 63, IQR 50-74) patients were younger and had different comorbidity profiles compared to Non-Hispanic White patients (median age 73, IQR 62-84; p<0.05 for both comparisons). Among inflammatory markers associated with COVID-19 mortality, there was a significant interaction between the Non-Hispanic Black population and interleukin-1-beta (interaction p-value 0.04). Conclusions: : This analysis of a multi-ethnic cohort highlights the need for inclusion and consideration of diverse popualtions in ongoing COVID-19 trials targeting inflammatory cytokines.
Subject(s)
COVID-19 , Hypoxia , White Muscle DiseaseABSTRACT
Objectives: To compare risk factors for COVID-19 mortality among hospitalized Hispanic, Non-Hispanic Black, and White patients. Design: Retrospecitve cohort study Setting: Five hosptials within a single academic health system Participants: 3,086 adult patients with self-reported race/ethnicity information presenting to the emergency department and hospitalized with COVID-19 up to April 13, 2020. Main outcome measures: In-hospital mortality Results: While older age (multivariable OR 1.06, 95% CI 1.05-1.07) and baseline hypoxia (multivariable OR 2.71, 95% CI 2.17-3.36) were associated with increased mortality overall and across all races/ethnicities, Non-Hispanic Black (median age 67, IQR 58-76) and Hispanic (median age 63, IQR 50-74) patients were younger and had different comorbidity profiles compared to Non-Hispanic White patients (median age 73, IQR 62-84; p<0.05 for both comparisons). Among inflammatory markers associated with COVID-19 mortality, there was a significant interaction between the Non-Hispanic Black population and interleukin-1-beta (interaction p-value 0.04). Conclusions: This analysis of a multi-ethnic cohort highlights the need for inclusion and consideration of diverse popualtions in ongoing COVID-19 trials targeting inflammatory cytokines.
Subject(s)
COVID-19 , HypoxiaABSTRACT
Importance The ACE D allele is more prevalent among African Americans (AA) compared to other races/ethnicities and has previously been associated with severe COVID-19 pathogenesis through excessive ACE1 activity. ACE-I/ARBs may counteract this mechanism, but their association with COVID-19 outcomes has not been specifically tested in the AA population. Objectives To determine whether the use of ACE-I/ARBs is associated with COVID-19 in-hospital mortality in AA compared with non-AA population. Design, Setting, and Participants In this observational, retrospective study, patient-level data were extracted from the Mount Sinai Health System’s (MSHS) electronic medical record (EMR) database, and 6,218 patients with a laboratory-confirmed COVID-19 diagnosis from February 24 to May 31, 2020 were identified as ACE-I/ARB users. Exposures Patients with an active prescription from January 1, 2019 up to the date of admission for ACE-I/ARB (outpatient use) and patients administered ACE-I/ARB during hospitalization (in-hospital use) were identified. Main Outcomes and Measures The primary outcome was in-hospital mortality, assessed in the entire, AA, and non-AA population. Results Of the 6,218 COVID-19 patients, 1,138 (18.3%) were ACE-I/ARB users. In a multivariate logistic regression model, ACE-I/ARB use was independently associated with reduced risk of in-hospital mortality in the entire population (OR, 0.655; 95% CI, 0.505-0.850; P=0.001), AA population (OR, 0.44; 95% CI, 0.249-0.779; P=0.005), and non-AA population (OR, 0.748, 95% CI, 0.553-1.012, P=0.06). In the AA population, in-hospital use of ACE-I/ARBs was associated with improved mortality (OR, 0.378; 95% CI, 0.188-0.766; P=0.006) while outpatient use was not (OR, 0.889; 95% CI, 0.375-2.158; P=0.812). When analyzing each medication class separately, ARB in-hospital use was significantly associated with reduced in-hospital mortality in the AA population (OR, 0.196; 95% CI, 0.074-0.516; P=0.001), while ACE-I use was not associated with impact on mortality in any population. Conclusion and Relevance In-hospital use of ARBs was associated with a significant reduction in in-hospital mortality among COVID-19-positive AA patients. These results support further investigation of ARBs to improve outcomes in AA patients at high risk for COVID-19-related mortality.
