Unpaid care and protected characteristics, England and Wales: Census 2021

• On Census Day 2021 (21 March 2021), more unpaid carers were disabled (27.5% and 29.8% in England and Wales, respectively) compared with non-carers (17.8% and 21.4% in England and Wales, respectively).

• In unpaid carers, the most common ethnic group was “White: English, Welsh, Scottish, Northern Irish or British” in both England (78.3%) and Wales (92.9%).

• In unpaid carers in England and Wales, the most common religion was “Christian” with 48.7% and 45.5% reporting this as their religion in England and Wales, respectively.

• The majority of unpaid carers aged 16 years and over were “married or in a registered civil partnership” in England and Wales (50.9% in England and 50.9% in Wales), compared with non-carers where 44.8% and 42.6% in England and Wales, respectively, were “married or in a registered civil partnership”.

• More unpaid carers aged 16 years and over identified with a lesbian, gay, bisexual or other (LGB+) sexual orientation in both England (3.9%) and Wales (4.0%) compared with non-carers (3.0% in both countries).

Census 2021 asked “Do you look after, or give any help or support to, anyone because they have long-term physical or mental health conditions or illnesses, or problems related to old age?”. People were asked to exclude anything they did as part of their paid employment. This is different from the 2011 Census question, which began “Do you look after, or give any help or support to family members, friends, neighbours or others”. Further information about changes to the unpaid care question between 2011 and 2021 can be found in Section 10: Question changes.

In 2021, people self-identified as an unpaid carer or non-carer by selecting one of six categories:

• No
• Yes, 9 hours a week or less
• Yes, 10 to 19 hours a week
• Yes, 20 to 34 hours a week
• Yes, 35 to 49 hours a week
• Yes, 50 or more hours a week

Please note that Census 2021 was conducted during the coronavirus (COVID-19) pandemic, for more information please see Section 11: Data Sources and Quality.

Age-standardised and age-specific percentages are used within this article. Age-standardised percentages account for different age structures in populations and are more appropriate than crude percentages when drawing comparisons over time and across areas. The numbers being reported here are the actual number who responded in each category. You can download both age-standardised and non-age-standardised data in the accompanying datasets. For further information on age-standardised percentages, see our Age standardising data: What does this mean and why does it matter? blog.

In line with the Equality Act (2010), people who assessed their day-to-day activities as limited by long-term physical or mental health conditions or illnesses were considered disabled. Those who stated that they had a condition which did not limit their day-to-day activities or had no condition were considered non-disabled. Further information on how we measured disability is available in Section 9: Glossary.

Figure 1: More unpaid carers were disabled compared with non-carers

Unpaid care (broken down by hours of care provided) by disability, England and Wales, 2021

Notes:

1. Age-standardised percentages have been calculated using counts rounded to the nearest 5. 
2. Graph represents usual residents aged 5 years and over. 
3. All figures are individually rounded; totals may not sum exactly because of this rounding. 

Download the data

.xlsx

More unpaid carers were disabled (27.5% and 29.8% in England and Wales, respectively) compared with non-carers (17.8% and 21.4% in England and Wales, respectively).

Unpaid carers providing 50 or more hours of unpaid care per week had the highest percentage of disabled people in England (15.6% limited a lot, 19.8% limited a little in their day-to-day activities) and Wales (16.8% limited a lot, 20.0% limited a little in their day-to-day activities).

Figure 2: More unpaid carers were disabled compared with non-carers in every age group apart from those aged 90 years and over

Unpaid care by age and disability, England and Wales, 2021

Notes:

1. Age-specific percentages have been calculated using counts rounded to the nearest 5. 
2. Graph represents usual residents aged 5 years and over. 
3. All figures are individually rounded; totals may not sum exactly because of this rounding.   

Download the data

.xlsx

There was a higher percentage of unpaid carers with a disability, compared with non-carers, across all ages apart from those aged 90 years and over in both England and Wales. This difference was most marked in the younger age groups. For example, in England, 23.6% of young carers aged between 5 and 17 years were disabled, compared with 8.0% of people aged between 5 and 17 years who did not provide unpaid care. In Wales, 20.9% of young carers aged between 5 and 17 years were disabled, compared with 8.6% of people aged between 5 and 17 years who did not provide unpaid care.

