Skip to main content
SearchLoginLogin or Signup

Review 1: "Inhaled corticosteroid use and risk COVID-19 related death among 966,461 patients with COPD or asthma: an OpenSAFELY analysis"

This well-conducted, high powered study provides strong evidence of that inhaled corticosteroids do not protect against COVID-19 related deaths.

Published onSep 10, 2020
Review 1: "Inhaled corticosteroid use and risk COVID-19 related death among 966,461 patients with COPD or asthma: an OpenSAFELY analysis"
1 of 2
key-enterThis Pub is a Review of
Inhaled corticosteroid use and risk COVID-19 related death among 966,461 patients with COPD or asthma: an OpenSAFELY analysis

Background: Early descriptions of the coronavirus outbreak showed a lower prevalence of asthma and COPD than was expected for people diagnosed with COVID-19, leading to speculation that inhaled corticosteroids (ICS) may protect against infection with SARS-CoV-2, and development of serious sequelae. We evaluated the association between ICS and COVID-19 related death using linked electronic health records in the UK. Methods: We conducted cohort studies on two groups of people (COPD and asthma) using the OpenSAFELY platform to analyse data from primary care practices linked to national death registrations. People receiving an ICS were compared to those receiving alternative respiratory medications. Our primary outcome was COVID-19 related death. Findings: We identified 148,588 people with COPD and 817,973 people with asthma receiving relevant respiratory medications in the four months prior to 01 March 2020. People with COPD receiving ICS were at a greater risk of COVID-19 related death compared to those receiving a long-acting beta agonist (LABA) and a long-acting muscarinic antagonist (LAMA) (adjusted HR = 1.38, 95% CI = 1.08 - 1.75). People with asthma receiving high dose ICS were at an increased risk of death compared to those receiving a short-acting beta agonist (SABA) only (adjusted HR = 1.52, 95%CI = 1.08 - 2.14); the adjusted HR for those receiving low-medium dose ICS was 1.10 (95% CI = 0.82 - 1.49). Quantitative bias analyses indicated that an unmeasured confounder of only moderate strength of association with exposure and outcome could explain the observed associations in both populations. Interpretation: These results do not support a major role of ICS in protecting against COVID-19 related deaths. Observed increased risks of COVID-19 related death among people with COPD and asthma receiving ICS can be plausibly explained by unmeasured confounding due to disease severity.

RR:C19 Evidence Scale rating by reviewer:

  • Reliable. The main study claims are generally justified by its methods and data. The results and conclusions are likely to be similar to the hypothetical ideal study. There are some minor caveats or limitations, but they would/do not change the major claims of the study. The study provides sufficient strength of evidence on its own that its main claims should be considered actionable, with some room for future revision.



This is a novel, useful contribution, of interest, which, as far as it is able, demolishes the rather straw man hypothesis that a protective effect of inhaled corticosteroids (ICS) might account for the observation that asthma, and to a lesser extent, COPD patients appeared under-represented in some of the (earlier) patient descriptions of COVID-19. The study is well conducted and the Conclusions appear sound. It demonstrates the value of the OpenSAFELY resource and its rapid exploitation in addressing specific hypotheses related to healthcare.

  1. As acknowledged, it is disappointing that no spirometric data are available for categorisation of severity more than for diagnosis of airflow obstruction. Even one-off blood eosinophil counts would bear on appropriateness of ICS and should be a covariate.

  2. The adopted COPD criteria are arguable; the diagnosis would be very unusual at age 35 (I know this was previously employed though not widely used). LAMA monotherapy (not counting SABA) should be relatively common in the COPD population and must be a valid comparison group (cf asthma groups). LABA  monotherapy is presumably more unusual.

  3. It has been well demonstrated that prescribed ICS dose correlates well with asthma severity, and while many patients are not prescribed or do not take ICS appropriately, and it is argued that many are inappropriately prescribed ICS (NICE), I do not accept the contention that the SABA alone group have similar severity to ICS users. The data also do not support this.

  4. There are a few omissions most of which could be addressed reasonably easily;

    • the denominator should be made clearer; if it is 24 million then the incidence of COPD and asthma appears rather low; at 291,805/24m =1.21% and 1.285/24m =5.35%. Why is this?

    • The time-course Figures (1a and 1b) are not very meaningful without also expressing the comparative whole population deaths over time (which should be readily available).

    • It would appear important to report time from Covid diagnosis to date of death (there has been concern that ONS reports Covid death in anyone who tested positive even months later, when it may be un-related).

    • With all the caveats, would it not be worth reporting incidence of positive Covid tests in the comparison groups to examine a further potential effect of ICS?

    • There needs to be some discussion of dosing with respect to adherence. Is there any measure of previous prescription collection? Why 1 ICS prescription in 4 months? What does this mean for the commonest inhalers ?60 puffs ?120 ?200 for low doses.

    • smoking histories (pack-years) or some estimate must be relevant and could be a covariate, as should current smoking.

    • In view of the recent major educational emphasis on ICS deprescribing in COPD, (and repeated pro ICS prescribing in asthma), it is possible that an unstated GP practice bias relates to the patients who continue to be prescribed ICS in COPD and who are not in asthma (practice prescription rates might be another covariate)? 

    • There is no mention of previous oral steroid usage; obesity, hypertension, DM are all increased in the HD ICS group.

    • How many of the CoVid deaths were unproven (U07.2)?

    • Box 2 contains no limitations of the study (though they are included in Discussion).

    • There is a large amount of data, particularly in asthma but also in COPD, that ICS should not be stopped abruptly and although this is included in Discussion should not this be emphasised under Implications in  Box 2 as well?

  1. Minor points

  • It is clear and reasonably well written.

  • The 1st sentence of the Introduction requires a date to be added.

  • I would argue that ‘people’ and ‘patients’ are a bit muddled and not entirely consistent.

  • The description of inhalers is very confusing eg ‘single therapy’ in Exposures, and descriptions in Tables 1 and 2 eg single SABA; need to distinguish drug alone vs combination or concomitant usage. LAMA/LABA is more conventional.

  • The y axis in Figures 1a and 1b is not clear.

  • In Table 1 for COPD should not 1st grouped age be 35-40 y?

  • There is a typo in the Title.

  • There is no mention of cardiovascular mortality related to Covid 19 and possible effects of ICS on this have not been considered (there is quite a previous literature).

  • The Referencing is good but incomplete for numbers 10,11,22,26,3.

No comments here
Why not start the discussion?