What are the risk factors for severe covid-19?
Author: Dr Mark Ali MBBS BSc FRCS CTh, Medical Director
The outbreak of the disease caused by COVID-19 was declared a pandemic by the WHO on March 11 2020. Since then, there have been at least 400,000 deaths globally including over 40,000 within the UK. Sadly, the more admissions to hospital and the more deaths due to COVID-19, the more data we gather about risk factors for severe COVID-19. Globally, there is a pattern of those at highest risk: we have become used to hearing in the news “underlying medical conditions” or certain ages. Knowing who are particularly at risk of severe disease helps to highlight the most vulnerable in our societies so that we can protect them. It may also elucidate some aspects of the pathophysiology of the disease.
Important considerations: rates, excess deaths and independent risk factors
Before discussing the different risk factors for severe disease, it is important to mention that it is difficult to give very accurate morbidity and mortality rates. For this, we need an accurate value for the denominator: the numbers infected. We really don’t know this- the vast majority of community cases (not requiring admission) have not been formally diagnosed by a swab until recently. In addition, many people have likely had the infection asymptomatically (the numbers vary widely by study). Antibody tests to detect evidence of past infection are starting to be rolled out. But even then it has become clear that many do not generate antibody responses detectable by these tests (it may be a test issue, or it may be that people fight COVID-19 using a different part of the immune system not requiring antibodies)[1].
We know that COVID-19 tends to hit those with “underlying conditions” and the elderly the hardest. Many of these people have an increased annual risk of death that may be equal to or exceed their risk of death from COVID-19- might they have passed away this year from causes other than COVID-19? This is why it is important to consider excess deaths- deaths in excess of what we might expect for each timepoint. Furthermore, it is likely that there will be excess deaths for non-COVID-19 reasons. For example, this may be due to delays in a presentation at hospital and therefore more severe illness.
We can begin to gather this data but a full analysis of COVID-19 and non-COVID-19 excess deaths will only be possible in several months' time when longer-term effects and additional data, both death registrations and other sources, can be considered.
Many risk factors for severe COVID-19 will likely be related. For example, the risk of severe COVID-19 increases with age and also cardiac disease. But cardiac disease increases with age. Analyses try to disentangle these related risk factors to show if they are independently associated with severe disease. As more data is collected, we will have a better idea of what is truly an independent risk factor.
Age
We know that as you get older, your risk of death increases- both your overall risk of death (separate from coronavirus) and risk of death from COVID-19. Studies globally have consistently shown that being older increases your risk of dying from COVID-19. A Public Health England report covering disparities showed that, among people already diagnosed with COVID-19, people who were 80 or older were seventy times more likely to die than those under 40[2]. This could be because, as we get older, the frequency of underlying conditions associated with severe COVID-19 increases. In addition, the immune system is likely to be less effective in older people; age is a risk factor, independent of underlying conditions.
Some serious complications and deaths have occurred in young and middle-aged people but most victims have been in the over-65s category with 12% of deaths under-65. Risk increases steeply with age: the mortality rate at ages 80-84 in England and Wales is about ten times greater than at ages 60-64. The graph below from Public Health England[3] shows the number of COVID-19 deaths in each age category.
During the pandemic, there are also more people dying from non-COVID-19-related causes. The office for national statistics (ONS) has also found that non-COVID-19 excess deaths occur predominantly in older age groups, increasing with age and especially for the frail and those with underlying health conditions. Undiagnosed COVID-19 may account for some of these deaths. But it likely does not account for all. For example, deaths due to causes such as asthma and diabetes have increased and occurred increasingly outside the hospital; this could suggest a delay in care for these conditions is leading to an increase in deaths. Time will show what will happen due to delayed diagnoses because of reduced/redirected care during the pandemic (e.g. delayed cancer presentations). Again, this will likely have an age-related effect.
Gender
The first signs of a sex difference in COVID-19 severity emerged from hospital records in Wuhan shortly after the city locked down. Since then, this observation has been verified globally. Men are more likely to be admitted to hospital than women (60% vs 40%) and the mortality rate for men is double that of women[4]. Below is a graph from the Office of National Statistics showing the age-specific mortality rates for men and women.
