Statistics from Altmetric.com
The Authors’ reply
We thank interested readers and their interesting views and opinions on our recently published study ‘Risk of severe COVID-19 disease with ACE inhibitors and angiotensin receptor blockers: cohort study including 8.3 million people’.1 Please find our responses to views and comments as follows.
In this study, the most recent smoking records available from general practices were used prior to cohort entry. Smoking status was recorded for 96% of patients included in the study, although recent changes in smoking status might not have been recorded. At the point of analyses of the study, negative test results were not yet available from Public Health England (PHE). However, we agree that it will be useful to compare clinical and demographic characteristics between COVID-19 positive and negative test groups to assess potential differences between both groups for biases caused by testing strategies. However, if smokers were more likely to be tested as Berlin2 suggests, this would tend to lead to an over-representation of smokers among those testing positive, which is clearly not the case. Also, the study referred to is from an online survey of people aged 13–24 in the USA so it is unlikely to be representative of testing patterns in UK general population.3 As stated in our paper, our findings for smoking status arose when smoking status was included as a confounder and should be interpreted cautiously. We are now carrying out a detailed study on smoking status and COVID-19 outcomes, and in this ongoing study we hope to look at negative test groups for ascertainment of smoking and COVID-19 outcomes with recent data released from PHE.4 With regard to alcohol use, we have not included this variable in our study so we cannot compare it with our findings for smoking status.
We agree that any causative link between smoking and COVID-19 infection or severity should be ascertained via prospective studies. Two studies are currently underway within our group to evaluate the association of smoking and COVID-19-related outcomes using (1) an epidemiological approach4 and (2) a causal inference approach, that is, Mendelian randomisation study using UK Biobank data. Causal estimates on the association of smoking and COVID-19-related outcomes and the dose-response effects will help inform future testing of nicotine replacement therapy in clinical trials as potential treatment and prophylaxis for COVID-19.
Contributors All authors contributed to this reply to reviewers.
Funding This study was funded by John Fell Oxford University Press Research Fund, Wellcome Trust (221514/Z/20/Z), Oxford Wellcome Institutional Strategic Support Fund (204826/Z/16/Z) and Cancer Research UK (C5255/A18085).
Competing interests JH-C reports grants from the National Institute for Health Research Biomedical Research Centre, Oxford, grants from John Fell Oxford University Press Research Fund, grants from Cancer Research UK (CRUK) grant number C5255/A18085, through the Cancer Research UK Oxford Centre, and grants from the Oxford Wellcome Institutional Strategic Support Fund (204826/Z/16/Z), during the conduct of the study; personal fees and other from ClinRisk (until 2019), outside the submitted work; and is an unpaid director of QResearch, a not-for-profit organisation which is a partnership between the University of Oxford and EMIS Health, which supply the QResearch database used for this work. PST reports consulting with AstraZeneca and Duke-NUS, outside the submitted work. CC has nothing to disclose.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Patient consent for publication Not required.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.