Subject(s)
COVID-19ABSTRACT
Importance: Alpha-1-adrenergic receptor antagonists (1-blockers) can abrogate pro-inflammatory cytokines and may improve outcomes among patients with respiratory infections. Repurposing readily available drugs such as 1-blockers could augment the medical response to the COVID-19 pandemic. Objective: To evaluate the association between 1-blocker exposure and COVID-19 mortality Design: Real-world evidence study Setting: Patient level data with 32,355 records tested for SARS-CoV-2 at the Mount Sinai Health System including 8,442 laboratory-confirmed cases extracted from five member hospitals in the New York City metropolitan area. Participants: 2,627 men aged 45 or older admitted with COVID-19 between February 24 and May 31, 2020 Exposures: 1-blocker use as an outpatient or while admitted for COVID-19 Main Outcomes and Measures: In-hospital mortality Results: Men exposed to 1-blockers (N=436) were older (median age 73 vs. 64 years, P<0.001) and more likely to have comorbidities than unexposed men (N=2,191). Overall, 758 (28.9%) patients died in hospital, 1,589 (60.5%) were discharged, and 280 (10.7%) were still hospitalized as of May 31, 2020. Outpatient exposure to 1-blockers was not associated with COVID-19 hospital outcomes, though there was a trend towards significance (OR 0.749, 95% CI 0.527-1.064; P=0.106). Conversely, inpatient use of 1-blockers was independently associated with improved in-hospital mortality in both multivariable logistic (OR 0.633, 95% CI 0.434-0.921; P=0.017) and Cox regression analyses (HR 0.721, 95% CI 0.572-0.908; P=0.006) adjusting for patient demographics, comorbidities, and baseline vitals and labs. Age-stratified analyses suggested greater benefit from inpatient 1-blocker use among younger age groups: Age 45-65 OR 0.384, 95% CI 0.164-0.896 (P=0.027); Age 55-75 OR 0.511, 95% CI 0.297-0.880 (P=0.015); Age 65-89 OR 0.810, 95% CI 0.509-1.289 (P=0.374). Conclusions and Relevance: Inpatient 1-blocker use was independently associated with improved COVID-19 mortality among hospitalized men. Clinical trials to assess the therapeutic value of 1-blockers in COVID-19 are warranted.
Subject(s)
COVID-19 , Respiratory Tract InfectionsABSTRACT
The global surge in COVID-19 cases underscores the need for fast, scalable, and reliable testing. Current COVID-19 diagnostic tests are limited by turnaround time, limited availability, or occasional false findings. Here, we developed a machine learning-based framework for predicting COVID-19 positive test status relying only on readily-available baseline data, including patient demographics, comorbidities, and common lab values. Leveraging a cohort of 31,739 adults within an academic health system, we trained and tested multiple types of machine learning models, achieving an area under the curve of 0.75. Feature importance analyses highlighted serum calcium levels, aspartate aminotransferase levels, and oxygen saturation as key predictors. Additionally, we developed a single decision tree model that provided an operable method for stratifying sub-populations. Overall, this study provides a proof-of-concept that COVID-19 status prediction models can be developed using only baseline data. The resulting prediction can complement existing tests to enhance screening and pandemic containment workflows.
Subject(s)
COVID-19ABSTRACT
Background: The global surge in COVID-19 cases underscores the need for fast, scalable, and reliable testing. Current COVID-19 diagnostic tests are limited by turnaround time, limited availability, or occasional false findings.Methods: In this study, we developed a machine learning-based framework for predicting COVID-19 positive test status relying only on readily available baseline data, including patient demographics, comorbidities, and common lab values. Leveraging a cohort of 31,739 adults within an academic health system, we divided the patient data into a training set (patient encounters through April 13, 2020) and a test set (patient encounters from April 13, 2020 through June 2, 2020). We trained our machine learning models on the training set and evaluated model performance on the test set.Findings: We trained and tested multiple types of machine learning models, achieving an area under the curve of 0·75 in the test set. Feature importance analyses highlighted serum calcium levels, aspartate aminotransferase levels, and oxygen saturation as key predictors. Additionally, we identified an optimal probability threshold for patient screening and developed a single decision tree model that provided an operable method for stratifying sub-populations.Interpretation: Overall, this study provides a proof-of-concept that COVID-19 status prediction models can be developed using only baseline data. Our machine learning models can be adapted to a variety of global pandemic scenarios, as the resulting prediction could complement existing tests to enhance screening and pandemic containment workflows.Funding: Icahn School of Medicine at Mount Sinai, New York, NY.Declaration of Interests: J.F. is an employ of Outco Inc. All other authors declare no competing financial interests.Ethics Approval Statement: This study utilized de-identified data extracted from the electronic health record and as such was considered nonhuman subject research. Therefore, this study was exempted from the Mount Sinai IRB review and approval process. All analyses were carried out in accordance with relevant guidelines and regulations.
