Sickness absence in the UK labour market: 2021

• The sickness absence rate in the UK in 2021 rose to 2.2%, from a record low of 1.8% in 2020; this is the highest it has been since 2010, when it was also 2.2%.

• An estimated 149.3 million working days were lost because of sickness or injury in the UK in 2021, equivalent to 4.6 days per worker.

• The most common reason for sickness absence in 2021 was "other" conditions, including accidents, poisonings, diabetes and coronavirus (COVID-19).

• COVID-19 accounted for nearly one in four of all occurrences of sickness absence in 2021.

• The groups with the highest rates of sickness absence in 2021 included women, older workers, those with long-term health conditions, people working part-time and people working in caring, leisure and other service occupations.

The coronavirus (COVID-19) pandemic has affected the sickness absence data in many ways. While the virus may have led to additional sickness absence, measures such as furloughing, social distancing, shielding and increased homeworking helped to reduce other causes of absence in 2020. However, the reduction to the furlough scheme and its eventual end in September 2021, coupled with the decline in homeworking, shielding, social distancing policies, and new COVID-19 variants, led to a rise in sickness absence in 2021 to above pre-pandemic levels.

More information on definitions and how these are used in our estimates can be found in the Glossary

Many people in employment were on furlough during 2020. Furloughed employments, as reported in HM Revenue and Customs' Coronavirus Job Retention Scheme statistics, increased from the start of the scheme to a peak of 8.9 million employments on 8 May 2020. A year later, in May 2021, only 2.8 million employments were still on furlough and when the scheme ended on 30 September 2021, 1.2 million employments were on furlough.

The number of people in employment shielding, that is staying at home because of underlying health conditions, reduced between 2020 and 2021. Labour market statistics show that, on average, 200,000 people were shielding between July and December 2020. This decreased to 173,000 on average between January and June 2021, becoming negligible after this time.

This group of people would normally have a higher sickness absence rate than those not shielding. If fewer from this group were present in the workplace in 2020, then they would be contributing less to the number taking days off because of sickness. This could be one explanation for the reduction in the sickness absence rate in 2020. Following this, an increase in the number of these workers returning to work in 2021 could partly explain the increase in rates in 2021.

Homeworking became more prevalent in 2020 during the coronavirus pandemic, with 8.4 million people working at home in the reference week up from 4.0 million in 2019. This may have had an impact on sickness absences in 2020 and 2021. For example:

• homeworking may have allowed people to work when they were a little unwell; they might not have travelled to a workplace to work but felt well enough to work from home

• government restrictions requiring people to observe social distancing and self-isolate, may have led to less exposure to germs and minimised some of the usual sickness absences in 2020

The subsequent relaxation of coronavirus restrictions during periods of 2021 and less requirement to work from home, may have contributed to an increase in sickness absence in 2021.

Estimates showing trends in homeworking and compliance with government guidance such as wearing a face covering, social distancing or self-isolating from March 2020 to April 2022 are available within our Public opinions and social trends bulletin.

The sickness absence rate (or the percentage of working hours lost because of sickness or injury) generally declined through the 2000s and has remained relatively flat throughout the 2010s, dropping to a record low in 2020. In 2021, the rate increased to 2.2%, the highest it has been since 2010. The increase in 2021 was driven by increases seen between October and December 2021.

The number of working days lost because of sickness or injury increased in 2021, to an estimated 149.3 million working days, equivalent to 4.6 days per worker. Caution should be taken when analysing total days lost for 2020 and 2021, because of the impact of furloughed workers and policy during the coronavirus (COVID-19) pandemic.

The main reasons for sickness absence over the time series have been minor illnesses and musculoskeletal problems, but this has changed in the recent periods because of the impact of the coronavirus (COVID-19) pandemic.

In 2021, "Other" conditions (which includes COVID-19) were the most common reasons for sickness absence, accounting for 26.9% of occurrences of sickness absence. This was followed by minor illnesses, at 21.9%, musculoskeletal problems, at 13.4%, and mental health conditions, at 9.8%.

COVID-19 accounted for 24.0% of all occurrences of sickness absence in 2021, up from 13.9% in 2020. The Chartered Institute of Personnel and Development (CIPD) reported that two-thirds of organisations include COVID-19 among the top three reasons of short term absence in 2021.

Sickness absence rates have been consistently lower for men than for women since 1995, with the rates for both sexes generally falling between 1995 and 2020. Men lost 1.8% of their working hours in 2021 (an increase of 0.3 percentage points from 2020) as a result of sickness or injury and women lost 2.6% of their working hours (an increase of 0.3 percentage points from 2020).

Sickness absence rates are higher among older workers as they are more likely to develop health problems. The rate increased for all age groups from 2020 to 2021. In 2021, the sickness absence rate for those aged 65 years and over was the highest on record (at 3.2%).

In 2021, as in previous years, workers living in Wales had the highest sickness absence rate (2.8%). Those living in London had the lowest sickness absence rate (1.7%), 0.5 percentage points below the UK average.

This can be largely explained by the younger age profile of people and workers living in London and the types of occupations that they do. There is a large concentration of high-skilled jobs in London, which tend to have lower rates of sickness absence.

In 2021, sickness absence rates for public and private sector workers were 3.0% and 1.9% respectively, both up 0.3 percentage points from their 2020 rates. Sickness absence rates for public sector workers have been consistently higher than those for private sector workers for every year on record.

