Unfortunately, your browser is too old to work on this website. Please upgrade your browser
Skip to main content

Key points

  • Unpaid carers play an important role in our society. They provide vital care, and many carers derive satisfaction and pride from their roles. However, while carers’ rights to state support are recognised under the 2014 Care Act, caring can come at a personal and financial cost.
  • Most people will have caring responsibilities at some point in their lives. More than 5 million people (9% of the population aged 5 and older) in England and Wales were providing unpaid care in 2021. 60% of carers were older than 50 and 60% of carers were women.
  • Caring relationships vary by age of carer. Carers aged up to 70 were most likely to be caring for their parents (53%), whereas more than half of those older than 70 cared for their partner.
  • Caring responsibilities can affect people’s ability to pursue paid work. We found that 4 in 10 carers under retirement age were not working as much as they might do otherwise due to their caring role. Carers providing more than 20 hours of care a week were also more likely to live in lower-income households than non-carers.
  • There are several types of state support for carers. Only a small percentage of carers approach their local authority for help: in 2021, only 8% of carers in England did so and of those 1 in 4 ended up receiving direct support. Measuring how many carers receive indirect support is much more challenging, because detailed electronic records are needed to identify carers and match them with the people they care for.
  • Both the NHS Long Term Plan and the government’s roadmap for adult social care data have recognised the need for better data to identify carers and understand whether they are being supported effectively. Improved recording and data collection would help local authorities identify their most vulnerable unpaid carers, develop a better understanding of what their needs are (and the needs of those they care for) and deliver more targeted support.
  • The Health Foundation, through its Networked Data Lab, is working with four local partners across the UK, using novel data linkages to improve the quality of their data on carers. This piece sets the scene for our work so far, and our full research findings will be published later this year.


Unpaid carers play a substantial and vital role in meeting social care needs. The care they provide has enormous value, both for the people they care for and for wider society. Many carers experience great satisfaction from their role, and through the help and support they provide to friends and family members they may also reduce the costs of formal social care provision.

At the same time, caring responsibilities can come at a high personal and financial cost, despite the 2014 Care Act giving carers the right to receive support. 1 in 5 carers report feeling socially isolated and 4 in 10 report financial difficulties because of their caring role.

Access to formal, paid social care is not easy. The adult social care workforce is bigger than the NHS workforce, but 11% of posts are unfilled, and staff experience low pay and difficult work conditions. Since 2010/11 funding for social care has not kept pace with demand and it has become harder to access care.

This means people may go without the care they need, pay more for their care, and/or turn to their family and friends for unpaid care. Increasing pressures in the NHS to discharge patients from hospital could lead to carers supporting relatives and friends with more complex needs, further adding to pressure on unpaid carers.

In this analysis, we explore national data on the number of unpaid carers and trends over time, as well as which groups are more likely to have caring responsibilities and who they provide care for. We give an overview of the types of support available to carers, and what we know – and don’t know – about how many carers are accessing support.

National data, however, are insufficient to help local areas understand the profile and needs of their carer population and to design and target the right support. There are other rich sources of information on local carers, including local administrative records from local authorities and electronic health records held by GP practices. These data sources are far from perfect, and despite ambitions to improve the recording of carer status, this is still generally under-recorded. For this reason, and due to challenges with data sharing and linkages across organisational boundaries, these data sources remain underexplored. The government’s new roadmap for better adult social care data recognises this, and includes plans to improve local data on carer support and to create a new carers survey.

Partner teams within the Networked Data Lab, funded by the Health Foundation, have been working to develop local data sources to explore key unanswered questions. This will include exploring which unpaid carers are known to local primary care and local authority services, how many receive support and how this varies between population groups. Research findings will be published later in 2023.


How many unpaid carers are there in England and Wales?

Estimates of the number of unpaid carers in the UK vary widely, drawing on different data sources, as explained below. The most reliable figures are based on the UK census, which takes place every 10 years (most recently conducted in England, Wales and Northern Ireland in 2021 and in Scotland in 2022). Almost all households complete the census, whereas other surveys usually recruit a smaller sample designed to be representative of the general population.

