Official Statistics

Survey of adult carers in England, 2025 to 2026: data quality report

Published 24 September 2026

Applies to England

Introduction

The survey of adult carers in England (SACE) is carried out every 2 years.

This document presents the data quality of SACE, 2025 to 2026 submitted by councils with adult social services responsibilities (CASSRs), which will be referred to as local authorities throughout this release. These outputs are published as official statistics.

This statement should be used in conjunction with the guidance for the survey of adult carers in England 2025 to 2026 and the following documents published as part of this statistical release:

• commentary
• question responses, time series and quality of life data tables
• data quality tables
• methodology

How these statistics can be used

These statistics can be used to:

• understand the experiences and perceptions of carers
• benchmark local authorities against each other
• gain insight into carers
• measure and monitor local performance
• gain further insight alongside other local data
• inform policy development

These statistics cannot be used to make judgments:

• about how effective local authorities are at providing social care and support services for carers
• on the social care sector as a whole

Data presentation

Percentages are rounded to one decimal place and may therefore not add up to 100%.

For presentational purposes, figures are rounded to the nearest 5. Estimated totals are rounded to the nearest 10. Exceptions to this are:

• where tables present data relating to numbers of local authorities
• to reconcile data quality issues where it is not possible to identify the service users nor infer any additional information

Relevance

The data released with this publication is used by:

• central government departments to monitor the impact of social care policy
• local government bodies to assess performance in relation to their peers

The data is also available for use by:

• researchers looking at local authority performance
• service users and the public to hold local authorities and the government to account

The data has also been used previously by the Care Quality Commission (CQC) for their annual performance assessment.

Assessment of user needs and perceptions

The survey and associated data collection were developed in collaboration with the Social Services User Survey Group (SSUSG). The survey content was included in 2 consultations:

• a consultation on adult social care data developments in 2012
• a consultation on the data requirements for the safeguarding adults return and adult social care user and carer surveys in response to the Care Act in 2014

Accuracy

Submissions and exempt local authorities

Returns for SACE 2025 to 2026 were submitted by 152 of the 153 local authorities. All 151 local authorities that are required to provide data did so.

The Isles of Scilly and City of London were exempt from the 2025 to 2026 survey as the number of carers within their area who met the survey eligibility criteria was too small (less than 150) to guarantee statistically robust results.

The City of London chose to still undertake the survey, and its data is included in this release and accompanying data tables. However, care should be taken when analysing and interpreting results that compare the City of London with other local authorities due to its small eligible population.

Missing data

The data quality tables provide an overview of the level of missing administrative data in local authority submissions. The administrative data is the data completed by the local authority on each of the carers in their sample. These tables provide an overview of the level of missing administrative data for:

• all those in the sample
• for those that responded to the survey

Survey responses

Across all local authorities, an eligible population of 311,660 carers was reported in 2025 to 2026. A sample of 153,455 (49.2%) carers were sent a questionnaire, of which:

• 43,460 (28.3%) completed and returned the questionnaire (response)
• 106,195 (69.2%) did not return the questionnaire (non-response)
• 3,800 (2.5%) returned a blank questionnaire (non-response)

Table 1: overall response rate achieved for the survey of adult carers in England by survey year

Year of survey	Response rate (%)
2012 to 2013	45.6
2014 to 2015	43.8
2016 to 2017	40.7
2018 to 2019	37.3
2021 to 2022	32.5
2023 to 2024	29.6
2025 to 2026	28.3

The data in table 1 can be calculated using table 5 of the accompanying ‘Survey of adult social care in England 2025 to 2026: data quality tables’ on the Survey of adult carers in England: 2025 to 2026 report page.

Table 1 shows that response rates have declined each year SACE has been carried out. There was variation in the response rates achieved for different questions and different local authorities. The data quality tables provide an overview of the response rates for each question submitted by each local authority.

Table 2: summary of overall response rates for local authorities, England 2025 to 2026

Response rate	Number of local authorities
Less than 20%	18
20% to 25%	22
25% to 30%	36
30% to 35%	36
35% to 40%	26
40% to 45%	11
More than 45%	3

The data in table 2 can be found in table 3 of the accompanying ‘Survey of adult carers in England 2025 to 2026: data quality tables’ on the Survey of adult carers in England: 2025 to 2026 report page.

