Research and analysis

Survey findings: Health-led Employment Trial Evaluation

Updated 8 April 2024

Applies to England, Scotland and Wales

DWP research report no.1035

A report of research carried out by the Institute for Employment Studies on behalf of the Department for Work and Pensions.

You may re-use this information (not including logos) free of charge in any format or medium, under the terms of the Open Government Licence.

To view this licence, visit the National Archives

Or write to:

Information Policy Team
The National Archives
Kew
London
TW9 4DU

Email: psi@nationalarchives.gov.uk

This document/publication is also available on our website at: Research at DWP

If you would like to know more about DWP research, email socialresearch@dwp.gov.uk

First published July 2023.

ISBN 978-1-78659-547-8

Views expressed in this report are not necessarily those of the Department for Work and Pensions or any other government department.

Executive summary

This report presents a descriptive analysis of the 12-month survey conducted as part of the Health-led Employment Trials (HLTs) evaluation. The survey was issued on a monthly rolling basis to the full sample of recruits – those randomised (that is, assigned to treatment or control groups) for the trials. The report outlines the profile of recruits responding to the survey and details employment, work search activities and self-efficacy, health conditions, barriers to work, and perceptions of support received as part of the Individual Placement and Support (IPS) services among respondents in different trial groups. Key findings are:

• the trial was designed with an understanding that while respondents were likely to continue experiencing health problems, these should not be so serious that it was unrealistic for them to move into work. The final survey results show that around a third of respondents did feel able to work.
• additionally, around 1 in 5 respondents from the out-of-work (OOW) trial groups were in work at the time of the final survey.
• comparisons between the treatment and control groups highlight that the support went some way to address fears and misconceptions about the potential negative impacts of employment on health and wellbeing.
• Overall, the survey results show a positive perception of the IPS services and the support received among respondents.
  • 68% of participants said that their employment specialist “understands my needs a lot”.
  • 69% of participants reported that their employment specialist “has the right skills and expertise”.
• despite these positive findings, respondents continued to face major health, wellbeing, and life satisfaction barriers to finding work.
• the results suggested that there may be gaps in the support for some respondents with particularly complex needs as they seek to find and sustain suitable employment. This may have emerged as the trial’s IPS services were designed to support people with mild-to-moderate health conditions.
• in line with a more holistic understanding of their complex needs, this may suggest the domains of the IPS services as configured for the trials could be extended to better account for a range of wider determinants of health.
• when comparing trial groups, the Sheffield City Region in-work (SCR IW) group, which reported higher levels of education, greater likelihood to own their own homes, and lengthier employment histories before the trial started, were in the strongest position to secure employment outcomes. Their position also suggested they were the least likely to be receiving any employment support which may have led to their strongly positive views of the support they received.
• there may be scope for further targeting of IPS services seen in the trials towards the needs and circumstances of people with health conditions who have the weakest employment histories and who have not spent long and stable periods of time in employment. However, the IPS services as operated in the trials may need some adjustments to meet more complex health needs.

The final report series for the trials covers:

• synthesis report – a high-level, strategic assessment of the achievements of the trial, drawing together the range of analyses from the evaluation.
• 4-month outcomes report covering: an analysis of implementation, a descriptive analysis of the survey findings 4 months post-randomisation, and an assessment of impact at 4 months following randomisation.
• 12-month survey report providing a descriptive analysis of the final survey, based on the theory of change for those in the treatment group.
• Context-mechanism-outcome (CMO) report, reporting evidence on outcomes from the trials and relating these to its theories of change.
• 12-month impact report covering the net effect on employment, health and wellbeing resulting from the trials 12 months after randomisation drawing on administrative and survey data.
• economic evaluation report exploring the costs and benefits arising from trial delivery, drawing on the administrative and survey data.
• the pandemic and the trial – an analysis of how the trial outcomes may have been affected by the onset of COVID-19.

