COVID-19 health outcomes

Summary

During the first wave, the UK had the fourth highest rate of excess deaths out of 33 OECD countries behind Spain, Mexico and the USA. Despite government measures to supress the virus, the UK experienced higher hospitalisations and excess deaths during its second wave. It also had one of the highest excess mortality rates among those younger than 65 in Europe.

The single biggest factor influencing differences in COVID-19 mortality between countries was the timing of restrictions. The virus spread throughout the UK before restrictions were applied due to higher levels of travel within and to the UK. Underlying population health also accounted for some of the variation in excess mortality between countries.

In the UK certain groups experienced disproportionate levels of exposure and deaths due to the virus. These included: older people, people from ethnic minority communities, disabled people and people working in certain occupations, including some keyworker roles.

Occupation, living conditions and the ability to access financial support affected risk of exposure to COVID-19. Low rates and coverage of statutory sick pay, and difficulty in accessing isolation payments reduced people’s ability to self-isolate and increased exposure. Being able to work from home reduced the exposure risk.

Once exposed, people’s pre-existing physical and mental health made them more vulnerable to severe outcomes. People with pre-existing conditions, such as diabetes, obesity, cancer, respiratory disease and underlying mental health conditions were at increased risk of death from COVID-19.

In England, for people younger than 65, the COVID-19 mortality rate for the 10% living in the most deprived areas is almost four times that of those in the least deprived areas. This partly reflects the poorer health of people living in more deprived areas.

The vaccination programme has been crucial in preventing many deaths and hospitalisations. Despite this some groups have lower access to and uptake of the vaccine.

COVID-19 health outcomes summary

The COVID-19 pandemic has had profound consequences for the health of people in the UK. Most directly – and visibly – through the numbers of deaths attributed to COVID-19 and those who have experienced severe symptoms. As explored in this section, the UK’s high COVID-19 death toll reflects a range of different factors. While some factors relate to how the pandemic was managed – such as the timing of lockdown measures – others reflect pre-existing aspects of the structure of UK society and the economy, such as the wide disparity in COVID-19 mortality rates for those younger than 65.

COVID-19 outcomes in the UK

COVID-19 was first detected in the UK in January 2020 with the first known death occurring at the beginning of March 2020. On 23 March, the UK government mandated a nationwide lockdown requiring certain businesses to close and for people to only leave their homes for essential purposes. By this time there were already 938 deaths recorded and around 5,000 hospitalisations attributed to COVID-19. Most of these deaths had occurred in England (876), with Wales (24), Scotland (29) and Northern Ireland (9) recording fewer deaths.

Box 2: COVID-19 waves and mortality

Where possible, this report uses the Office for National Statistics (ONS) definition of first and second waves. The first wave refers to the period between 24 January and 11 September 2020. The second wave refers to the period from September 12 2020. Where our analysis diverges from this definition, this is clearly flagged.

A measure of excess mortality for COVID-19 deaths in the first wave is used – this is the number of deaths in a given period over and above the number expected, such as the average number recorded in the same week of the past 5 years. This was the international gold standard used to measure mortality during the pandemic, and the measure we used for international comparisons.

This is more reliable than measuring cases of COVID-19 because testing and the reporting of deaths vary between countries. This also matters in comparing between waves in the UK, given testing was less available during the first wave and conducted mainly for hospitalisations, leaving high numbers of potential infections uncounted.

For the second wave, COVID-19 deaths are a more accurate measure of COVID-19-related mortality in the UK. This is due to increased testing and because excess mortality was reduced due to lower numbers of deaths than usual from flu during winter 2020–21.

The nationwide lockdown saved thousands of lives but by the end of the first wave the UK had experienced 137,466 hospitalisations due to COVID-19 and 58,824 excess deaths. With tight restrictions in place, the number of COVID-19 cases and deaths began to fall from mid-April 2020. By the end of June, weekly deaths were back to levels consistent with previous years.

