Adult social care workforce survey: background quality and methodology
Published 17 December 2021
Applies to England
© Crown copyright 2021
This publication is licensed under the terms of the Open Government Licence v3.0 except where otherwise stated. To view this licence, visit nationalarchives.gov.uk/doc/open-government-licence/version/3 or write to the Information Policy Team, The National Archives, Kew, London TW9 4DU, or email: [email protected].
Where we have identified any third party copyright information you will need to obtain permission from the copyright holders concerned.
This publication is available at https://www.gov.uk/government/statistics/adult-social-care-workforce-survey-december-2021/adult-social-care-workforce-survey-background-quality-and-methodology
About these statistics
These statistics are badged as experimental statistics as they are newly developed and have considerable immediate value to users.
Data is being published to provide insight into current workforce pressures in the social care sector compared to 6 months previously. This includes issues relating to recruitment, retention, staff morale and use of agency staff.
This report is being published as part of the Department of Health and Social Care (DHSC) in adult social care settings, which includes data on care home staffing levels. Data for both this workforce survey and the monthly statistics are taken from the Capacity Tracker.
For more details on the Capacity Tracker data collection and the methodology for the monthly statistics please see adult social care in England statistics: background quality and methodology
Introduction
The workforce survey was a voluntary survey completed by CQC-registered care homes and domiciliary care providers via the Capacity Tracker. It ran for a month from 13 September 2021 to 14 October 2021 and aimed to gain insight into the scale of workforce challenges and specific areas of concern.
This document primarily sets out information on the data sources and methodology used to generate the statistics.
These statistics cover:
- retaining staff: scale of challenges in retaining staff; reported causes of challenges in retaining staff
- recruiting staff: scale of challenges in recruiting staff; reported changes in number, quality, and suitability of applications; harder to recruit roles
- maintaining staff morale: scale of challenge in maintaining staff morale; reported consequences of a more challenging morale situation
- accessing agency staff: scale of challenge in accessing agency staff; frequency of agency staff usage; reported challenges of using agency staff
- assistance from government: forms of government assistance with recruitment or staffing that care settings report they would find useful
Data sources and collection
The data in the publication is taken from voluntary survey returns submitted by CQC-registered care homes and domiciliary care settings in England to a data collection and insight tool called Capacity Tracker.
Capacity Tracker was originally developed by NHS England and the Better Care Fund to enable the system to better manage hospital discharges by identifying available capacity in care homes. It enables care homes to share their vacancies in real-time, meaning hospital discharge teams and other health professionals can rapidly search availability throughout England.
Capacity Tracker was identified as a suitable tool for COVID-19 data collection from care providers. Its re-purposing was announced via the government’s action plan (15 April 2020) and joint letter (17 April 2020) from DHSC and NHS England and Improvement (NHSEI) together with the Care Quality Commission (CQC) and the Care Provider Alliance.
In total, 8,941 care settings responded to the survey, of which:
- 4,051 were care homes
- 4,440 were domiciliary care providers
This represents a response rate of 27% of all CQC-registered care homes and 44% of all CQC-registered domiciliary care providers. When calculating percentages, those that responded N/A or left the question blank are not included as this indicated that the question was not relevant to the respondent. Responses of ‘unsure’ or ‘don’t know’ are included as they indicate a relevant question for which the respondent is not sure of the answer.
As an example, for a question on staff recruitment, a respondent who has not attempted recruitment recently may put N/A or blank, as they cannot speak to the challenges since they haven’t experienced the situation, and therefore their response should not be included in a calculation of percentage. A location which is recruiting but isn’t sure on the level of challenges for whatever reason should put ‘don’t know’. This should be included as, whilst they aren’t sure of the situation, insight can still be gained by seeing that one of the other answers wasn’t picked.
When respondents could give multiple responses, percentages are calculated in 2 different ways. ‘Percentage of respondents’ gives what percentage of respondents include the response as one of their answers. Because respondents could select multiple responses this will not sum to 100%. ‘Percentage of responses’ gives that answer’s share of all the responses provided. This column will sum to 100%. Where only one option can be selected, these 2 percentages will be the same.