Authors’ credits

Joe Crowley is a Senior Researcher in the Health and Social Care team at NatCen Social Research, whose main methodological expertise lies in survey data collection and analysis. He has worked across a variety of large-scale cross-sectional surveys, including NatCen’s flagship British Social Attitudes survey and the Family Resources Survey. He has also worked on a range of mixed-methods projects, combining qualitative and quantitative methods.

Imogen Martin is a Researcher in the Health and Social Care team at NatCen Social Research. She has worked on areas across the policy spectrum including healthcare, disabilities and welfare. As part of the Health-led Employment Trials, Imogen has assisted with the delivery and analysis of the interim and final surveys.

Phoebe Weston-Stanley is a Research Assistant in the Health and Social Care team at NatCen Social Research. She has worked on projects using both qualitative and quantitative methodologies across the policy spectrum. As part of the Health-led Employment Trials, Phoebe has assisted with the analysis of the final surveys.

Dr Rosa Lau is a Research Director in the Health and Social Care team at NatCen Social Research. She has extensive experience of delivering complex applied research and service evaluations using mixed methods on areas including healthcare, public health, employment, housing and welfare.

Dr Priya Khambhaita is Co-Director of Health and Social Care at NatCen Social Research. She has 12 years of experience in conducting policy and academic research on health, social care, work, and income.

Becci Newton is Director of Public Policy and Research at the Institute for Employment Studies (IES) and specialises in research on unemployment, inactivity, health, skills and labour market transitions. Becci has managed the evaluation since its design and contributed to the process evaluation. She has led multiple evaluations for DWP including of the 2015 ESA Reform Trials and the Work Programme.

Rosie Gloster is a Principal Research Fellow at IES. She supported the management of the evaluation consortium and contributed to the process evaluation. She is a mixed-methods researcher specialising in employment, and careers. She has authored several reports for the Department for Work and Pensions (DWP), including the Evaluation of Fit for Work.

Rebecca Duffy and Zoe Gallagher are Project Support Officers at IES who led proofing and formatting of this report.