As restrictions eased in the summer, COVID-19 cases again began to rise, leading to a second wave. Between 12 September 2020 and 30 April 2021, the UK experienced a further 96,763 COVID-19 deaths and 326,352 hospitalisations. The second wave was a more prolonged outbreak.

Figure 1: Excess deaths and deaths involving COVID-19 by week registered, per 100,000 people, UK, January 2020 to June 2021

Source: Health Foundation analysis of ONS, Weekly provisional figures on deaths registered in England and Wales; National Records of Scotland, Deaths involving coronavirus (COVID-19) in Scotland, Weekly data on Births and Deaths in Scotland; NISRA, Weekly deaths bulletin; Mid-year population estimates via NOMIS.

Experiences of COVID-19 have varied across the UK, from one nation to the next and between different regions within nations. England had the highest excess deaths during the first wave. In the second wave, the rates were more similar across the four nations except for Northern Ireland, which had a slightly lower level of excess deaths.

At first, each UK nation imposed the same rules. But over time these varied – in timing and stringency – as measures were eased at the start of June 2020. However, by the end of June, Leicester was back in lockdown and by August other parts of the country had returned to stricter measures. Between the end of the first lockdown (July 2020) and the beginning of the second lockdown (November 2020), some regions and countries in the UK had considerably higher levels of excess mortality. Northern Ireland had 36 excess deaths per 100,000, and the north east and north west of England had 19 per 100,000. London had fewer excess deaths than average, and the south west had 1.6 per 100,000, highlighting the regional variation in outbreaks during this period.

Figure 2: Excess deaths as a share of expected deaths by week registered, for regions and countries of the UK, January 2020 to June 2021

Source: Health Foundation analysis of ONS, Weekly provisional figures on deaths registered in England and Wales; National Records of Scotland, Deaths involving coronavirus (COVID-19) in Scotland, Weekly data on Births and Deaths in Scotland; NISRA, Weekly deaths bulletin; Mid-year population estimates via NOMIS.

In the north of England as well as the East Midlands, deaths began to increase from October 2020 compared with November/December in London and parts of the south of England. These regional variations and tiered restrictions continued until a new national lockdown was imposed in January 2021.

“Across the 33 countries in the OECD with comparable data, the UK had the fourth highest excess mortality rate during this period behind Spain, Mexico and the USA.”

Long COVID

The effect of COVID-19 on people’s health should not only be perceived in terms of hospitalisations and deaths. By May 2021, initial data showed that an estimated 1 million people self-reported being affected by ‘long COVID’ – ongoing symptoms persisting for more than 4 weeks after infection. Of these, 65% reported that persisting symptoms affected their ability to carry out day-to-day activities, including almost 20% who reported these activities had been limited a lot. People with a pre-existing, activity-limiting health condition, women, workers in the health and care sectors and those from more deprived backgrounds have reported higher incidences of long COVID.

Comparing UK COVID-19 outcomes internationally

Outcomes from COVID-19 during 2020 – measured by excess deaths – were significantly worse in the UK than in most other comparable OECD countries. Across the 33 countries in the OECD with comparable data, the UK had the fourth highest excess mortality rate during this period behind Spain, Mexico and the USA. In spring 2020, the UK had a higher peak in the rate of excess deaths than neighbouring or similar countries other than Spain (Figure 3).

In the autumn, excess deaths began to rise and then fall again following short-term restrictions in November, but then rose rapidly through December and January 2021. The 82,305 COVID-19 deaths between 28 November 2020 and 30 April 2021 can be attributed partly to the rapid spread of the alpha variant and a failure to introduce lockdown measures sooner. Few other countries experienced the level of excess deaths that the UK saw at the peak of its second wave.

Restrictions began to be eased again on 12 April 2021. Deaths have remained low in the UK following tight restrictions and the introduction of the vaccination programme. But the threat of emerging variants that may be more transmissible or resistant to existing vaccines remains.

