Study population and setting
This paper evaluates the outcomes of COVID-19 among 75,463 hospitalized patients with and without underlying respiratory conditions. Data originated from the International Severe Acute Respiratory and emerging Infection Consortium (ISARIC) WHO Clinical Characterisation Protocol UK (CCP-UK) and included people who were admitted to a hospital in England, Scotland, or Wales between January 17 and August 3, 2020 with confirmed SARS-CoV-2 infection by RT-PCR or highly suspected COVID-19 prior to widespread availability of RT-PCR. The main exposures of interest were asthma and chronic pulmonary disease, based on self-report at baseline, and the main outcome was in-hospital mortality. Additional analyses evaluated admission to critical care units and use of invasive mechanical ventilation, non-invasive ventilation, or oxygen. Patients also reported use of asthma medications, including inhaled corticosteroids, short-acting beta-agonists (SABAs), long-acting beta-agonists (LABAs), or theophylline or leukotriene receptor agonists (LRTAs), within 2 weeks of admission. People with asthma who were taking inhaled corticosteroids, LABA, and another maintenance asthma medication were considered to have severe asthma. The main analyses were stratified by age group: 16-49 (n=8950), and ≥50 (n=65,653).
Summary of Main Findings
Among the 8950 individuals who were 16-49 years old, 1867 (21%) had asthma (55% female, 6.5% died) and 7083 (79%) did not have asthma (43% female, 5.4% died). After adjusting for age, sex, ethnicity, Index of Multiple Deprivation, smoking status, obesity, chronic cardiac disease, and malignant neoplasm, only severe asthma was associated with an increased hazard of death (hazard ratio (HR): 1.96 (95% CI: 1.25-3.08). However, overall asthma was associated with increased odds of critical care admission, invasive mechanical ventilation, non-invasive ventilation, and oxygen use (adjusting for the same covariates as above). Among the 65653 patients who were ≥50 years old, 47398 (72%) didn’t have any respiratory conditions (42% female, 34% died), 5918 (9%) had asthma alone (44% female, 28% died), 10266 (16%) had chronic pulmonary disease but no asthma (41% female, 42% died), and 2071 (3%) had a co-diagnosis of asthma and chronic pulmonary disease (52% female, 36% died). After adjusting for age, sex, ethnicity, Index of Multiple Deprivation, smoking status, obesity, malignant neoplasm, chronic cardiac disease, and chronic kidney disease, people with chronic pulmonary disease with and without inhaled corticosteroid use and people with both asthma and chronic pulmonary disease but without inhaled corticosteroid use were at increased risk of in-hospital mortality (HR: 1.16 (95% CI: 1.12-1.22), 1.10 (95% CI: 1.04-1.16), and 1.13 (95% CI: 1.01-1.28), respectively). Asthma with inhaled corticosteroid use was protective for in-hospital mortality, relative to no respiratory disease (HR: 0.86, 95% CI: 0.80-0.92). In adjusted logistic regression, asthma was consistently associated with increased odds of critical care admission, invasive mechanical ventilation, non-invasive mechanical ventilation, and oxygen; the results for chronic pulmonary disease (with and without asthma) were more inconsistent, showing protective associations with critical care admission and invasive mechanical ventilation, though this was potentially due to bias.
Missing data were multiply imputed. Results were consistent in sensitivity analyses that excluded patients with obesity, restricted analyses to only people with PCR-confirmed SARS-CoV-2 infection, and adjusted for additional medications.
Much of the exposure and confounder data, including comorbidities and mediation use, were based on baseline self-report, given linkage to primary or secondary care records was not available. This likely resulted in misclassification, which could bias results either toward or away from the null. Additionally, it is not clear when individuals began corticosteroids. It’s possible people who were healthier and engaged in health care were more likely to be on corticosteroids, explaining some of the protective effects seen (confounding by indication). Additionally, sample size was small in some exposure levels, particularly for severe asthma, which may have led to spurious associations. The logistic regression analyses for outcomes other than mortality did not account for the competing risk of death (which occurred at greater frequency than any of the other outcomes), which could have biased the results in such a way that a protective association was observed when the true relationship was null or positive. The study only included hospitalized people, and thus results are likely not generalizable to COVID-19 managed at home or less severe disease.
This was the largest known study to date of the relationship between asthma and underlying respiratory conditions and COVID-19 outcomes among hospitalized patients.
This review was posted on: 12 June 2021