The difference doesn’t appear to be caused by differential rates of infection. For example, a study in New York[5] found that equal numbers of men and women catch the virus. But men are more likely to progress to severe illness and death.
There are a few hypotheses for this gender disparity. Some have suggested smoking may play a part. In China, over half of men smoke but only 5% of women do. Tobacco smoking increases the number of ACE2 receptors, the very receptors COVID-19 uses to enter cells. This may mean that smoking makes cells more susceptible to the virus. However, the data doesn’t support this: there is mixed evidence on the role of smoking in COVID-19 and associated outcomes.
It may be that men, in particular older men, are in worse health than women. They tend to have higher rates of underlying conditions that might make them more susceptible to COVID-19 e.g. cardiovascular disease, obesity, diabetes.
It may be that men, in particular older men, are in worse health than women. They tend to have higher rates of underlying conditions that might make them more susceptible to COVID-19 e.g. cardiovascular disease, obesity, diabetes.
There is also some evidence that female sex hormones, oestrogen and progesterone, boost the immune system but we don’t know this specifically in relation to COVID-19.
Pre-existing conditions
We have heard a lot about pre-existing or underlying conditions in the news: about 90% of COVID-related deaths occur in people with pre-existing conditions. These include cardiovascular disease, diabetes, non-asthmatic chronic pulmonary disease, renal disease and other conditions weakening the immune system. The UK government published a list of conditions that likely increase the risk of severe illness from COVID-19. The high risk and moderate risk groupings are based on theory and also updated data from around the world. The lists can be seen here: https://www.nhs.uk/conditions/coronavirus-covid-19/people-at-higher-risk/whos-at-higher-risk-from-coronavirus/
A study of over 20,000 hospital inpatients[6] In the UK showed that the commonest comorbidities at admission were a chronic cardiac disease, uncomplicated diabetes, non-asthmatic chronic pulmonary disease and chronic kidney disease. Chronic cardiac disease, non-asthmatic chronic pulmonary disease, chronic kidney disease, liver disease and obesity were associated with higher in-hospital mortality.
Another study conducted by the OpenSAFELY Collaborative[7] showed similar results. In the largest cohort study of its kind, scientists analysed NHS health data from 17.4 million UK adults between 1st February and 25th April 2020. Among the adults in this sample, there were 5,683 deaths in hospital attributable to COVID-19. Death was strongly associated with being male, older age and deprivation, uncontrolled diabetes, severe asthma and various other prior medical conditions. Compared to people with ethnicity recorded as white, black and Asian people were at higher risk of death. This is discussed below.
People with type 1 diabetes were found to have 3.5 times increased risk of death and those with type 2 diabetes were twice as likely to die as those without diabetes in the hospital with coronavirus. However, these figures show those who have died in the hospital, not the many thousands who have recovered at home or been discharged from the hospital. The majority of people, whether they have diabetes or not, don’t need to go to the hospital: the absolute risk is low but the relative risk is increased.
Raised BMI is consistently associated with poorer outcomes. The above study showed that the risk of COVID-related hospital death increased from between 1.5 to 2 times for those with a BMI of 30 and 2 times for those with a BMI of 40 or more. This may be due to confounding factors such as higher rates of other comorbidities like cardiovascular disease. Obesity itself may increase risk by weakening the immune system and causing a pro-inflammatory state. We know that those who do badly from COVID-19 have an exaggerated inflammatory response. Obesity may exacerbate this. Furthermore, obesity makes ventilation more difficult due to reduced lung capacity relative to body size. Intubation is often more tricky as are scans and general patient care.
Ethnicity
Death rates from COVID-19 are highest amongst people from Black, Asian and Minority Ethnic groups. The Public Health England report into the disparities in the risk and outcome from COVID-18 showed that, after accounting for the effect of sex, age, deprivation and region, people of Bangladeshi ethnicity had around twice the risk of death than people of White British ethnicity. People of Chinese, Indian, Pakistani, Other Asian, Caribbean and Other Black ethnicity had between 10 and 50% higher risk of death when compared to White British.
Some of this increased risk will be accounted for by the presence of other important risk factors for acquiring COVID-19 and risk of death e.g. comorbidities, occupation, obesity. When comorbidities are included, the difference in risk of death amongst hospitalised patients is greatly reduced. However, this increased risk of death is only partially attributable to these factors.
