select search filters
roundups & rapid reactions
factsheets & briefing notes
before the headlines
Fiona fox's blog

expert reaction to ONS figures on deaths in England and Wales in week ending 3 April 2020

The Office for National Statistics (ONS) have released the latest figures, week ending April 3rd, for deaths in England and Wales.


Prof Sheila Bird, Formerly Programme Leader, MRC Biostatistics Unit, University of Cambridge, said:

“Today’s report by the Office for National Statistics on 16 387 death-registrations for England and Wales in week 14 (to 3 April) makes stark reading.

“First is ONS’s notable comparison with the average number of death registrations in the corresponding week over the past 5 years, which was 10 305. Whatever quibbles there may be about how COVID-mention-deaths are established, there is no doubt that the extra toll in lives lost to the pandemic is substantial. It is noteworthy, however, that around 2,600 of the additional 6082 deaths were not coded as “COVID-mention-deaths”.

“Given that COVID-19 is a new disease, I’ve chosen to represent the COVID-19-mention deaths as a percentage of the non-COVID-mentioned deaths (as if all of the latter were the deaths that might have occurred in an absence of coronavirus).

“COVID-mention-deaths, as a percentage of non-COVID-mentioned deaths, differ markedly by gender and age-group. Some of this difference may be indirect (e.g. due to difference in coding the cause of death)

Persons: For persons under 44 years of age, COVID-mention-deaths (43) add 14% to non-COVID-mention-deaths (317) but the percentage-added increases to 22% at 45-54 years (106/490). Thereafter, with some variation, the extra toll is about one third per age-group (55-84 years: 2 163/6 840) until we reach the two oldest age-groups [85-89 years and 90 years or older] when the extra lives lost (as certified) represent respectively 26%  (617/2 398) and 19% (546/2 866) of non-COVID-mention deaths.

Females: From age 60 onwards, COVID-mention-deaths are a substantially higher percentage of non-COVID-mention deaths for males than for females, such as 34% and 39% (males) versus 26% and 24% (females) at 70-74 years and 80-84 years respectively.”


Dr Amitava Banerjee, Associate Professor in Clinical Data Science and Honorary Consultant Cardiologist, Institute of Health Informatics, University College London (UCL), said:

“We know that COVID-19 has direct and indirect effects on health and mortality. The direct effects are directly due to the infection, and should be recorded in death certification. As long as we are including community-based as well as hospital-based deaths, and all sources such as nursing homes, this will be relatively well recorded. However, the indirect effects of the COVID-19 public health emergency are more difficult to capture and estimate. Partly the strain on the health system means that there are challenges for treatment and prevention of non-COVID diseases in the current context. This could be due to barriers to healthcare for some groups of people being shielded in order to protect themselves from infection and therefore presenting late or not at all to health services. It could also be both patients and health professionals thinking less about conditions such as cardiovascular disease and cancer, due to focus on COVID-19. Moreover the stress of the COVID-19 emergency and lock-down may in themselves have adverse health consequences beyond COVID-19 infection, such as increased incidence of cardiovascular disease. We cannot be sure at present and we need to look at the underlying causes recorded in death certificates in greater detail as well as presentation of other conditions to both primary and secondary healthcare settings.

“We need a mass mobilisation of our routine healthcare data in order to understand which of these factors are most important in causing excess deaths. A first step is to model the background risk of deaths based on underlying conditions, as we have done for England, suggesting an excess of 35 000 to 70 000 deaths depending on the population infection rate and the impact of COVID-19 on the health system. (


Prof Tom Dening, Professor of Dementia Research, University of Nottingham, said:

“It is worrying that there appears to be a sharp increase in deaths in the community that are not known to be due to Ccovid-19. So far, we don’t have good data on the possible reasons for this but there are a number of possibilities. The first is simply that many of these are in fact caused by Covid-19 that wasn’t diagnosed. Testing remains extremely limited outside of hospitals so we probably won’t ever know how many people had the virus during this period.

“There are probably multiple reasons for other deaths. These include people not feeling able to attend their GP surgeries, call an ambulance or attend A&E as they may have done in the past. Therefore, some serious conditions may present too late for effective treatment. Concern has been expressed by doctors working in children’s emergency care that they are not currently seeing the usual range of childhood emergencies.

“Another possibility is that some people with serious conditions, like cancer or chronic kidney disease, are either unable or unwilling to attend hospital on the usual regular basis, so their treatment regimes may lapse. Managing conditions like unstable diabetes will be much harder remotely than with face to face attention. Or there may be people who would have been referred to specialists for assessment of potentially serious conditions, where there is now a delay in offering appointments or indeed clinics may simply have been cancelled.

“Some people confined to their homes are likely to be drinking and smoking more, or eating less healthily, and this may also contribute to health problems, including accidents, around the home.

“Finally, there is already much known about the psychological effects of quarantine, especially when this is prolonged beyond a couple of weeks. Most of the mental health consequences will have a longer timescale and probably haven’t contributed to the excess mortality currently being seen. There don’t seem to be any data that so far suggest there is a wave of suicides. I think we might have seen more anecdotal cases in the news media if this was so. It is however quite possible that we will see an increase in suicides and self-harm over time, the longer the lockdown goes on. People with existing serious mental health issues may find the current circumstances particularly difficult, as they may have limited social networks or difficulty in accessing their usual sources of support.”


Prof Sarah Harper, Professor of Gerontology, University of Oxford, said:

“I think the differences in reported numbers is pretty well-understood and explained in the ONS document, but it is indeed interesting that there would appear to be a larger number of deaths in absolute terms in the period than the 5 year average for this period. But this need not necessarily translate into excess mortality. For example, it may be due to underlying differences in the age distribution of the population between the two periods. Even if the age distribution has only changed slightly this could lead to these increased numbers.

“If there were, however, excess mortality, it would be tempting to look at the causes of these deaths, which may then be attributed to the health services not being able to respond as quickly as normal because of COVID-19. Deaths reported as elsewhere could be homeless, or in prisons, on ships…….in other words anywhere without a “real” address.”


Prof Martin Hibberd, Professor of Emerging Infectious Disease, London School of Hygiene and Tropical Medicine, said:

“These new figures, for deaths up to 3rd of April from the ONS, clearly show the impact of COVID-19 for the first time. The weekly total for all deaths (16,387) is the highest recorded since records began in 2005 and represent a 37% increase more than expected. Reports for 21% of these deaths mentioned “novel coronavirus”. As mentioned by Chris Whitty yesterday, this weekly number will become a very important number for evaluating the impact of COVID-19 and our response, as it essentially summates all the possible effects together. We know that for some situations, we may be overplaying the role of COVID-19, for example where COVID-19 was mentioned in a death but where it may have actually played only a minor role. Whereas in other situations, a death may not mention coronavirus, even though it may have contributed, possibly as a result of the indirect consequences of the lock down and stretched healthcare services. This is the problem of getting the COVID-19 strategy balanced correctly to minimise these total figures and save as many lives as possible. We look forward to further analysis of these numbers and next week’s release of data to get a clear idea of the consequences of the COVID-19 outbreak.”


All our previous output on this subject can be seen at this weblink:


Declared interests

None received.

in this section

filter RoundUps by year

search by tag