On the last census day (21 March 2021), 5 million people aged 5 or older were providing unpaid care in England and Wales – 1 in 10 (9%) of the population. In Northern Ireland, it was a slightly higher percentage (12%). While this figure provides a useful snapshot, it is an underestimate of the true impact of caring: other research shows that over the course of their lives, two-thirds of people in the UK will care for a loved one at some point in time, and half of all people will provide care before they turn 50.

The census also revealed that there is large geographical variation in the proportion of people who are carers, with the highest rates in Northern Ireland, Wales and the north of England (Figure 1). Reasons for this variation include differences in demographics, economic opportunities, social relationships and the availability of care services.

Figure 1

The census also showed that the percentage of people who provide unpaid care has unexpectedly decreased in the past 10 years. After adjusting for population changes during this period, the percentage of unpaid carers fell from 11% of the population in 2011 to 9% in 2021.

Against the backdrop of an ageing population and a struggling social care system, this fall may be a cause for concern if it indicates that more people with care needs might be going without support. This trend should also spark debates about what is needed to prevent a potential shortfall of unpaid caring in the future.

There may be other explanations for this apparent decrease in unpaid caring. The 2021 census was held against a backdrop of COVID-19 lockdown restrictions, which limited mixing between different households. For example, those who shared caring responsibilities with a sibling may have taken on that role alone during the pandemic. Another explanation could be the high number of COVID-19 deaths in 2020 and 2021, which may have resulted in a reduction in the overall need for care. Changes in the wording of census questions may also have played a role.

Driving this reported decrease in unpaid caring was a marked drop in the number of carers providing up to 19 hours of care per week. This suggests that, while overall fewer people now provide unpaid care than 10 years ago, carers in 2021 were more likely to provide a higher number of hours of care (Figure 2). Today’s carers may therefore need more support.

Figure 2

Making sense of different estimates of the number of carers

Results from other surveys provide a useful alternative to census figures and have the benefit of being conducted annually with consistent wording and methodology. However, these surveys return a wide range of estimates of the number of unpaid carers.

Several large, nationally representative surveys such as Understanding Society, the General Practice Patient Survey and the Health Survey for England reported a much higher number of carers than the census (Figure 3). Understanding Society, for example, identified many more carers providing up to 19 hours of care per week than the census.

Figure 3

Different surveys have varying definitions of caring and sometimes select respondents from different populations (eg only certain age groups or only users of certain services). For instance, the GP Patient Survey only interviews people registered with a GP. Other factors, such as questionnaire design and delivery mode, can also influence estimates. For example, it should be clear from a survey questionnaire that unpaid care does not include any care provided as part of paid employment.

Another barrier to identifying carers in surveys is that not everyone who provides support for friends or family thinks of themselves as a carer, and it can take a long time for people to do so. According to a Carers UK survey, half of carers (51%) took over a year to recognise their role and more than a third (36%) took over 3 years.

It is also possible that respondents who take part in the same surveys annually, including longitudinal surveys (such as Understanding Society) with modules for carers, are more sensitised to the topic of unpaid care and more willing to recognise themselves as carers. Surveys that involve interviews with health care professionals, such as the Health Survey for England led by nurses, may be helping respondents identify as carers by providing specific examples of unpaid care.


Inequalities in caring: who are unpaid carers, and who are they caring for?

Unpaid care was already known to have both a gender and an age gap, and the 2021 census confirmed that carers were more likely to be older – 60% were older than 50, compared with 38% of non-carers – and more likely to be women (60% of carers). However, the census lacks information on the nature of the caring relationship, and on the relationship between caring and people’s wider social and financial circumstances. We therefore used data from Understanding Society, a large, nationally representative, UK-wide household survey.

Our analysis showed that not only are carers themselves a diverse group, but so are their caring relationships. We found that less than half of carers (46%) provided care inside their own household, for example for a partner or child, with the majority of carers (63%) supporting someone living elsewhere, for example parents, other relatives or friends (Figure 4).