Table 2 shows a summary of the overall response rates for the 152 local authorities that took part in SACE 2025 to 2026. The table shows that 18 local authorities had a response rate of less than 20%. Due to this data being grouped using unrounded figures, the figures represented in table 2 may not match exactly those derived from aggregating the relevant column in the data quality tables.

Demographic profile of carers

Local authorities are required to complete a data return with administrative data on all the carers in their sample. If a carer responds to the survey, then their questionnaire responses are then added to the data return.

Of the 153,455 carers who were sent a questionnaire:

• gender was known for 145,135 carers (94.6%), of which 101,650 were female (70.0%), 43,355 were male (29.9%) and 130 had ‘other’ gender (0.1%)
• age was known for 137,400 carers (89.5%), of which 77,925 (56.7%) were 18 to 64 years old and 59,475 (43.3%) were 65 and over
• ethnic group was known for 122,000 carers (79.5%) of the sample, of which 98,345 (80.6%) were White and 23,655 (19.4%) were non-White

Of the 43,460 carers who responded to the survey:

• gender was known for 43,110 carers (99.2%), of which 29,855 were female (69.3%), 13,230 were male (30.7%) and 28 had ‘other’ gender (0.1%)
• age was known for 42,325 carers (97.4%), of which 17,480 (41.3%) were 18 to 64 years old and 24,840 (58.7%) were 65 and over
• ethnic group was known for 39,495 carers (90.9%) of the sample, of which 34,100 (86.3%) were White and 5,395 (13.7%) were non-White

Note that figures may not add up to totals due to rounding.

Bias

Random sources of bias: confidence intervals and margin of error

Surveys produce statistics that are estimates of the real figure for the whole population which would only be known if the entire population was surveyed. Therefore, estimates from the sample surveys are always surrounded by a confidence interval which assesses the level of uncertainty caused by only surveying a sample of carers.

A 95% confidence interval gives the range within which it would be expected that the true indicator value would fall 95 times if 100 samples were selected.

SACE is designed so that the 95% confidence interval around an estimate of 50% is less than 5 percentage points above or below this estimate. For example, this means that if the survey gives an answer of 50%, we can be confident that the true figure is between 45% and 55%. However, it should be noted that:

• for local authorities with very small numbers of service users, it’s particularly difficult to achieve the margin of error requirement
• margins of error are much smaller at national level than at local authority level as they are based on more respondents

When comparing 2 estimates, only where confidence intervals do not overlap are the estimates considered statistically different.

In a confidence interval, the range of values above and below the sample statistic is called the margin of error. In the example given above, the margin of error is 5 percentage points.

Table 3: summary of margins of error for local authorities, England 2025 to 2026

Margin of error	Number of local authorities
Less than 5%	65
5% to 5.5%	30
5.5% to 6%	15
More than 6%	42

The data in table 3 can be found in table 5 of the accompanying ‘Survey of adult carers in England 2025 to 2026: data quality tables’ on the Survey of adult carers in England: 2025 to 2026 report page.

Table 3 shows a summary of the margin of errors achieved by each of the local authorities taking part in SACE 2025 to 2026. The margins of error were taken at the 95% confidence level around an estimate of 50%. It shows that 65 local authorities had a margin of error less than 5% and 42 local authorities had a margin of error greater than 6%. Due to this data being grouped using unrounded figures, the figures represented in table 3 may not match exactly those derived from aggregating the relevant column in the data quality tables.

The margin of error for individual questions varies greatly at local authority level and can be considerably higher. See table 4 of the ‘Survey of adult carers in England 2025 to 2026: data quality tables’ on the Survey of adult carers in England: 2025 to 2026 report page for the margin of error for each question at local authority level. Users are advised to refer to this data before analysing results at local authority level.

Non-response and sampling bias

Non-response and sampling bias can occur if response rates are low and if subgroups of the population are more likely to respond than others. The response rates for each question for each local authority are provided in the data quality tables.

