Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

The first large study to report the effect of bodyweight on risk of worse outcomes from COVID-19 across the full range of body mass index (BMI) is published today in The Lancet Diabetes and Endocrinology journal.

illustration of three normal people all wearing face coverings

The study, led by Nuffield Department of Primary Care Health Sciences researchers, based on more than 6.9 million people living in England and including data from more than 20,000 COVID-19 patients who were hospitalised or died during the first wave, found that the risk of worse outcomes from COVID-19 start rising in people with a BMI above 23kg/m2, which is considered to be in the healthy range. The risks of hospitalisation were 5% higher for each one unit increase in BMI and the risk of ICU admission was 10% higher for each unit increase. People who were underweight (BMI less than 18.5) also experienced worse outcomes from COVID-19.

The effect of excess weight on the risk of severe COVID-19 was greatest in young people aged 20 to 39 years of age and decreased after age 60, with increasing BMI having very little impact on the risk of severe COVID-19 in people aged over 80 years (increase in risk of hospitalisation per BMI unit: 20-39 years, 9%; 40-59 years, 8%; 60-79 years, 4%; 80-100 years 1%).

However, the incidence of worse outcomes in people aged 20-39 years was very low, which meant the risk of severe COVID-19 attributable to increased BMI was greatest in people of middle age because worse outcomes were more common in people aged 40-59 (for example, Out of 2,384,223 people in the 20 to 39 age group, there were 922 patients hospitalised with COVID-19 [0.04%],127 ICU admissions [<0.01%] and 32 deaths [<0.01%]; while in patients in the 40-59 age group, out of a total of 2,444,011 there were 2,845 people hospitalised [0.1%], 582 ICU admissions  [0.02%] and 378 deaths [0.02%])

The risks associated with higher BMI were greater for black people compared with white people (increase in risk of hospitalisation per BMI unit: black, 7% vs white 4%; increase in risk of death: black 8% vs white 4%). There was no evidence that the risks for other ethnic groups differed from those of the white populations.

Dr Carmen Piernas, lead author of the study, from the University of Oxford’s Nuffield Department of Primary Care Health Sciences, UK, said: “Our study shows that even very modest excess weight is associated with greater risks of severe COVID-19 complications and the risks rise sharply as BMI increases. Crucially we also show that the risks associated with excess weight are greatest in people aged under 50 years, while weight has little to no effect on your chances of developing severe COVID-19 after age 80. These findings suggest that vaccination policies should prioritise people with obesity, especially now the vaccine is being rolled out to younger age-groups.”

We don’t yet know that weight loss specifically reduces the risk of severe COVID-19 outcomes, but it is highly plausible, and will certainly bring other health benefits. Losing weight is hard and the recent NHS investment to improve access to weight management programmes could help to reduce the severity of COVID-19 at a population level and reduce the pressure on health care systems, while also lowering the risks for type 2 diabetes and some cancers.” - Professor Paul Aveyard, who co-led the study, from University of Oxford’s Nuffield Department of Primary Care Health Sciences

Previous studies have reported that obesity is associated with  more severe outcomes after infection with the SARS-CoV-2 virus, but most studies only included people who had been admitted to hospital with COVID-19 symptoms, which may bias the findings towards more severe cases.

The latest study is the first to examine the consequences of excess weight on COVID-19 outcomes across the full range of BMI. It is based on anonymised health records from 6,910,685 community-based patients in the QResearch database of routinely collected electronic patient health records in England. All of the participants included in the study were 20 years or older and had at least one BMI measurement on their record. The average BMI across the whole study group was 26.8kg/m2.

The researchers analysed records between 24 January and 30 April 2020 for outcomes linked to severe COVID-19 disease. They found that during the study period, 13,503 patients were admitted to hospital with COVID-19, 1,602 patients required treatment in intensive care (ICU), and 5,479 patients died. Most people with severe COVID-19 were aged over 60 years (72.1% of people admitted to hospital, 9,736/13,503; 55.7% of ICU admissions, 892/1,602; 92.5% of deaths, 5,069/5,479).

Across all age groups, the risk of hospitalisation from COVID-19 increased by around 5% for every increase in BMI unit above 23kg/m2, which is not attributable to other pre-existing health conditions, including type 2 diabetes (number of hospital admissions as a proportion of population: healthy weight BMI 18.5 - <25: 0.1% [3,655/2,713,189]; overweight BMI 25 - <30: 0.2% [4,593/2,306,897]; class 1 obesity BMI 30-34.9: 0.3% [2,823/1,072,777]; class 2 and 3 obesity BMI >35: 0.3% [2,070/608,335]).

Being underweight (BMI <18.5kg/m2) was also linked to an increased risk of hospitalisation and death from COVID-19, which the researchers say may be linked to other aspects of poor health associated with having a low BMI (number of hospital admissions as a proportion of population: 0.2% [362/209,497]; number of deaths: 0.1% [275/209,947]).

The authors note several limitations to their study. The analysis of the impact of BMI may be limited by the smaller sample of people with recent BMI measurements. However, the findings did not change when the researchers excluded BMI measurements that were more than a year old at the start of the study period.

They also note that daily fluctuations in weight and errors caused by clothing will cause imprecision in the estimates of BMI, therefore it may not be possible to precisely define the BMI associated with minimum risk based on their results. However, they say this is unlikely to differ by more than one BMI unit.  


Associations between body-mass index and COVID-19 severity in 6·9 million people in England: a prospective, community-based, cohort study
Min Gao, Carmen Piernas, Nerys M Astbury, Julia Hippisley-Cox, Stephen O’Rahilly, Paul Aveyard, Susan A Jebb
Lancet Diabetes Endocrinol S2213-8587(21)00089-9


Linked comment in the Lancet


Contact our communications team

Opinions expressed are those of the authors and not of Oxford University. Readers' comments will be moderated - see our guidelines for further information.