Explaining the differences in COVID-19 outcomes internationally

The single biggest factor in the extent to which COVID-19 spread in the UK compared with other countries is the timing of lockdowns and stringency of restrictions. Higher levels of regional and international travel, increasing burden of disease and worsening social conditions also played a role. Once the virus had spread the extent to which different groups were affected within the UK reflected variations in underlying health and socioeconomic factors that increased risk of exposure and worse outcomes.

Figure 3: Cumulative excess mortality for select countries, January 2020 to May 2021

Source: Health Foundation analysis of Our World in Data excess mortality data, population data via OECD Stat for 2018.

Timing of lockdowns and travel

Differences in policy interventions were more important than socioeconomic variations in explaining the varying death rates among OECD countries. One study points to timing of lockdowns accounting for around 40% of the international variation in deaths – highlighting the importance of policy responses in suppressing the spread of the virus.

The extensive spread of the virus observed in the UK reflects three factors:

  • COVID-19 entered the country in a number of regions over a similar period of time
  • the UK is more densely populated than many developed countries,
  • the relatively high level of travel between different UK regions, for instance there were more than 1.7 billion domestic trips in 2019 (versus 260 million in France).

The virus arrived and spread in different places in the UK, leading to excess deaths across all regions. In other countries, the spread of the virus and deaths tended to be isolated to specific regions. For example, in Italy only three northern regions experienced high excess mortality outside of the main cities.

An ageing population and poorer underlying health

The total number of excess deaths during the first wave was highest for older age groups, with 41% of all deaths among those aged 85 and older. Once the virus had spread, the age and health of the UK population (including underlying health conditions and comorbidities) played a role in the severity of outcomes.

The UK has an ageing population, so could expect more deaths than countries with a younger age profile. But deaths among older age groups were far higher in the UK than other countries even though its proportion of people aged 80 and older is below average for a developed country. As healthy life expectancy improvements have not kept pace with life expectancy in the UK, people are living more years in poor health – particularly in older age. This could have led to greater vulnerability to COVID-19.

While there appears to be little relationship between pre-existing levels of life expectancy or healthy life expectancy and excess deaths from COVID-19 across different countries, Figure 4 shows that greater improvement in healthy life expectancy over the past decade is associated with lower COVID-19 mortality. Countries with the greatest improvements in healthy life expectancy in the previous two decades, such as Latvia and Slovenia, experienced lower excess mortality during the pandemic. A similar pattern holds for life expectancy improvements.

Improvements in healthy life expectancy are a meaningful measure of broader and historical social conditions because factors such as adequate income, good-quality jobs and housing are necessary ingredients for good health. The decline in improvement in healthy life expectancy partly reflects the erosion of these social conditions in the UK in the decade preceding the pandemic – affecting certain groups to a greater extent. This reduced the population’s opportunities for good health and is highly likely to have weakened resilience to COVID-19.

Figure 4: Change in healthy life expectancy at birth, 2010–2019 compared with excess mortality as a share of expected deaths during 2020, for people aged younger than 65, selected OECD countries

Source: The Human Mortality Database,  Short-term Mortality Fluctuations (STMF) data series; WHO Global Health Observatory data repository.

Note: Greece and Germany calculations are based on 2016-2019 data. Data from 2000 to 2018 are final and results for 2019 and 2020 are provisional. Data for the last 3 weeks of 2020 are estimated. Data for 2020 and 2021 are preliminary. Expected deaths are an average of deaths between 2015 and 2019.

Increasing burden of disease in the UK

Research within the UK,  and internationally, shows that certain underlying health conditions increase risk of more severe outcomes from COVID-19. For example, UK adults with diabetes were 1.31–1.95 times more likely (depending on blood sugar levels) to die than those without diabetes, after adjusting for age, sex and other health conditions. Similarly, those with dementia, obesity, cancer, respiratory disease, reduced kidney function or COPD had an increased risk of mortality.,  In the years leading to the pandemic, the UK experienced declines in health – particularly in the conditions associated with poor COVID-19 outcomes. In 2016, 27.8% of adults in the UK were obese, the highest in Europe. Prevalence data from the Global Burden of Disease study also show that the UK has higher age-standardised prevalence of asthma, COPD, chronic respiratory conditions and diabetes. It is lower for stroke, cardiovascular disease and chronic kidney conditions. In the decade to 2019, diabetes prevalence increased faster in the UK than in the EU.