The response rates for each question for each local authority are provided in table 3 of the accompanying ‘Survey of adult carers in England 2025 to 2026: data quality tables’ on the Survey of adult carers in England: 2025 to 2026 report page.

Help completing the survey

Respondents were allowed to have assistance when completing the questionnaire and around 7.9% of respondents did so, this was a decrease proportionally from 8.1% in 2023 to 2024. Although not ideal, allowing this as part of the survey design is essential to help to make the survey representative of as many carers as possible.

Survey design sources of bias: method of collection

Of those who responded and where the method of collection is known, 99.4% returned questionnaires were completed by post and 0.6% were conducted by telephone interview.

Table 4: summary of overall responses by method of collection

Method of collection	Number of responses	Percentage of responses (%)
Post	43,215	99.4
Face-to-face	5	0.0
Telephone	240	0.6

The data in table 4 can be calculated using the CSV data in the accompanying ‘Survey of adult carers in England 2025 to 2026: data pack’ on the Survey of adult carers in England: 2025 to 2026 report page.

Table 4 shows the total number of responses for each method of collection. Figures are rounded so may not sum up to the total number of respondents. As 99.4% of surveys were returned by post, there is minimal bias caused by the different methods of data collection.

Survey design sources of bias: adding or modifying questions

Some local authorities who requested to add or modify questions to gain specific information from carers using the additional question request form.

The survey guidance makes it clear that if local authorities wish to add questions to the questionnaire, then they must seek approval from the Department of Health and Social Care (DHSC), and local research governance processes should be followed. Further modifications must not be made to any section of the survey materials that are not highlighted as requiring input from the local authority unless consent has been given by DHSC.

This aims to limit variation between local authorities conducting the survey and to help guard against order effects - for example, how the inclusion of additional questions may impact on responses to subsequent questions.

The data from the additional questions was not returned to DHSC and did not contribute to this publication.

Survey design sources of bias: reviewing materials

To ensure that local authorities were consistently following the survey process, DHSC reviewed the local authority materials to check any changes from the template had been pre-approved and that only approved additional questions were added.

Timescales of fieldwork

Local authorities were required to extract their eligible population during July to September 2025. The data quality tables provide details of the extract dates used.

Local authorities were required to distribute the questionnaires to a random sample of carers who are eligible for the survey in October and November 2025. A fixed period is recommended to minimise the impact of any wider contextual issues such as national news stories, that may influence the views expressed by respondents.

Completeness of carers data

The data quality tables contain the percentage of missing administrative data and missing question data for all local authorities.

Accuracy of eligible population

The eligible population for SACE survey includes all carers who are aged 18 and over and who are caring for an adult aged 18 and over. Carers must have either:

• received carer support at any point in the last 12 months
• been assessed in the last 12 months but received no direct support, irrespective of whether their cared-for person received respite care

The eligible population is the same as previous years and, despite the shift from the short and long term (SALT) return to client level data (CLD) collection, the population should still align with the previous SALT LTS003 table 1 cohort.

The validation reports sent to local authorities highlighted where the difference between the 2023 to 2024 SALT eligible population and SACE eligible population was greater than 20%.

SALT collection ceased in 2023 to 2024, with this collection being the last available. Comparison to 2023 to 2024 SALT is still offered here as broad comparison, until other data sources can fully replicate 100% coverage.

Translated materials

Less than 0.2% of returned questionnaires used translated versions of the survey. This was a similar percentage to 2023 to 2024.

Validations at source

When the questionnaires were returned to local authorities, they were entered onto a data return and, as local authorities completed their data returns, a range of validation checks and flags appeared. These validations were carried out using conditional formatting on the record-level data and a system of flags against the summary data. The record-level validations included:

• the highlighting of invalid and contradictory entries
• missing responses for mandatory questions

The data was summarised into tables, where validations on proportions and values were performed, and flags were provided to local authorities to assist them in assessing the quality of their data before submitting it.

Password protection was applied to the data return to protect the structure of the form and limit what changes could be made.