Comparisons with the 2011 Census

This section compares the prevalence of unpaid carers who were disabled and non-disabled in 2021 and 2011. While the disability and unpaid care questions in each census had some differences, they remain broadly comparable across the time periods. Further information on question changes can be found in Section 10: Question changes. 

The percentage of unpaid carers with a disability increased between 2011 and 2021. This is despite an overall reduction in the proportion of people with a disability between the 2011 Census and Census 2021. The increase is mainly driven by an increase in the percentage of disabled unpaid carers aged 54 years and under between 2011 and 2021, while there was a decrease in the percentage of disabled unpaid carers aged 60 years and over. The largest increase is evident in young unpaid carers aged between 5 and 17 years, where 9.4% in England and 9.5% in Wales were disabled in 2011 compared with 23.6% in England and 20.9% in Wales in 2021. In line with this, in the general population there have been increases in disability prevalence between 2011 and 2021 in the younger age groups and decreases in the older population. For more information please see our Disability by age, sex and deprivation, England and Wales: Census 2021 article.

Further breakdowns by disability are available in the accompanying dataset.

The ethnic group question has two stages. First, a person identifies through one of the following five high-level ethnic groups:

• Asian, Asian British, Asian Welsh
• Black, Black British, Black Welsh, Caribbean or African
• Mixed or Multiple
• White
• Other ethnic group

Second, a person identifies through one of the 19 available response options, which include categories with write-in response options. This analysis will look at the 19 ethnic groups.

The ethnicity of unpaid carers largely follows the ethnic-group distributions in the whole population. In people who identified as unpaid carers and non-carers, the most common ethnic group identified with was "White: English, Welsh, Scottish, Northern Irish or British" in both England (78.3% and 73.8% in unpaid carers and non-carers, respectively) and Wales (92.9% and 90.2% in unpaid carers and non-carers, respectively).

In England, the least common ethnic groups reported in people who identified as unpaid carers were “White: Roma” (0.1%) and “White: Gypsy or Irish Traveller” (0.2%). These were also the least common in non-carers (0.2% and 0.1% in the “White: Roma” and “White: Gypsy or Irish Traveller” ethnic groups, respectively).

In Wales, in people who identified as unpaid carers, the least common ethnic groups identified with were “Black, Black British, Black Welsh, Caribbean or African: Caribbean”, “Black, Black British, Black Welsh, Caribbean or African: Other Black”, and “White: Roma” (0.1% in all three groups). These were also the least common reported in non-carers alongside “White: Gypsy or Irish Traveller” (0.1% in all four groups).

We have also analysed the difference in unpaid care provision within each ethnic group.

Figure 3: Of all ethnic groups, the “White: Gypsy or Irish Traveller” ethnic group had the highest percentage of unpaid carers in females and males

Nineteen category ethnic group by unpaid care status and sex, England and Wales, 2021

Notes:

1. Age-standardised percentages have been calculated using counts rounded to the nearest 5.
2. Graph represents usual residents aged 5 years and over. 
3. All figures are individually rounded; totals may not sum exactly because of this rounding.  

Download the data

.xlsx

The “White: Gypsy or Irish Traveller” ethnic group had the highest percentage of people who identified as unpaid carers; 15.4% of females and 10.7% of males in England and 16.3% of females and 11.9% of males in Wales. The “Asian, Asian British or Asian Welsh: Chinese” ethnic group had the lowest percentage of people who identified as unpaid carers in females (5.6%) and males (4.5%) in England. In Wales, the “Asian, Asian British or Asian Welsh: Chinese” ethnic group had the lowest percentage of people who identified as unpaid carers in females (5.7%) and the “Asian, Asian British or Asian Welsh: Other Asian” had the lowest percentage of people who identified as unpaid carers in males (4.9%).

Comparisons with the 2011 Census

Similar patterns were seen across both the 2021 and the 2011 Census. There was no “White: Roma” ethnic group tick-box in 2011 so it is not possible to know how people who identified as “White: Roma” in 2021 defined their ethnic group in 2011.