Table 1: Age-standardised prevalence of certain health conditions associated with COVID-19 outcomes, UK and EU, 2019 and change since 2009

UK age-standardised prevalence 2019

EU age-standardised prevalence 2019

Difference

Change in the UK since 2009

Change in the EU since 2009

Cancer

2.2%

2.1%

0.1%

0.1%

0.0%

Diabetes

9.4%

6.5%

2.9%

1.9%

0.9%

Chronic kidney disease

5.7%

6.6%

-0.9%

0.0%

0.1%

Chronic obstructive pulmonary disease

4.3%

3.2%

1.1%

-0.1%

-0.1%

Chronic respiratory conditions

13.3%

8.8%

4.5%

-1.1%

-0.5%

Cardiovascular disease

6.1%

6.6%

-0.5%

0.1%

-0.1%

Asthma

10.1%

6.1%

4.0%

-1.2%

-0.5%

Stroke

0.7%

0.9%

-0.1%

-0.1%

-0.1%

Source: Institute for Health Metrics and Evaluation, Global Burden of Disease Study, 2019 via GBD Results Tool.

People with underlying mental health conditions have also had worse COVID-19 outcomes. Evidence from the UK Biobank study between 31 January and 26 July 2020 found that after adjusting for age, sex, ethnicity, deprivation, BMI, smoking status and certain comorbidities, people with existing psychiatric disorders were 2.0 times more likely to die from COVID-19 than those without. The risk ranged from 1.2 for substance misuse to 3.5 for psychotic disorders. People with more than one psychiatric diagnosis had an even higher risk of dying from COVID-19.

A range of factors contribute to worse COVID-19 outcomes among those with mental health conditions:

  • higher prevalence of other underlying health conditions, such as diabetes
  • greater likelihood of living in poor living environments
  • difficulties appraising health information
  • difficulties physically distancing or complying with social distancing restrictions
  • stigma resulting in barriers to accessing health care
  • differences in immune responses.

A working age population at greater risk

Compared with other European countries, excess mortality among people younger than 65 in the UK was second only to Bulgaria. Figure 5 illustrates the differences between some of these countries. More detailed analysis using data for within the UK shows the rate of excess deaths for 55–64 year olds (45–64 year olds in Scotland) was higher than for 65–75 year olds. As will be discussed further, higher COVID-19 mortality in this younger age group was determined by poorer pre-existing health due to socioeconomic deprivation, as well as increased risk of exposure.

Figure 5: Excess mortality by age, selected European countries, 2020

Source: ONS, Comparisons of all-cause mortality between European countries and regions: 2020.*

*Note: Latest data available for Italy is 30 October.

What made people more vulnerable to COVID-19 in the UK?

Underlying health and comorbidities are important factors in shaping outcomes from COVID-19, but they do not provide a complete picture of why the UK fared as it did. Other factors, such as someone’s sex, the type of work they do, where they live and their housing situation, also shaped people’s exposure and vulnerability to COVID-19.

Sex

In England and Wales 54% of COVID-19 deaths were among men. In Scotland and Northern Ireland deaths were evenly distributed among men and women.,  However a comparison of age-standardised rates of excess deaths shows that across 29 high income countries men were at greater risk of mortality. In England and Wales excess mortality rates were 1.5 times higher for men, 1.9 times higher in Scotland and 1.2 times higher for Northern Ireland.

“Across 29 high income countries men were at greater risk of mortality.”

Overall differences in mortality reflect that men are at a higher risk of more severe symptoms and worse outcomes. Men have higher prevalence of certain comorbidities (such as cardiovascular disease and diabetes). And there are hormonal differences too, with research showing that oestrogen can enhance the immune system while testosterone can suppress it.