Deprivation and occupation
A report from Public Health England has shown that the impact of COVID-19 has replicated existing health inequalities and, in some cases, has increased them. People who live in deprived areas have higher diagnosis rates and higher death rates than those living in less deprived areas. The mortality rates from COVID-19 in the most deprived areas were more than double the least deprived areas, for both males and females. One analysis[8] showed that little of this excess risk is accounted for by comorbidities or other risk factors. Related to this is occupation. The occupation will affect the risk of exposure to COVID-19. There is also overlap between occupation and other risk factors such as deprivation, ethnicity and comorbidities.
The ONS analysed 2,494 deaths in the working-age population (20-64) in England and Wales up to and including 20th April 2020[9]. Men working in the lowest skilled occupations had the highest death rate compared with the rate amongst those of the same sex and age. Men working as security guards had one of the highest rates. Social care workers (care workers and home carers) also had significantly raised death rates for both men and women. However, healthcare workers such as doctors and nurses were not found to have higher death rates compared with the same age and sex. This may be due to increased Covid-19 testing and personal protective equipment in these settings. For men, other specific at-risk occupations with raised death rates included: taxi drivers, chauffeur, bus and coach drivers, chefs and sales and retail assistants.
Smoking- mixed evidence
Smoking worsens outcomes from respiratory infections. Therefore the expectation is that smoking would predispose to worse outcomes from COVID-19. However, there is mixed evidence on the role of smoking in COVID-19 infection and associated outcomes. Some studies have detected fewer people who smoke than would be expected in hospitalised patients with COVID-19.
It is not clear if this is genuinely due to a protective effect of smoking in COVID-19 or misreporting, confounding and bias. Nicotine may have a biologically plausible protective effect but the clinical significance is unclear.
Genetics
There is likely to be a genetic component to the symptoms experienced due to COVID-19 and risk of severe disease. Some of this will be due to genetic factors influencing comorbidities whilst others may be down to differences in the immune system.
A study of twins[10]- by King’s College, London, showed a genetic effect on COVID-19 symptoms. The findings are based on data collected via the COVID-19 symptom tracking app. The team asked thousands of twins already enrolled in other research projects (TwinsUK) to download the app and log their symptoms. The researchers looked at the similarities in symptoms or non-symptoms between identical twins (sharing 100% of their genes) and non-identical twins (sharing half of their genes). If there is a genetic factor to how symptoms are experienced, then you would expect a greater similarity amongst the identical twins. The study had 2,600 twins and took into account whether households were shared. The results showed that 50% of the differences in symptoms could be explained by genetics. They found a big genetic influence for symptoms of fever, diarrhoea, delirium, and losses of taste and smell. Other symptoms (hoarse voice, cough, reduced appetite, chest and abdominal pain) were not linked to genetic makeup.
This may help to explain the observation that COVID-19 has very different presentations in different people. Another study of more than 750,000 people by 23andme, the genetics testing company, showed that people with type O blood were 9-18% less likely to test positive for COVID-19. Similarly, the ABO blood group associated with respiratory failure in an Italian-Spanish genome-wide association study[11]. The study showed a higher risk for group A and protective effect of group O, which was observed in other studies[12]. Many genetic studies are underway, which will help to elucidate the reasons for such variable symptomatology and severity amongst different people. This may help to protect the most vulnerable and identify targets for therapies.
[1] https://www.cebm.net/covid-19/covid-19-what-proportion-are-asymptomatic/
[5] https://www.medrxiv.org/content/10.1101/2020.04.08.20057794v1.full.pdf
[6] https://www.bmj.com/content/369/bmj.m1985
[7] https://www.medrxiv.org/content/10.1101/2020.05.06.20092999v1.full.pdf
[8] https://www.medrxiv.org/content/10.1101/2020.05.06.20092999v1.full.pdf
[10] https://covid.joinzoe.com/post/genetics-covid
[11] https://www.medrxiv.org/content/10.1101/2020.05.31.20114991v1.full.pdf
[12]https://www.medrxiv.org/content/10.1101/2020.03.11.20031096v2