Figure 4

Age was strongly correlated with different types of caring relationships, resulting in different support needs. Carers aged up to 70 were most likely to be caring for their parents (53%), whereas more than half of carers older than 70 cared for their partner.

Caregiving where the carer and the cared-for person lived together was also more often of higher intensity: 49% provided more than 20 hours of care a week, compared with only 10% of those caring outside the household. Research by Carers UK shows that older adults caring for a partner often face their own specific set of challenges, as they are more likely to be supporting someone with mobility issues or dementia and may be managing their own health conditions at the same time.

Younger carers also face unique challenges. This group includes people with caring responsibilities for their child, for example due to disability, which applies to 1 in 5 people providing care in their household. These carers were often younger, with a median age of 47, and cared for extremely long hours, with 41% caring for more than 50 hours a week. According to another survey by NHS Digital, carers of young adults also reported lower satisfaction with social care services, and were more likely to have financial difficulties due to their role as a carer.

The number and frequency of these different types of caregiving relationships, along with other factors, will shape the support needs of the carer population in a local area. Local authorities, which have a legal obligation to support carers, therefore need to understand the profile and needs of their local carer population to provide the right balance of support.

Figure 5


How does unpaid care affect carers’ ability to earn an income?

Being an unpaid carer can have an impact on people’s ability to pursue paid work. Our analysis of Understanding Society data showed that 4 in 10 carers under retirement age were not working as much as they might otherwise due to their caring responsibilities. This was more frequently the case for carers who care for more hours, such as those caring for a child (6 in 10 working less) or those caring for multiple people (5 in 10 working less).

As a result, being an unpaid carer can have stark financial implications. Among people under retirement age, unpaid carers providing more than 20 hours of care per week were more likely to be living in lower-income households, compared with non-carers or carers providing lower intensity care (Figure 6). Importantly, this is the same group of carers that grew since the 2011 census (Figure 2). Unpaid care prevents some carers from working, and the people they care for may also be less likely to work themselves, having a dual impact on total household income.

Not only does caring often get in the way of paid work, but carers can also face additional costs relating to care equipment or higher fuel costs. A recent survey from Carers UK, covering the period affected by the current cost-of-living crisis, found that a quarter of carers (25%) were cutting back on essentials such as food or heating, almost twice as many as in 2021 (13%).

Figure 6


What support are carers entitled to and how many carers receive support?

Various types of support for carers are offered by local and central government in the UK, with the majority being subject to needs- and means-testing (Table 1). An estimated 17% of carers in England were receiving carer’s allowance from the Department for Work and Pensions (DWP) in 2021, and an additional 6% of carers applied but were considered ineligible, often because they were already receiving other state benefits. Similar estimates for Scotland will be available once their census data are released in 2023 (the DWP does not release counts on carer’s allowance claims in Northern Ireland).

The Conservative 2019 general election manifesto pledged to give unpaid carers a legal entitlement to unpaid leave from work although, as of April 2023, this bill has not yet been passed by parliament. In 2022, the Welsh government offered unpaid carers a one-off £500 payment to recognise the essential role they played during the pandemic and to help with the higher caring costs during the cost-of-living crisis.

All carers are legally entitled to receive an assessment of their support needs from their local authority. If a carer is found to have eligible needs, support may be provided to the carer directly (direct support) or to the person they care for (indirect support), as shown in Table 1.

Table 1: Types of support available to carers in the UK

Direct support Indirect support
Financial support Primary care Replacement care

Carer's allowance and carer benefits

The main source of financial support for carers comes in the form of benefits, such as the carer's allowance and other care-related components of Universal Credit and national insurance. People caring for a child with a disability are also entitled to a disability living allowance for children.

To be entitled to carer's allowance, carers must provide more than 35 hours of care a week, if the cared-for person is receiving certain benefits (eg the disability living allowance). It is reserved for carers with low incomes (less than £132 per week) and amounts to £69.70 per week. Being in full-time education or receiving other benefits can also preclude carers from receiving it. If the carer also receives Universal Credit, this amount is deducted from their Universal Credit payments.