Occupational risk factors

Occupation type

Among people aged 20–64, mortality differed among those working in different occupations. By the end of December 2020, the highest relative risk of mortality was for men and women working in elementary occupations, process, plant and machine operatives and in caring, leisure and other service occupations. The pattern of these COVID-19 mortality rates by occupation largely follows the pattern of all-cause mortality. This suggests that people in these occupations had poorer pre-existing health.

Figure 6 shows how risk of occupational exposure affected mortality. It compares first wave mortality rates before and after the first lockdown by occupation. Mortality rates proportionally reduced the most for people more likely to be able to work from home, or for those who tended to work in sectors that shut down.

People working in social care had a significantly higher rate of death compared with those working in health care by December 2020. For men, the death rate in health care was 44.9 per 100,000 people, rising to 79.1 per 100,000 for nurses, compared with 109.9 per 100,000 in social care. For women, the death rate for health care was 17.3 per 100,000, rising to 24.5 for nurses, and compared with 47.1 in social care.

Figure 6: Age standardised mortality rates for deaths related to COVID-19 among 20–64 year olds by occupation and sex, England and Wales, March to June 2020

Source: ONS, Coronavirus (COVID-19) related deaths by occupation, before and during lockdown, England and Wales: deaths registered between 9 March and 30 June 2020.

A study from Scotland using national data on health care workers for the period 1 March to 6 June 2020 found that patient-facing workers were 3.3 times more likely to be admitted to hospital for COVID-19 than non-patient-facing workers, after adjusting for age, sex, ethnicity, deprivation and comorbidity. Household members of patient-facing workers were also 1.79 more likely to be admitted.

Figure 7 highlights the relationship between the ability to work from home and COVID-19 mortality for those younger than 65 across local areas in England during the second wave. Areas with low rates of homeworking and high mortality rates are shown in dark red; areas with high rates of homeworking and low mortality rates are shown in dark blue. The pattern suggests that those able to work from home were able to limit their exposure to the virus and subsequently their risk of mortality. Some areas, however, had high rates of homeworking and COVID-19 mortality among those younger than 65. In London and the south east in particular this partly reflects exposure to the alpha variant in late 2020. More rural areas and the south west tended to have low rates of both.

Figure 7: Mortality rates for deaths due to COVID-19 among those younger than 65 with share of workers that work from home, England and Wales, September 2020 to May 2021

Source: ONS, ad hoc requested data on COVID-19 mortality by lower tier local authority and MSOA by age; Homeworking in the UK, work from home status.

Access to financial support

The UK government and devolved governments launched several schemes to support individuals and businesses through periods of economic closure and reduced activity. This included the Coronavirus Job Retention Scheme. By May 2021, over 11 million jobs had been furloughed, with people receiving financial support to replace their earnings while not being able to work.

For people going out to work, inadequate levels of sick pay led to people continuing to work even if unwell or exposed to COVID-19. UK statutory sick pay covers only a quarter of the average worker’s earnings and 2 million of the lowest paid workers are ineligible. Additional self-isolation payments of £500 were introduced but access to the scheme was limited – between September/November 2020 and 15 January 2021, only one-third of people applying for isolation payments were successful.

Financial incentives to keep working are also strong, with UK unemployment benefits replacing only 12% of average wages, compared with the 50–60% replacement offered by most other western European countries. Existing UK social security design is geared towards incentivising work rather than allowing a minimum standard of living for those out of work.

The CORSAIR study, which surveyed 53,880 people aged 16 and older between 2 March 2020 and 27 January 2021, found that adherence to self-isolation was low (20.2%). ‘Going  to work’ was reported as one of the most common reasons for not self-isolating.

Access to PPE

Within organisational settings, containment measures such as social distancing and access to PPE played a role in reducing deaths among key workers. For example, in the first wave there were higher deaths among social care workers compared with health care workers. Social care workers also had lower access to PPE.

