ABSTRACT
Study Objective: To determine if COVID-19 markers of severity, positive test, and hospitalization, differ among racial groups. Additionally, to examine whether these differences are associated with mortality and to identify predictive variables for potential prevention and intervention. Methods: This is a retrospective cohort design studying those tested for COVID- 19. A multistate model was created using Trinity Health electronic health records in the US (January 1 - June 30, 2020). The primary outcome variable was mortality and secondary outcomes were COVID-19 positivity and hospitalization. Predictive variables included age, sex, race, insurance, income status, BMI, zip code population density and measures of comorbidities using the Charlson Comorbidity Index (CCI). Adjusted treatment effects were estimated using logistic regression. Results: The data included 181,199 patients of which 18,083 patients (9.95%) were Black and 133,452 (73.2%) were White. COVID-19 testing was positive in 13.7% of African Americans (AA) and 4.97% of Whites. AA patients had higher rates of comorbidities (p < 0.001), lower rates of commercial insurance (p < 0.001) and higher population densities (p < 0.001) as compared to White patients. Unadjusted logistic regression shows that AA patients have higher odds of infection (OR = 3.033, p < 0.001), mortality (1.3% vs 0.8%, OR = 1.656, p < 0.001), and hospitalization (OR = 1.165, p = 0.031) compared to white patients. After adjusting for predictors, the odds of SARS-CoV-2 infection are higher for AA (OR = 1.744, p < 0.001). There is no significant difference in the odds of mortality between AA patients and White patients who were COVID positive (OR = 0.740, p = 0.09), after adjusting for the other predictive variables. Conclusion: In this large multi-state study of COVID-19 tested patients, African Americans were infected much more often and had greater mortality than Whites before adjusting for covariates. The rate of hospitalization was lower for COVID positive AAs than Whites, and mortality was nearly the same as Whites after adjusting for predictors such as comorbidities. Our study identifies variables associated with COVID-19 morbidity and mortality, highlighting the disproportionate impact of COVID-19 on the African American community. This analysis may provide opportunities to employ preventive medicine approaches and mitigate systemic inequities to improve the health of vulnerable populations. [Formula presented]
ABSTRACT
Study Objectives: Approximately 10-30% of older patients in the emergency department (ED) exhibit delirium, which goes unrecognized by up to 75% of providers. Delirium is linked to increased lengths of stay, in-hospital falls, cognitive decline, and mortality, yet in a recent national survey of ACEP members, only 14% reported having a protocol addressing delirium in the ED. We conducted a feasibility pilot of a delirium toolkit developed to improve screening and management of delirium in the ED. Methods: Supported by a monthly workgroup, four EDs used the toolkit to develop and implement distinct quality improvement (QI) initiatives contextually appropriate to their ED (sites represented a range of ED environments). QI initiatives included delirium screening (using the CAM, bCAM, and/or DTS instruments) as well as delirium management strategies. Toolkit feasibility testing included assessment of implementation speed, protocol adherence, and qualitative feedback. Sites implemented and reported on process metrics for their QI initiatives from July – November 2020. Results: Findings reflect data from three sites (the fourth site did not contribute quantitative data). Over 73% of ED staff received delirium protocol training across sites in the first month of implementation, and staff participation in additional monthly trainings continued at a lower intensity over time. A total of 7,107 delirium screenings were conducted (representing 43% of older adults visiting the three EDs during the study period) and 4.5% of delirium screenings were positive. Over time, the monthly number and proportion of older adults screened for delirium trended slightly downwards, while the proportion of positive delirium screenings trended upwards. The sites provided 1,460 instances of delirium management activities (some patients received more than one). These activities were grouped into over a dozen different categories, with documenting an updated diagnosis or disposition being most common (300 instances), followed by orientation (239 instances) and hydration/nutrition interventions (196 instances). Conclusion: All pilot sites leveraged the ED-Delirium Toolkit to develop QI initiatives, with three of the four sites contributing data demonstrating successful implementation. These delirium QI initiatives were seen as complementary activities to the concurrent pandemic priorities given the recognition of delirium as both a presenting symptom as well as a common complication of COVID-19. ED nurses may have been able to improve targeted screening of patients over time based on the increasing positivity rate and declining proportion and number of screenings conducted. Given the number of staff trained and scale of delirium management activities, use of the toolkit increased awareness of and interventions for addressing delirium in the ED.