Subject(s)
COVID-19ABSTRACT
Background: Little is known about risk factors for COVID-19 outcomes, particularly across diverse racial and ethnic populations in the United States. Methods: In this prospective cohort study, we followed 3,086 COVID-19 patients hospitalized on or before April 13, 2020 within an academic health system in New York (The Mount Sinai Health System) until June 2, 2020. Multivariable logistic regression was used to evaluate demographic, clinical, and laboratory factors as independent predictors of in-hospital mortality. The analysis was stratified by self-reported race and ethnicity. Findings: A total of 3,086 COVID-19 patients were hospitalized, of whom 680 were excluded (78 due to missing race or ethnicity data, 144 were Asian, and 458 were of other unspecified race/ethnicity). Of the 2,406 patients included, 892 (37.1%) were Hispanic, 825 (34.3%) were black, and 689 (28.6%) were white. Black and Hispanic patients were younger than White patients (median age 67 and 63 vs. 73, p<0.001 for both), and they had different comorbidity profiles. Older age and baseline hypoxia were associated with increased mortality across all races. There were suggestive but non-significant interactions between Black race and diabetes (p=0.09), and obesity (p=0.10). Among inflammatory markers associated with COVID-19 mortality, there was a significant interaction between Black race and interleukin-1-beta (p=0.04), and a suggestive interactions between Hispanic ethnicity and procalcitonin (p=0.07) and interleukin-8 (p=0.09). Interpretation: In this large, racially and ethnically diverse cohort of COVID-19 patients in New York City, we identified similarities and important differences across racial and ethnic groups in risk factors for in-hospital mortality.
Subject(s)
Diabetes Mellitus , Hypoxia , Obesity , COVID-19ABSTRACT
Objective: To identify sex-specific effects of risk factors for in-hospital mortality among COVID-19 patients admitted to a hospital system in New York City. Design: Prospective observational cohort study with in-hospital mortality as the primary outcome. Setting: Five acute care hospitals within a single academic medical system in New York City. Participants: 3,086 hospital inpatients with COVID-19 admitted on or before April 13, 2020 and followed through June 2, 2020. Follow-up till discharge or death was complete for 99.3% of the cohort. Results: The majority of the cohort was male (59.6%). Men were younger (median 64 vs. 70, p<0.001) and less likely to have comorbidities such as hypertension (32.5% vs. 39.9%, p<0.001), diabetes (22.6% vs. 26%, p=0.03), and obesity (6.9% vs. 9.8%, p=0.004) compared to women. Women had lower median values of laboratory markers associated with inflammation compared to men: white blood cells (5.95 vs. 6.8 K/uL, p<0.001), procalcitonin (0.14 vs 0.21 ng/mL, p<0.001), lactate dehydrogenase (375 vs. 428 U/L, p<0.001), C-reactive protein (87.7 vs. 123.2 mg/L, p<0.001). Unadjusted mortality was similar between men and women (28.8% vs. 28.5%, p=0.84), but more men required intensive care than women (25.2% vs. 19%, p<0.001). Male sex was an independent risk factor for mortality (OR 1.26, 95% 1.04-1.51) after adjustment for demographics, comorbidities, and baseline hypoxia. There were significant interactions between sex and coronary artery disease (p=0.038), obesity (p=0.01), baseline hypoxia (p<0.001), ferritin (p=0.002), lactate dehydrogenase (p=0.003), and procalcitonin (p=0.03). Except for procalcitonin, which had the opposite association, each of these factors was associated with disproportionately higher mortality among women. Conclusions: Male sex was an independent predictor of mortality, consistent with prior studies. Notably, there were significant sex-specific interactions which indicated a disproportionate increase in mortality among women with coronary artery disease, obesity, and hypoxia. These new findings highlight patient subgroups for further study and help explain the recognized sex differences in COVID-19 outcomes.