Further breakdowns by ethnic group are available in the accompanying dataset.

In the census data, religion refers to a person’s religious affiliation. This is the religion with which they connect or identify, rather than their beliefs or active religious practice. As the question is voluntary, we should be cautious when comparing figures between different areas or between censuses because of varying response rates.

In unpaid carers and non-carers in England, the most common religion identified with was “Christian” (48.7% and 47.6% in unpaid carers and non-carers, respectively). The least common were “Buddhist” (0.4% and 0.5% in unpaid carers and non-carers, respectively) and “Jewish” (0.5% in unpaid carers and non-carers).

In Wales, the most common religions identified with in unpaid carers were “Christian” (45.5%) and “No religion” (45.4%), and the most-common religion in non-carers was “No religion” (46.6%) followed by “Christian” (43.6%). The least common religions were “Jewish” and “Sikh” (0.1% in unpaid carers and non-carers).

We have also analysed the difference in unpaid care provision within each religion.

Figure 4: Of all religions, people who identified with an “Other religion” had the highest percentage of unpaid carers in females and males

Religion by unpaid care and sex, England and Wales, 2021

Notes:

1. Age-standardised percentages have been calculated using counts rounded to the nearest 5.
2. Graph represents usual residents aged 5 years and over. 
3. All figures are individually rounded; totals may not sum exactly because of this rounding.  

Download the data

.xlsx

People who identified with an “Other religion” had the highest percentage of unpaid carers; 14.7% and 11.4% in females and males, respectively, in England and 17.4% and 13.6% in females and males, respectively, in Wales. In England, those who described their religion as “Buddhist” had the lowest percentage of unpaid carers with 8.4% and 7.0% in females and males, respectively. In Wales, people identifying as “Hindu” had the lowest percentage of unpaid carers with 7.5% and 5.9% in females and males, respectively.

Comparisons with the 2011 Census

The "Christian" group was most common in unpaid carers in England and Wales in 2011, similar to the pattern seen in 2021. In both countries, there has been an increase in the percentage of unpaid carers identifying with "No religion" in 2021 compared with 2011, this religion pattern has also been seen in the usual resident population.

Further breakdowns by religion are available in the accompanying dataset.

Same-sex marriage was introduced in 2014 and opposite-sex civil partnerships were introduced in 2019. The questions on marriage and civil partnerships were updated for 2021 to reflect this. This analysis looks at usual residents aged 16 years and over.

Figure 5: There was a higher percentage of unpaid carers who were “Married or in a registered civil partnership” compared with non-carers

Unpaid care (broken down by hours of care provided) by legal partnership status, England and Wales, 2021

Notes:

1. Age-standardised percentages have been calculated using counts rounded to the nearest 5. 
2. Graph represents usual residents aged 16 years and over. 
3. All figures are individually rounded; totals may not sum exactly because of this rounding.  
4. Marriages of same-sex couples have been possible in England and Wales from 29 March 2014. 
5. The first civil partnerships (CPs) between opposite-sex couples took place in England and Wales on 31 December 2019. 

Download the data

.xlsx

The breakdown of marital status in unpaid carers is broadly similar to the breakdown of marital status in the population. The majority of unpaid carers aged 16 years and over were "married or in a registered civil partnership" in both England and Wales. People who provided 50 or more hours of unpaid care had the highest percentage of people who were "married or in a registered civil partnership" in England (53.0%) and Wales (53.5%). Non-carers had the lowest percentage of people who were "Married or in a registered civil partnership" in England (44.8%) and Wales (42.6%) and this was lower than in all unpaid care hours categories.

Comparisons with the 2011 Census

Between 2011 and 2021, there has been an increase in the percentage of unpaid carers who were "Never married and never registered a civil partnership" in England (30.0% in 2011 and 36.5% in 2021) and Wales (30.3% in 2011 and 36.8% in 2021). The percentage of unpaid carers who were "Married or in a registered civil partnership" decreased the most out of all legal partnership statuses between 2011 and 2021 in England (56.6% in 2011 and 50.9% in 2021) and Wales (56.8% in 2011 and 50.9% 2021). This marital status pattern has also been seen in the usual resident population.