Glossary of terms

Term	Definition
Base	The number of observations or cases in a sample. For example, a survey may have a base=2,300 respondents. During analysis the base may become smaller, for example if not all respondents answer a particular question, or when analysing responses from a subset of the full sample.
Baseline data collection	Data from the baseline assessment completed by provider staff who recruited people to the trial.
Causal link	The connection between a cause and an effect.
Controlling for	In statistical modelling with multiple variables and factors, keeping one variable constant in order to examine and test the relationship and effect between other variables of interest in the model.
Dataset	A collection of data or information such as all the responses to a survey or all the recordings from a set of research interviews.
Demographic	A particular section of the population. Also refers to characteristics of an individual of interest for research, such as age, gender, and ethnicity.
Descriptive analysis	Producing statistics that summarise and describe features of a dataset such as the mean, range and distribution of values for variables.
EuroQol-5D-5L (EQ5D5L)	Descriptive system for health-related quality of life states in adults, consisting of five dimensions (mobility, self-care, usual activities, pain & discomfort, anxiety & depression), each of which has five severity levels described by statements appropriate to that dimension
Employment specialists	Staff employed by the trials to undertake randomisation appointments, provide IPS support to the treatment group, and undertake employer engagement.
Final survey	The survey completed by recruits 12 months after randomisation.
Health-led Employment Trials	Two trials, funded by the Work and Health Unit, to test a new model of employment support for people with long term health conditions.
4-month survey	The survey completed by trial recruits four months after starting the trial.
Intervention	The work and health support provided in Sheffield City Region and the West Midlands Combined Authority as part of the trial.
In employment/working	Those in employment full-time, part-time, or less than 16 hours a week; those who are self-employed.
In paid work	Those in those in employment full-time, part-time, or less than 16 hours a week, not those who are self-employed
Individual Placement and Support (IPS)	IPS is a voluntary employment programme that is well evidenced for supporting people with severe and enduring mental health needs in secondary care settings to find paid employment.
Job Search Self Efficacy Scale	Nine item scale to measure self-efficacy relating to finding employment.
Provider staff	Those working in provider organisations including employment specialists delivering IPS support, as well as managers and administrators
p-value	Used as a measure of statistical significance. Low p-values indicate results are very unlikely to have occurred by random chance. p<0.05 is a commonly cited value, indicating a less than 5 per cent chance that results obtained were by chance. Research findings can be accepted with greater confidence when even lower p-values are cited, for example p<0.01 or p<0.001.
Recruits	People who agreed to take part in the trials and who were randomised to either the treatment or control group
Refer / referral	A recommendation that an individual should be considered for the trial, facilitated by a means to directly connect them to a trial provider
Respondents	Trial recruits from the treatment or control group who were invited to take part in the evaluation and took part in the surveys. As such the descriptive analysis of the survey identifies treatment group respondents and control group respondents
Site	The trials were delivered in two combined authorities, which are termed sites.
Statistical significance	Statistical significance indicates that the result or difference obtained following analysis is unlikely to be obtained by chance (to a specified degree of confidence) and that the finding can be accepted as valid. A study’s defined significance level is the probability of the study rejecting the null hypothesis (that there is no relationship between two variables), demonstrated by the p-value of the result.
Short Warwick-Edinburgh Mental Well-being Scale	The SWEMWBS is a short version of the Warwick–Edinburgh Mental Wellbeing Scale (WEMWBS). The WEMWBS was developed to enable the monitoring of mental wellbeing in the general population and the evaluation of projects, programmes and policies which aim to improve mental wellbeing
Survey	A research instrument used to collect data by asking scripted questions or using lists or other items to prompt responses. Can be conducted in person face-to-face, by telephone, or by postal or web-based questionnaire.
Tenure	Housing arrangement or status of an individual, for example owner occupier, private renter, or local authority or housing association renter.
Theory of Change (ToC)	A description and illustration of how and why a desired change is expected to happen in a particular context. It sets out the planned major and intermediate outcomes and how these relate to one another causally.
Trial group(s)	Three trial groups are referred to in the report: 2 out-of-work (OOW) groups (1 in each combined authority), an in-work (IW) group in Sheffield City Region (SCR).
Variable	A variable is defined as any individual or thing that can be measured.
Weighting	During analysis of survey data, adjusting for over- or under-representation of particular groups, to ensure that the results are representative of the wider population.

1. Introduction

This report explores the situation of respondents at the time of the final survey across a range of themes, in order to understand the full context in which respondents took part in the trial and the trial’s effects, described in the impact report. In the rest of the introduction, we outline the sections included in this report and the analysis conducted to inform it. We also provide a summary of the fieldwork and methods used in conducting the final survey for the Health-led Employment Trials (HLTs).

1.1. Report outline

This report seeks to provide information on how close to being in employment respondents in this trial, both control and treatment groups, were at the time of the final survey. Among those in employment, it explores how secure and stable their employment was; how satisfied they were with their jobs; and how well suited these jobs were to respondents. It also describes important background characteristics of respondents which may be important in understanding how ready or able they were to find work, particularly in terms of health but also across other areas, such as confidence in their ability to apply for jobs. Finally, it reports on the perceptions of respondents to the treatment group about the trial support (the IPS service), in terms of how effective it was in helping them get ready for work and manage their health conditions and impairments.

1.1.1. Report structure

The report is structured into the following sections. It first describes the demographic characteristics of final survey respondents (based on data from the baseline survey) (Section 2). It then explores their employment status at the final survey, work history during the 12 months of the trial, and, among people in work, attitudes to the quality of their jobs (Section 3). Respondents’ work search activities, as well their perception of their own ability to achieve their aims (self-efficacy) are also reported, both in general and in work-related contexts (Section 4). Wellbeing and health at the time of the final survey are then described, including both physical and mental health problems (Section 5). Barriers to employment experienced by respondents are then reported (Section 6), including the extent to which health problems or impairments limit their ability to work, alongside a range of work-related and other practical barriers (such as a lack of transportation). Finally, among respondents in the treatment group who received the trial support, Section 7 outlines respondent perceptions of that support.