For an example of this, take 2 respondents, A and B, responding to a question with 4 possible answers. They are told to select all options which apply:
- respondent A selects options 1, 2 and 3
- respondent B selects only option 1
- for option 1, ‘percentage of respondents’ is 100%, but ‘percentage of responses’ is 50%
- for options 2 and 3, ‘percentage of respondents’ is 50%, but ‘percentage of responses’ is 25%
- for option 4, both are 0%
For free text analysis, multiple options were considered, however only a few tools were appropriate, namely term frequency and bigrams. Generally, more sophisticated tools such as sentiment analysis and topic modelling were either inappropriate (in the case of sentiment analysis) or required much larger datasets (topic modelling), however insight can still be gained from the basic tools that were used.
When providing wordcounts, responses were first filtered for unique responses only. These responses were then run through a tool to count word frequency. This list then had ‘stopwords’, such as ‘to’, ‘is’ and ‘the’ removed. Finally, words that appeared in the question, and therefore would appear in every answer were manually curated. This included words such as ‘staff’ and ‘care’ which were essential to providing answers and therefore did not provide any insight.
Data coverage
All data in this publication refers to CQC-registered care homes and domiciliary care providers in England only.
Data was self-reported by care settings on a voluntary basis between 13 September 2021 and 14 October 2021.
In total, 8,941 settings responded to the survey, of which 4,051 were care homes and 4,440 were domiciliary care providers. This represents a response rate of 27% of all CQC-registered care homes and 44% of all CQC-registered domiciliary care providers.
Size of respondents
Below is how the sample used compares to the population, when using the number of staff to represent size of a respondent. A maximum difference of 2.2 percentage points was found when splitting the sizes into the bins, as shown below.
| Total number of staff | Proportion of settings responding to the survey | Proportion of all CQC-registered settings | Difference (percentage points) |
|---|---|---|---|
| 0 to 10 | 14.3% | 14.7% | 0.5 |
| 11 to 25 | 33.4% | 31.6% | 1.9 |
| 26 to 50 | 29.7% | 29.6% | 0.2 |
| 51 to 100 | 16.4% | 18.6% | 2.2 |
| Over 100 | 6.2% | 5.5% | 0.7 |
Also considered was the resident to staff ratio. Ratios were calculated as ‘number of residents’ divided by ‘number of staff’ and then sorted into bins with a width of 0.2. The results of this analysis are shown below.
| Resident to staff ratio (R) | Proportion of settings responding to the survey | Proportion of all CQC-registered settings | Difference (percentage points) |
|---|---|---|---|
| R ≤ 0.2 | 4.8% | 4.9% | 0.2 |
| 0.2 < R ≤ 0.4 | 20.2% | 18.3% | 1.9 |
| 0.4 < R ≤ 0.6 | 23.6% | 21.6% | 2.0 |
| 0.6 < R ≤ 0.8 | 25.5% | 27.0% | 1.5 |
| 0.8 < R ≤ 1.0 | 17.0% | 18.8% | 1.8 |
| R > 1.0 | 8.9% | 9.4% | 0.6 |
Respondent Location
Below are the comparisons between the sample and population for regions. These are split by care type. Please note that the shortage in domiciliary care providers in London is due to a known issue with returning data.
Care homes
| NHS region | Proportion of settings responding to the survey | Proportion of all CQC-registered settings | Difference (percentage points) |
|---|---|---|---|
| East of England | 9.0% | 11.4% | 2.4 |
| London | 10.8% | 8.8% | 2.0 |
| Midlands | 19.2% | 20.5% | 1.3 |
| North East and Yorkshire | 17.0% | 15.5% | 1.5 |
| North West | 10.6% | 12.0% | 1.4 |
| South East | 18.4% | 18.7% | 0.3 |
| South West | 15.0% | 13.1% | 1.9 |
Domiciliary care
| NHS region | Proportion of settings responding to the survey | Proportion of all CQC-registered settings | Difference (percentage points) |
|---|---|---|---|
| East of England | 13.0% | 12.1% | 0.8 |
| London | 7.1% | 15.4% | 8.3 |
| Midlands | 21.6% | 21.1% | 0.5 |
| North East and Yorkshire | 15.5% | 12.6% | 2.8 |
| North West | 12.4% | 11.5% | 0.9 |
| South East | 19.1% | 17.3% | 1.8 |
| South West | 11.4% | 10.0% | 1.4 |
Data quality
This section measures the adult social care statistics against the dimensions of quality set out by the Government Statistical Service for statistical outputs.