Further breakdowns by marital status are available in the accompanying dataset.

The census question on sexual orientation was new for 2021; it was a voluntary question asked of those aged 16 years and over.

Figure 6: There was a higher percentage of unpaid carers who identify with a lesbian, gay, bisexual or other (LGB+) sexual orientation compared with non-carers

Unpaid care (broken down by hours of unpaid care provided) by sexual orientation, England and Wales, 2021

Notes:

1. Age-standardised percentages have been calculated using counts rounded to the nearest 5. 
2. Graph represents usual residents aged 16 years and over. 
3. All figures are individually rounded; totals may not sum exactly because of this rounding.  

Download the data

.xlsx

Overall, 3.9% of unpaid carers in England and 4.0% of unpaid carers in Wales, identified with a “gay or lesbian”, “bisexual”, or “other sexual orientation” (LGB+) orientation, compared with 3.0% in non-carers in both countries. Unpaid carers who provide nine hours or less of unpaid care had the highest percentage of people who identified with an LGB+ orientation.

We have also analysed the difference in unpaid care provision by sexual orientation.

Figure 7: In people who identified with a lesbian, gay, bisexual or other (LGB+) sexual orientation there was a higher percentage of unpaid carers, compared with those who identified with a straight or heterosexual orientation, in all age groups

Sexual orientation by age and unpaid care status, England and Wales, 2021

Notes:

1. Age-specific percentages have been calculated using counts rounded to the nearest 5. 
2. Graph represents usual residents aged 16 years and over. 
3. All figures are individually rounded; totals may not sum exactly because of this rounding.    

Download the data

.xlsx

Across all age-groups in England and Wales, people who identified with an LGB+ orientation provided more unpaid care compared with people who identified with a straight or heterosexual orientation. This difference was not statistically significant in people aged between 60 to 64 years and 65 years and over in Wales.

Age-specific percentage

Age-specific percentages are used to allow comparisons between specified age groups. 

Age-standardised percentage

Age-standardised percentages allow for comparison between populations over time and across geographies, as they account for differences in the population size and age structure. The 2013 European Standard Population is used to standardise proportions. This is a hypothetical population and assumes that the age structure is the same in both sexes, therefore allowing comparisons to be made between the sexes as well as between geographical areas and over time. The standard populations are listed in Annex F of the Revision of the European Standard Population: Report of Eurostat's task force and Revised European Standard Population: 2013 ESP.

For further information on age-standardised percentages, see our Age standardising data: What does this mean and why does it matter? blog.

To calculate an age-standardised percentage, an age specific percentage is first calculated for each age group:

where:

The age-standardised percentage of unpaid carers is that which would have occurred if the observed age-specific percentage of unpaid carers had applied in the European Standard Population (ESP).

thus:

Age-standardised percentage =

where:

Census Day 

The census is carried out every 10 years and gives us the most accurate estimate of all the people and households in England and Wales. In 2021, the census was conducted on 21 March. At this time, most of the coronavirus (COVID-19) lockdown restrictions were still in place in England and Wales. The coronavirus pandemic may have affected estimates of people providing unpaid care.

Disability

In Census 2021, people who assessed their day-to-day activities as limited by long-term physical or mental health conditions or illnesses were considered disabled. This definition of a disabled person meets the Government Statistical Service harmonised standard for measuring disability and is in line with the Equality Act (2010). This requires that a person has a physical or mental impairment, and that the impairment has a substantial and long-term adverse effect on a person's ability to carry out day-to-day activities. Therefore, those reporting conditions that limited their day-to-day activities a little or a lot were classified as disabled. People who had no long-term physical or mental health conditions, or who had conditions that did not limit their day-to-day activities were classified as non-disabled. Details on changes to questions between 2011 and 2021 are given in Section 10: Question changes.

For more information on the disability variable please see Disability variable: Census 2021.

Ethnic group

The ethnic group that the person completing the census feels they belong to. This could be based on their culture, family background, identity or physical appearance.