Once a successful data return submission was made, an automated validation process performed several checks on the data submitted and sent a report to the data submitter. The checks included:

• are the values entered valid for the admin data item or question?
• have question responses been provided for a non-response record?
• are multiple choice selections valid?
• has an extract date been provided in the ‘survey process information’ worksheet?

Local authorities could make numerous submissions and receive a validation report for every submission made. The most recent submission received when the window closed was the one processed for the publication.

Post-submission validations

All local authorities that submitted a valid data return by the mandated deadline received an individual data quality summary report. The data quality summary report provided summary data for the individual local authority:

• alongside their previous year’s data
• against information collated on data submitted by all local authorities

This report flagged where a local authority’s 2025 to 2026 data was more than 2 standard deviations from the overall local authority average.

The collection window was reopened for local authorities to submit revised data. Local authorities were then given until 2 April 2026 to make any changes and resubmit their data return.

In addition to the validation and data quality summary report, manual checks were also carried out and queries were sent to local authorities as required. As part of this validation process, DHSC also provided support to local authorities who had queries with their reports or required further guidance or assistance with their resubmission.

Additional checks were carried out during the processing of the final data, and a report was produced which flagged data quality issues. The report included invalid responses. The invalid codes were excluded from the data tables but the other responses for the record were still included.

Local authorities not submitting a valid data return by the mandated deadline

The mandated deadline for submitting data returns was 26 February 2026.

The City of London, Portsmouth and Cumberland did not submit a data return by the mandated deadline. DHSC worked with the local authorities to ensure they were able to submit for the final deadline.

Local authorities that do not submit by the mandated deadline do not receive a data quality summary report. As a result, their data is not subject to the same validation process as data submitted by the mandated deadline. Users of the data may wish to consider this when making comparisons between local authorities.

Other known data quality issues

Weightings

Weightings were used to calculate national, regional and local authority type estimates which made the calculation of confidence intervals for these aggregated results more complicated. Details of how these weights were calculated are found in the accompanying ‘Survey of adult carers in England 2025 to 2026: methodology’ on the Survey of adult carers in England: 2025 to 2026 report page.

Geography

The local authority-level data tables contain disaggregation by local authority type and region, in alignment with the Ministry of Housing, Communities and Local Government (MHCLG) definitions.

The local authority, region names and codes are also in alignment with those set out in the Office for National Statistics (ONS) guidance for administrative geographies - for full guidance, visit the open geography portal. However, the regions are also slightly different from those defined by the Association of Directors of Adult Social Services (ADASS) - Milton Keynes sits in the South East whereas, for ADASS, it’s in the East Midlands.

It should be noted that the classification of local authority type also differs - the MHCLG groupings used in this publication classify Greenwich as Inner London, and Haringey and Newham as Outer London, whereas the ONS administrative geographies classify Greenwich as Outer London, and Haringey and Newham as Inner London.

Timeliness and punctuality

The data in this publication relates to the financial year 2025 to 2026.

The survey fieldwork was carried out during the period October to November 2025. The final survey data was submitted to DHSC by 26 February 2026, and the final data was submitted by 2 April 2026.

This final SACE 2025 to 2026 statistical release was published on 24 September 2026.

Accessibility and clarity

There are no restrictions to access the published data.

All data is published at individual level as part of this release in a CSV format, alongside guidance for using the data - see the accompanying ‘Survey of adult carers in England 2025 to 2026: data pack’ zip file on the Survey of adult carers in England: 2025 to 2026 report page.

Some sensitive and personal variables for unique rows were removed from the data set for data protection and disclosure reasons. More information is given in the guidance for using the data.

This publication has been published on GOV.UK and consists of question responses and demographic data tables, data quality tables and time series data in ODS format, alongside HTML information on the various aspects of the statistical release.

Coherence and comparability

The data was derived from consistent data sources and a collection method that was consistent with the 5 previous survey years (2016 to 2017 through to 2025 to 2026) over which SACE was carried out - therefore, there was a high degree of coherence.

Compared to the 2023 to 2024 survey, the ‘Reported health condition: autism spectrum disorder (ASD) (voluntary field)’ question has been made mandatory with the previous 2 separate fields on reported health condition for autism and Aspergers removed.