Any feedback on these statistics is welcome and can be sent to [email protected].
Relevance
This is a one-off voluntary survey conducted and published by DHSC of statistics on workforce challenges amongst adult social care settings in England. This statistical bulletin provides an overview of reported issues relating to recruitment, retention, maintaining staff morale, and the use of agency staff.
These statistics are published to provide insight into current workforce pressures in the social care sector compared to 6 months previously, and to provide transparency.
These statistics complement DHSC’s monthly adult social care statistics, which provide an overview on a range of information on social care settings, with a focus on the impact of COVID-19.
These statistics are being badged as experimental statistics whilst DHSC scopes out user and stakeholder needs for these statistics and assess data quality.
Accuracy and reliability
Accuracy
Coverage is limited to care settings registered with Capacity Tracker and responding to the survey.
Data is self-reported by care settings and as this was a voluntary survey focussing specifically on workforce challenges, may be subject to self-selection bias.
There are some situations in which large providers upload data for all care homes and domiciliary care providers that they operate. This may involve uploading the same response for different settings in different regions which may be unlikely in practice, for example, a care home and a domiciliary care provider are unlikely to have the same level of staff shortages in the same positions.
Data from large providers is still included in the responses but is excluded from the analysis of free text responses to avoid a single large provider skewing the prominence of specific words by reporting multiple identical answers. This applies particularly to 2 large providers with over 200 responses each. The third largest provider reported for 42 locations.
Reliability
The data tables and analysis for this publication were produced in Microsoft Excel and subject to rigorous quality assurance by a professional analyst.
Timeliness and punctuality
These statistics are compiled from data gathered between 13 September 2021 to 14 October 2021 and ask care settings to compare their current situation to April 2021. These statistics are being published in December 2021 to ensure they are released whilst still relevant. The lag since the survey ended has allowed for DHSC to analyse, disseminate and quality assure the data.
More timely data on the adult social care workforce in England is available through other sources, more details on this can be found in the ‘Data sources and collection’ section above.
Accessibility and clarity
These statistics are freely available on the gov.uk website with all documents published in an accessible format.
The statistical reports and background quality and methodology documents are published in html and accompanying data tables are published in Open Document Spreadsheet (ODS) format.
This background quality and methodology document is published to ensure users have sufficient information on how these statistics can be used.
The commentary is written with the aim of being clear and impartial. DHSC will continue to scope user and stakeholder needs to ensure the commentary sufficiently meets their needs.
Other sources of adult social care COVID-19 data
These statistics are being published as a part of a wider landscape of statistics on adult social care. The Government Statistical Service (GSS) compiles a UK adult social care database of official statistics on adult social care across the 4 nations of the UK. This is updated on a monthly basis.
DHSC adult social care statistics
This is being published as part of the DHSC monthly statistics on adult social care. This publishes information on COVID-19 in adult social care settings, which provides insight into how COVID-19 has impacted the social care sector and its response to the pandemic.
Data for both this publication and the adult social care statistics are taken from the Capacity Tracker.
The adult social care statistics includes some data on care home staffing levels. In order to measure the workforce pressures on settings during the pandemic, each respondent is asked to assess their workforce pressures based on their agreed staffing ratios. More details on how staffing ratios are agreed can be found within the Care Quality Commission (CQC) guidance on staffing.
The workforce survey complements the regular collection by providing the opportunity to have a deeper exploration of workforce issues which is possible in an ad-hoc survey but would be too burdensome for the regular collection. The workforce survey collected information from both care homes and domiciliary care providers, but the monthly statistics only contain information for care homes.
Skills for Care
Skills for Care publish a range of workforce intelligence regarding adult social care in England.
This includes the annual state of the workforce report which provides an analysis of the adult social care workforce in England and the characteristics of those working in it.
Skills for Care also publish monthly tracking information for key workforce measures relevant to the COVID-19 pandemic, including total staff, vacancies, occupancy rates and sickness.