In 2011, respondents could choose 1 out of 18 tick-box response categories. In 2021 respondents could choose 1 out of 19 tick-box response categories as a new Roma category was added next to the Gypsy or Irish Traveller tick-box within the White category.

For more information on the ethnic group variable, please see Ethnic group variable: Census 2021.

Legal partnership status

Classifies a person according to their legal marital or registered civil partnership status on Census Day, 21 March 2021.

It is the same as the 2011 Census variable “Marital status” but has been updated for Census 2021 to reflect the Civil Partnership, Marriages and Deaths (Registration etc) Act 2019, which made provision for the extension of civil partnerships to couples who are not of the same sex.

For more information on the legal partnership status variable, please see Legal partnership status variable: Census 2021.

LGB+

An abbreviation used to refer to people who identify as lesbian, gay, bisexual and other minority sexual orientations (for example, asexual).

Religion

The religion people connect or identify with (their religious affiliation), whether or not they practise or have belief in it.

This question was voluntary, and the variable includes people who answered the question, including "No religion", alongside those who chose not to answer this question.

This variable classifies responses into the eight tick-box response options. Write-in responses are classified by their "parent" religion affiliation, including "No religion", where applicable. Percentages have been calculated out of the overall population as opposed to out of the population who answered the religion question. This aids comparison across time and between areas, as the percentage of the population who answer the question varies.

For more information on the religion variable, please see Religion variable: Census 2021.

Sexual orientation

Sexual orientation is an umbrella term covering sexual identity, attraction, and behaviour. For an individual respondent, these may not be the same. For example, someone in an opposite-sex relationship may also experience same-sex attraction, and vice versa. This means the statistics should be interpreted purely as showing how people responded to the question, rather than being about whom they are attracted to or their actual relationships.

We have not provided glossary entries for individual sexual orientation categories. This is because individual respondents may have differing perspectives on the exact meaning.

The question was voluntary and was only asked of people aged 16 years and over. For more information on the sexual orientation variable please see Sexual orientation variable: Census 2021.

Statistically significant

The term "significant" refers to statistically significant changes or differences. Significance has been determined using the 95% confidence intervals, where instances of non-overlapping confidence intervals between estimates indicate the difference is unlikely to have arisen from random fluctuation. See our statistical uncertainty page.

Unpaid care  

A person is a provider of unpaid care if they look after or give help or support to anyone because of long-term physical or mental health conditions or illnesses, or problems related to old age. This does not include any activities as part of paid employment. No distinction is made about whether any care that a person provides is within their own household or outside the household.  

The wording of the question was changed in 2021 and used the Government Statistical Service (GSS) harmonised standard for unpaid care. Information on steps taken to consult on question changes can be found in our Health and unpaid care question development for Census 2021.  

For more information on the unpaid care variable, please see Provision of unpaid care variable: Census 2021.

Usual resident  

A usual resident is anyone who on Census Day, 21 March 2021, was in the UK and had stayed or intended to stay in the UK for a period of 12 months or more or had a permanent UK address and was outside the UK and intended to be outside the UK for less than 12 months. 

Unpaid care

The unpaid care question was first asked in the 2001 Census.

In Census 2021 we asked:

“Do you look after, or give any help or support to, anyone because they have long-term physical or mental health conditions or illnesses, or problems related to old age?

Exclude anything you do as part of your paid employment.”

People answered the 2021 question by selecting one of six categories:

• No
• Yes, 9 hours a week or less
• Yes, 10 to 19 hours a week
• Yes, 20 to 34 hours a week
• Yes, 35 to 49 hours a week
• Yes, 50 or more hours a week

In the 2001 and 2011 Census we asked: “Do you look after, or give any help or support to family members, friends, neighbours or others because of either:

• long-term physical or mental ill-health/disability?
• problems related to old age?

Do not count anything you do as part of your paid employment.”