In November 2020 the National Audit Office found that of the total PPE items distributed, only 14% were given to social care providers – equal to 10% of the estimated need – compared with NHS trusts, which received 81% of the PPE distributed – equal to 80% of the estimated need. The fragmentation of the social care sector further exacerbated poor coordination and lack of access to PPE for staff especially during the first wave.

Deprivation

Before the pandemic, women living in England’s most socioeconomically deprived 10% of local areas were already expected to live 19 fewer years in good health than those in the least deprived 10%. COVID-19 mortality also follows a similar pattern in deaths by deprivation seen before the pandemic.

For people younger than 65, the COVID-19 mortality rate for the 10% living in the most deprived areas is almost four times (3.7) that of those in the least deprived areas (Figure 8). This partly reflects the poorer health of people living in more deprived areas. 50–69 year olds living in the 10% most deprived areas are more than twice as likely to have two or more long-term conditions, with a similar risk differential for diabetes and chronic lung conditions. Differences in occupation (for those in work) and housing conditions also increased risk of exposure to the virus in areas with greater deprivation.

Figure 8: Age-standardised mortality rates for deaths due to COVID-19, deprivation decile relative to the least deprived decile by age, England, March 2020 to May 2021

Source: ONS, ad hoc requested data on COVID-19 mortality by lower tier local authority and MSOA by age and MySociety IMD2019 Maps Local Authority MSOA-level file.

Housing

Going into the pandemic, one in three households in England (32% or 7.6 million households) had at least one major housing problem relating to overcrowding, affordability or poor-quality housing. These factors increase exposure to poor living conditions, reduce financial resilience, and exacerbated susceptibility to COVID-19 through overcrowding.

In 2019/20, the rate of overcrowding in the private and social rented sectors in England was at its highest recorded, with 9% of social renters and 7% of private renters living in overcrowded accommodation.

Household size increases the risk of COVID-19 infection and mortality, with one study finding that household size increased the odds of a positive test by 9%. Another study found that households with nine or more residents had more than five times the risk of deaths compared with single-occupant households.

Multigenerational households are also associated with higher rates of COVID-19 infection and mortality. For example, for women aged 65 and older, the risk of mortality if living in a multigenerational household with children was 1.2 times that of people living with another adult. Research also shows that the proportion of people aged 70 and older living with people of working age within a local authority explains some of the variation in COVID-19 cases across England – suggesting the risk of exposure was greater and that those of working age may not have had the space to self-isolate in their own homes.

The COVID-19 vaccination programme

The NHS vaccination programme has been a crucial intervention in suppressing COVID-19’s spread, saving lives and marking the beginnings of recovery in the UK. The initial rollout prioritised reducing mortality. As age is a key predictor of COVID-19 mortality, the focus was on making sure the oldest people were vaccinated first along with health and social care staff and those with underlying health conditions. By 30 April 2021, almost 35 million people in the UK had received their first dose and almost 15 million had received both doses.

Analysis by Public Health England found that the vaccination programme had prevented 33,000 hospitalisations among people aged 65 and older, and 11,700 deaths among people aged 60 and older in England to the end of April 2021.

Nevertheless, there are emerging inequalities in access to and uptake of the vaccine. Figure 9 shows that by 12 April 2021, fewer people in socioeconomically deprived areas, people from ethnic minority groups, or those for whom English is not their main language have been vaccinated – despite the pandemic disproportionately affecting these groups. Only 88% of people aged 50 and older in the poorest areas had received a first dose, compared with 95% in the least deprived. Innovative approaches are being applied by local councils to close this gap, including providing information and training to community and faith leaders, interactive webinars and follow-up phone calls with people not taking up vaccinations.

Ensuring a fair recovery across all areas and groups will require concerted local and national efforts to mitigate these differences in vaccine access and uptake.

Figure 9: COVID-19 vaccination rates among people age 50 and older by socio-demographic group, England and Wales, 12 April 2021

Source: ONS, COVID-19 vaccination rates and odds ratios by socio-demographic group. Data cover those aged 50 and over up to 12 April 2021.

Previous Next