28.05.2020

Hospitalization and Mortality among Black Patients and White Patients with Covid-19

Prognosis Transversal
Price-Haywood EG et al
NEJM

Main result

  • A total of 3481 patients were included in the study after exclusion of 145 subjects. The median age was 54 years, 60% were female and 70.4% were African-American, although they usually represent only 31% of patients in the area.
  • African-American subjects more often presented comorbidities: obesity, diabetes, hypertension, chronic renal failure and were more often symptomatic on admission (hyperthermia, cough and dyspnea).
  • Hospitalization: the hospitalization rate for the cohort was 39.7%. In multivariate analysis, being African-American compared to Caucasian was associated with an over-risk of hospitalization (odds ratio -OR- 1.96, 95% confidence interval: 1.65 - 2.37) as was age, having public health insurance (Medicare or Medicaid), and living in a low-income area. Female sex, on the contrary, reduces the risk of hospitalization (OR 0.56, 95% CI: 0.48-0.65).
  • In-hospital deaths: 23.6% of patients were hospitalized (n=326/1382). In multivariate analysis, ethnicity (African-American or non-Hispanic Caucasian) was not associated with the risk of death, unlike age, presence of tachypnea on admission, elevation of venous lactates, creatinine, procalcitonin, or lowering of platelets or lymphocytes.

Takeaways

In this study, African-American COVID-19 patients were more likely than Caucasian COVID-19 patients to visit a group of hospitals in Louisiana. Being African-American appears to be associated with an increased risk of hospitalization but not hospital death. These results remain to be confirmed in more robust studies and compared with other socio-economic, clinical and possibly biological data.

Strength of evidence Weak

- retrospective study
- study limited to a group of hospitals that may not necessarily reflect the entire patient population in the region
- numerous imputed missing data
- Incomplete follow-up for subjects who remained hospitalized at the end of the study.

Objectives

To compare the clinical characteristics and length of hospitalization of Caucasian and African-American subjects with COVID-19.

Method

  • Retrospective multi-center analytic cohort study conducted in Louisiana.
  • Inclusion of patients managed in one of the 40 hospitals in the group and tested positive for SARS-CoV-2 between 01/03/2020 and 11/04/2020 by RT-PCR.
  • Exclusion criteria: Neither African-American nor non-Hispanic Caucasian subjects or missing ethnicity data.
  • Data collection from the computerized file. Determination, based on the postal code, of residence or not in an area where more than 39.5% of the population has low income.
  • Biological examinations not performed in all subjects and therefore presented as subgroup analyses.
  • Study of the factors associated with hospitalization by multivariate logistic regression models and of the factors associated with death by Cox model taking into account the delay until death.
  • The highest level of adjustment took into account the following covariates for the two judgment criteria: age, sex, Charlson comorbidity score, residence in a low-income population area, obesity. The model studying hospitalization was also fitted on the type of health insurance of the patients and the model studying death on vital parameters at admission and biological parameters. Some interactions with ethnicity were considered but were not significant.
  • Parameters with more than 25% missing data were imputed, representing up to 683 patients for BMI.

bibliovid.org and its content are bibliovid property.

Legal Notice