There are several factors to consider when examining the differences between the public and private sectors, including:

• there are differences in the types of jobs between the sectors, and some jobs have higher likelihoods of sickness than others

• workers in the private sector are more likely not to be paid sickness absence than those in the public sector

• the analysis only counts someone as sick if they work fewer hours than they are contracted for and would exclude someone who makes up lost hours at a later point in the week; individuals in smaller workforces, which are more prominent in the private sector, may be under more pressure to make up any lost hours, but no data are collected on hours made up following sickness absence

Managers and senior officials (for example, chief executives and financial managers) had the lowest sickness absence rates in 2021, at 1.1%. Workers in caring, leisure and other service occupations had the highest rates, at 3.8%. Occupational groups will have been affected differently by the coronavirus (COVID-19) pandemic, and the ability to work from home would be very different across these groups. This can be seen in our article Which jobs can be done from home? published in July 2021.

The sickness absence rate for part-time workers has been consistently higher than the rate for full-time workers. This can be explained in part by higher numbers of women working part-time, as they also tend to have higher rates of sickness absence. The sickness absence rate for part-time workers was 3.1% in 2021, a 0.5 percentage point increase on the previous year. The sickness absence rate for full-time workers was 2.0% in 2021, an increase of 0.3 percentage points on the previous year.

COVID-19

The name used to refer to the disease caused by the SARS CoV-2 virus, which is a type of coronavirus. The Office for National Statistics (ONS) takes COVID-19 to mean presence of SARS-CoV-2 with or without symptoms.

Definitions of reason for sickness absence

• "Musculoskeletal problems" includes back pain, neck and upper limb problems, and other musculoskeletal problems.
• "Other" includes the total number of days lost to accidents, poisonings, infectious diseases, skin disorders, diabetes and anything else not covered.
• "Mental health conditions" includes stress, depression, anxiety and serious mental health problems.
• "Minor illnesses" includes coughs, colds, flu, sickness, nausea and diarrhoea.
• "Disability status" is defined as whether a person meets the legal definition of being disabled according to the Equality Act 2010.
• From April 2020, interviewers were advised to code any mention of coronavirus (COVID-19) as "Other", however, it is believed people could self-report this in "Minor illnesses" or "Respiratory conditions".

Furlough

A temporary absence from work allowing workers to keep their job during the coronavirus pandemic.

Number of days lost per worker

The number of days lost per worker is proportional to the number of days lost divided by the number of persons in employment aged 16 years and over.

The total number of days lost because of sickness or injury and the number of days lost per worker both saw a significant fall in 2020. The data that feed into the total number of days lost, and days lost per worker, include furloughed workers. Therefore, some of this fall will be because fewer people are in work to be taking days off because of sickness or injury. In addition, those employed but on furlough are included in the denominator for the number of days lost per worker, as they are defined as in employment, but would not be contributing to the numerator as they are away from work.

Sickness absence rate

Sickness absence rate is proportional to the total hours lost because of sickness or injury divided by total hours multiplied by 100, as seen in the following formula.

The sickness absence rate will not be affected by furloughed workers in the same way as number of days lost. This is derived as the total hours lost as a proportion of total hours worked. While there will be less hours lost and less hours worked, those contributing to both totals are consistent, therefore this measure should still be comparable over time. Since this measure is least affected by the coronavirus pandemic policies, it is therefore the most appropriate measure to use for sickness absence analysis to ensure comparability over time.

Working day

The working day is defined as 7 hours and 30 minutes.

The estimates included in this release have been produced using the Labour Force Survey (LFS). They relate to people aged 16 years and over in employment and are for the whole of the UK. Estimates are available from 1995 onwards for most breakdowns. Estimates calculated using the LFS are based on annual averages across quarters for each calendar year: taking an average of the January to March, April to June, July to September and October to December datasets within a calendar year.

The total number of days lost is presented in millions, unless otherwise stated.

The sickness absence rate is presented as a percentage throughout these analyses.

Quality and Methodology Information reports

The Labour Force Survey (LFS) Quality and Methodology Information (QMI) reports contain important information on:

• the strengths and limitations of the data and how it compares with released data

• the uses and users of the data

• how the output was created

• the quality of the output including the accuracy of the data

More information on the quality, methodology and history of the LFS can be found in the Labour Force Survey - user guidance.

Accuracy in estimates

Estimates of sickness absence at the UK level in 2021 were based on a sample size of approximately 1,000 averaged across the year. There is therefore a limit to how many breakdowns can be made of the data without significantly reducing quality or requiring an element of disclosure control.

The numbers quoted throughout this release are based on sample sizes of three or higher. Within the main tables, any estimate based on fewer than three people has been suppressed. In instances where secondary disclosure was possible, the "Prefers not to say" category has also been suppressed. Any estimate based on a sample size of between 3 and 25 has been highlighted (with grey shading), to emphasise the potential quality limitation of the estimate. These have been applied as per our Measuring and reporting reliability of Labour Force Survey and Annual Population Survey Estimates methodology.

Because of the coronavirus (COVID-19) pandemic, a high number of people in employment were on furlough (away from work which they expect to return to) during 2020 and 2021. This has affected some of the measures of sickness absence; some of the falls seen in 2020 will be because there were fewer people in work, therefore fewer people taking days off because of sickness or injury. However, decreases in furlough in 2021 could have led to higher levels of sickness absence as people return to work.

Impact of the coronavirus pandemic on the Labour Force Survey

The LFS design is based on interviewing households over five consecutive quarters. Generally, the first of these interviews, called Wave 1, takes place face-to-face, with most subsequent interviews, for Waves 2 to 5, conducted by telephone. During March 2020, we stopped conducting face-to-face interviews, instead switching to using telephone interviewing exclusively for all waves. This initially caused a significant drop in response. New measures have been introduced to improve this, which have increased sample sizes, although they are still below normal LFS sample sizes.