City Research Online

The Impact of Covid-19 on Future Higher-Age Mortality

Cairns, A. J. G., Blake, D. ORCID: 0000-0002-2453-2090, Kessler, A.R. & Kessler, M. (2020). The Impact of Covid-19 on Future Higher-Age Mortality. London, UK: Pensions Institute.


Covid-19 has predominantly affected mortality at high ages. It kills by inflaming and clogging the air sacs in the lungs, depriving the body of oxygen ‒ inducing hypoxia ‒ which closes down essential organs, in particular the heart, kidneys and liver, and causes blood clots (which can lead to stroke or pulmonary embolism) and neurological malfunction.

Evidence from different countries points to the fact that people who die from Covid-19 are often, but not always, much less healthy than the average for their age group. This is true for England & Wales – the two countries we focus on in this study. The implication is that the years of life lost through early death are less than the average for each age group, with how much less being a source of considerable debate. We argue that many of those who die from coronavirus would have died anyway in the relatively near future due to their existing frailties or co-morbidities. We demonstrate how to capture this link to poorer-than-average health using a model in which individual deaths are ‘accelerated’ ahead of schedule due to Covid-19. The model structure and its parameterization build on the observation that Covid-19 mortality by age is approximately proportional to all-cause mortality. This, in combination with current predictions of total deaths, results in the important conclusion that, everything else being equal, the impact of Covid-19 on the mortality rates of the surviving population will be very modest. Specifically, the degree of anti-selection is likely to be very small, since the life expectancy of survivors does not increase by a significant amount over pre-pandemic levels.

We also analyze the degree to which Covid-19 mortality varies with socio-economic status. Headline statistics suggest that the most deprived groups have been disproportionately affected by Covid-19. However, once we control for regional differences in mortality rates, Covid-19 deaths in both the most and least deprived groups are also proportional to the all-cause mortality of these groups. However, the groups in between have approximately 10-15% lower Covid-19 deaths compared with their all-cause mortality.

We argue that useful lessons about the potential pattern of accelerated deaths from Covid-19 can be drawn from examining deaths from respiratory diseases, especially at different age ranges. We also argue that it is possible to draw useful lessons about volatility spikes in Covid-19 deaths from examining past seasonal flu epidemics. However, there is an important difference. Whereas the spikes in seasonal flu increase with age, our finding that Covid-19 death rates are approximately proportional to all-cause mortality suggests that any spike in Covid-19 mortality in percentage terms would be similar across all age ranges.

Finally, we discuss some of the indirect consequences for future mortality of the pandemic and the ‘lockdown’ measures governments have imposed to contain it. For example, there is evidence that some surviving patients at all ages who needed intensive care could end up with a new impairment, such as organ damage, which will reduce their life expectancy. There is also evidence that many people in lockdown did not seek a timely medical assessment for a potential new illness, such as cancer, or deferred seeking treatment for an existing serious illness, with the consequence that non-Covid-19-related mortality rates could increase in future. Self-isolation during lockdown has contributed to an increase in alcohol and drug consumption by some people which might, in turn, reduce their life expectancy. If another consequence of the pandemic is a recession and/or an acceleration in job automation, resulting in long-term unemployment, then this could lead to so-called ‘deaths of despair’ in future. Other people, by contrast, might permanently change their social behaviour or seek treatments that delay the impact or onset of age-related diseases, one of the primary factors that make people more susceptible to the virus – both of which could have the effect of increasing their life expectancy. It is, however, too early to quantify these possibilities, although it is conceivable that these indirect consequences could have a bigger impact on future average life expectancy than the direct consequences measured by the accelerated deaths model.

Publication Type: Monograph (Discussion Paper)
Subjects: G Geography. Anthropology. Recreation > GF Human ecology. Anthropogeography
H Social Sciences > HC Economic History and Conditions
H Social Sciences > HD Industries. Land use. Labor
H Social Sciences > HM Sociology
Q Science > QR Microbiology > QR355 Virology
Departments: Bayes Business School > Finance
[thumbnail of wp2007.pdf]
Text - Published Version
Download (2MB) | Preview


Add to AnyAdd to TwitterAdd to FacebookAdd to LinkedinAdd to PinterestAdd to Email


Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login