People answered the 2001 and 2011 question by selecting one of four categories:

• No
• Yes, 1 to 19 hours a week
• Yes, 20 to 49 hours a week
• Yes, 50 or more hours a week

The question change in 2021 was based on research and testing, see our Health and unpaid care question development for Census 2021 report. To make comparisons between 2011 and 2021 data, some of the answer categories for 2021 have been grouped to match to the four answer categories for 2011. It should be noted that when the "provides 9 hours or less" and "provides 10 to 19 hours" of unpaid care is combined in 2021, this creates a slightly different category ("provides 19 hours or less") to the category this is compared with in 2011 ("provides 1 to 19 hours of unpaid care").

Disability

In 2021, people were asked “Do you have any physical or mental health conditions or illnesses lasting or expected to last 12 months or more?”. If they answered “yes”, a further question “Do any of your conditions or illnesses reduce your ability to carry out day-to-day activities?” was asked. The three response options were:

• Yes, a lot
• Yes, a little
• Not at all

Combining responses to these two questions enabled identification of disabled people who are either limited a little or limited a lot in their day-to-day activities. It also enabled identification of a non-disabled group who had no conditions or illnesses and a non-disabled group who had a condition or illness that did not limit their day-to-day activities. This aligns with the Equality Act (2010) definition of disability, which requires that an impairment has an adverse effect on a person's ability to carry out day-to-day activities.

In 2011, people were asked “Are your day-to-day activities limited because of a health problem or disability which has lasted, or is expected to last, at least 12 months? Include problems related to old age”. The three response options were:

• Yes, limited a lot
• Yes, limited a little
• No

Caution should be taken when examining trends of disability over time, because of the question changes. Nonetheless, comparison with previous censuses, when appropriately caveated, can be useful for policymakers, and as such are included in this release.

To allow comparisons between 2021 and 2011, we compared those who answered “No” in 2011 with those considered as non-disabled in 2021.

The wording of the question was changed in 2021 to more closely align with the Equality Act (2010) and the Government Statistical Service's harmonised standard. Further, reference to mental health was included and the visible prompt "include problems related to old age" was removed. Information on steps taken to consult on question changes can be found in our Health and unpaid care question development for Census 2021 report.

The census provides the most detailed picture of the entire population, with the same core questions asked to everybody across England and Wales. Census results can be more reliable than survey results based on a sample of the population, because the whole population is included. The UK Statistics Authority has assigned National Statistics status to Census 2021 outputs, providing assurance that these statistics are of the highest quality and value to users.   

Census 2021 achieved a very high response rate of 97%. We ensure the census results reflect the whole population by using statistical methods to estimate the number and characteristics of people who were not recorded on a census response. This means that the census statistics are estimates rather than simple counts of responses, so they have some statistical uncertainty associated with them. We take numerous steps to minimise possible sources of error.

Additionally, we apply statistical disclosure control to protect the confidentiality of census respondents. Differences in the methods used for statistical disclosure control may result in minor differences in data totals between census products. As we round all figures individually, table totals may not sum exactly.  

Quality considerations along with the strengths and limitations of Census 2021 more generally are provided in our Quality and Methodology Information (QMI) for Census 2021. Read more about the specific quality considerations for unpaid care, health and disability. Further information on our quality assurance processes is provided in our Maximising the quality of Census 2021 population estimates report.

In England and Wales combined, an estimated 5.0 million usual residents aged 5 years and over provided unpaid care in 2021. This is a percentage of 9.0%, a statistically significant decrease from 11.4% in 2011. This could be because of:

• coronavirus (COVID-19) guidance on reducing travel and limiting visits to people from other households
• unpaid carers who previously shared caring responsibilities may have taken on all aspects of unpaid care because of rules on household mixing during the pandemic
• there has been an increase in the percentage of people reporting very good health and a decrease in the percentage of people that were disabled in 2021 compared with 2011, which could have led to a reduction in the need for unpaid care
• excess deaths were highest in the older population and peaked at the beginning of 2021 which could have led to a reduction in the need for unpaid care
• changes in the question wording between 2011 and 2021 may have had an impact on the number of people who self-reported as unpaid carers

These differences should be considered when comparing previous census data used within this release.

Further analysis on gender identity will be included in subsequent articles following additional work we are undertaking. For more information, read our Gender identity in Census 2021 statement and our Sexual orientation and gender identity quality information for Census 2021 methodology.