The definition of the eligible population for the 2025 to 2026 survey was the same as that used from 2016 to 2017. All surveys after and including 2016 to 2017 are therefore comparable, subject to data quality issues listed within this report and the accompanying data quality tables.

Before 2016 to 2017, a different definition was used for the eligible population, so caution should be used when comparing the results of earlier surveys. The 2016 to 2017 data quality report contains full details of the change in definition and analysis of its impact.

In 2014 to 2015, there were some important changes to the methodology. These are explained in the methodological change notice of the 2014 to 2015 report, and they should be kept in mind when making comparisons to the 2012 to 2013 data. In the time series document that accompanies this report, the 2012 to 2013 results have been recalculated using the new weighting methodology so comparisons can be made.

We do not recommend making comparisons to the 2009 to 2010 pilot survey - see the accompanying methodology document for further information.

Performance, cost and respondent burden

SACE collection was mandated for all local authorities with the exception of the Isles of Scilly and City of London for the reasons listed in the ‘Submissions and exempt local authorities’ above.

The data collection process used in this publication is subject to the Data Standards Assurance Service (DSAS) on behalf of the Data Co-ordination Board. This was previously carried out by the Challenging Burden Service. The process ensures that data collections:

• do not duplicate other collections
• minimise the cost to all parties
• have a specific use for the data collected

Visit NHS England’s governance of data for more information on the DSAS.

The burden of SACE has been assessed and approved. The burden of any changes to the collection are similarly assessed, to ensure that they do not create undue burden for local authorities.

Confidentiality, transparency and security

The data contained in this publication is collected and prepared in line with the Code of Practice for Official Statistics.

The content of SACE outputs was reviewed as part of the joint Health and social care statistical outputs consultation, which closed on 5 March 2024. Following review of the adult social care consultation feedback, the joint DHSC and NHS England action plan agreed that SACE will continue with a reduced commentary.

Relevant NHS England policies and guidance are as follows:

• Statistical Governance Policy
• Data sharing standard 7b - Duty of Confidentiality
• Freedom of Information process
• ISB1523: Anonymisation Standard for Publishing Health and Social Care Data
• Data Access Request Service
• Data protection policy

Users and uses of the statistics

This section contains a summary of users that have found the information in the SACE publication useful for the purposes set out.

Adult social care outcomes framework (ASCOF)

SACE data is used to populate several outcome measures from the ASCOF statistical collection.

DHSC

SACE data is used to:

• inform policy monitoring
• inform speeches and briefings for ministers and senior officials
• answer Parliamentary questions and Prime Minister’s questions
• answer media enquiries and inform other correspondence

Local authorities

Local authorities will use the survey in different ways but there will be some commonality between them. Ways in which local authorities use the survey will include:

• benchmarking against other local authorities
• measuring and monitoring local performance
• policy development
• service development, planning and improvement
• management information, local reporting and accountability
• informing business cases
• identifying any immediate priorities or areas of concern

Data from this survey is used to understand how local progress is being made towards the adult social care priorities for local authorities.

Care Quality Commission (CQC)

SACE data is used in CQC State of Care reports. The reports look at trends in health and social care, shares examples of good and outstanding care and highlight where care needs to improve. SACE data has been used for the following:

• highlight an understanding of care satisfaction among carers
• support the claim that there is difficulty getting information about services for particular cohorts of users
• support the claim that satisfaction varies regionally

Academics and other known users

The data is used by the Personal Social Services Research Unit at the University of Kent to explore and understand variations in quality and outcomes in social care services. The results of these analyses are used to feed into social care policy and practice - in particular, ASCOF.

Unknown users

The survey data is free to access online and therefore many users will access this publication without being known. It is important to understand how these users are using the statistics and to gain feedback on how to make the data more useful.

Contact

We’d like feedback from our users about how you use our products, how well these products meet your needs and how they could be improved.

For feedback and any further questions, contact asc.statistics@dhsc.gov.uk.

Help us improve SACE

We are interested in understanding more about how the outputs and data from this report are used, and the decisions they inform. This is important for us so we can provide a high-quality publication that meets your needs.

Complete the survey to share your feedback.