Payments by local authorities

Local authorities can also provide carers with direct, means-tested financial support. These funds are managed by the carer or the local authority, and carers can use them to employ agency care workers, help with housework, pay for training courses or cover certain travel expenses.

Health checks, immunisations and care plans

By registering as a carer with their GP practice, carers can also benefit from annual carer health checks, access to immunisations and help with implementing care plans.

Services for the cared-for person

Indirect support from local authorities helps carers by providing services for the cared-for person. Such services could include providing a care worker or arranging a temporary stay in a care home or day centre (respite care).

This type of carer support may arise from a carer assessment (when a carer requests support) or not (for example if the cared-for person is referred to adult social care services after a hospital stay).


We only have access to data on how many carers request and receive support from their local authority for England. Based on local authority finance reports, we know that only a small percentage of carers in England approached their local authority for help: in 2021, only about 8% of carers did so (Figure 7). In many cases, carers may be coping well without formal support and therefore do not reach out, but in other cases carers with support needs are not being properly signposted to services.

Figure 7

In England, despite a strong policy focus on improving support for carers since the 2014 Care Act, accessing support remains difficult because of strict eligibility requirements for financial support and budget pressures on local authorities. Of the carers who did request support in 2021/22, only 27% ended up receiving direct support. This is a lower rate than when the data series started in 2016/17, when 31% of requests ended in direct support (Figure 8). This is likely to be a consequence of local authority gross expenditure on services for carers decreasing by 11% in real terms between 2015/16 and 2020/21.

Figure 8

It is much more difficult to estimate how many carers receive indirect support (where support is provided to the cared-for person), as this would require records that allow us to match carers to the person they care for. The new client-level adult social care data collection from local authorities, which is currently being piloted and will become mandatory from April 2023, may support this kind of analysis.


What additional answers can local datasets provide?

Improvements in recording and data collection could help deliver enormous benefits. With better data, local authorities could identify their most vulnerable unpaid carers, develop a better understanding of what their needs are (and the needs of those they care for) and deliver more targeted support. We recommend that local and national decision makers better utilise their local data sources, to commission services based on their local population’s needs.

Through the Networked Data Lab, the Health Foundation is working to answer two pressing research questions using local datasets:

  • How many carers are accessing local authority support, and is access equitable? Despite advances in data collection in social care, we still don’t know how many carers are being indirectly supported by local authorities (through support for the cared-for person) because electronic records do not usually link carers to the cared-for person. We also don’t know if carers in more deprived areas are less able to access support.
  • Who is known to local services as a carer, and who isn’t? We know that the use of voluntary markers to identify unpaid carers in electronic records is still underdeveloped, which means that NHS and local authority teams can’t identify and reach out to the carers they have a duty to support.

These two questions are difficult to answer because linkages between NHS data and local authority data require cross-sector working, and personal identifiers such as NHS numbers are not often recorded in non-health data.

Our four local teams in Liverpool and Wirral, Wales, Leeds and north-west London are working with GP records and carer assessment data from local authorities to answer these questions. Unpaid carers can register with their GP and they can also request support from their local authority, who are mandated to conduct carers’ assessments. Both of those activities leave a trail of data which isn’t routinely used for research purposes and will help our analysts build a picture of who unpaid carers are, what health and care services they are accessing and who is – and isn’t – accessing the support they need.

The analytical tools developed to analyse these data will also be packaged for use by other local areas, which should be particularly useful once the new client-level social care data pilot scheme is expanded nationally. Using these novel data linkages, the findings from the Networked Data Lab’s research will be published later in 2023.

Further reading

Kjell-bubble-diagramArtboard 101 copy

Get social

Follow us on Twitter
Kjell-bubble-diagramArtboard 101

Work with us

We look for talented and passionate individuals as everyone at the Health Foundation has an important role to play.

View current vacancies
Artboard 101 copy 2

The Q community

Q is an initiative connecting people with improvement expertise across the UK.

Find out more