1.1.2. Theory of change

There are three theories of change informing the trials:

• an intervention level theory of change, which shows how change for the treatment group is expected to be achieved;
• a health systems theory of change, which shows how change in health systems is expected to be achieved;
• an employer theory of change which shows how change in employer behaviour is expected to be achieved.

The survey mainly refers to the intervention level theory of change. In the discussion section, we explain how some of the results from the survey relate to the intervention level theory of change where appropriate - the CMO report for the trials’ evaluation covers the qualitative evidence on the three ToCs in detail. In this report, the particular focus is on addressing the following causal pathways, illustrating the steps through which the outcomes and activities work to achieve the desired impact:

Participant causal pathway 1:

The trial will increase or sustain participant income, through the treatment group entering and sustaining employment. This will be supported through the IPS worker achieving the best possible job match for the individual based on their skills and interests, which in turn will result in satisfaction in work and thus support work sustainment.

Participant causal pathway 2:

Through being in work, and as a result of the IPS service, the treatment group will improve their health (physical and/or mental) and make more effective use of health services, thereby reducing the costs on health services in the long run.

1.1.3. Analysis groups

In this report, respondents are split into the three main trial groups, as outlined in Table 1.1 below. These are based on the 2 trial sites, dividing respondents into those from the Sheffield City Region (SCR)^{[footnote 1]} and the West Midlands Combined Authority (WMCA). The SCR respondents are then further grouped depending on the criteria under which they joined the trial; those who joined the trial in-work (SCR IW) were analysed separately to those who joined the trial out-of-work (SCR OOW).

Table 1.1: Health-led employment trials: participant groups

Trial group	Health condition	Work status
SCR in-work (SCR IW)	Mild to moderate physical or mental health condition or impairment (excluding those on a Care Programme, receiving Community Mental Health Team support or identified on a Serious Mental Illness register)	In any kind of employment for 16 or more hours of work a week, including those who were self-employed but struggling or off sick due to their condition(s)
SCR out-of-work (SCR OOW)	Mild to moderate physical or mental health condition or impairment (excluding those on a Care Programme, receiving Community Mental Health Team support or identified on a Serious Mental Illness register)	Not in employment with an interest in moving into employment
WMCA (WMCA OOW)	Any health condition or impairment (excluding those with a moderate to severe learning disability or dementia)	Not in employment for over four weeks prior to the trial

These groups are referred to using the abbreviations above in the rest of this report. Comparisons are drawn between these trial groups where differences are statistically significant at the 95% level. In other words, the differences reported have no more than a 5% probability of occurring by chance. Using these groups, the report explores disparities both between the SCR and WMCA trial sites and the SCR groups, comparing those who joined the trial in-work (SCR IW) and out-of-work (SCR OOW).

In addition, this trial used a randomised controlled trial methodology, so in each of these groups, participants were randomly assigned into treatment and control groups, with 50% of eligible participants in each group. The final survey findings have therefore also been presented by treatment and control group (where survey questions were posed to both groups), within each of the three trial groups.

For some of the results included in this report the differences between the treatment and control group findings are not presented^{[footnote 2]}. This is because these are discussed in the 12-month impact report, although these results are presented in the appendices. Where our findings cover variables not included within the impact report, differences between treatment and control group respondents are reported where these are statistically significant at the 95% level.

1.1.4. The final survey

The final survey was conducted 12 months after recruits were randomised into the trial. Fieldwork was conducted in 17 monthly batches between 12 June 2019 and 25 November 2020. The final survey was issued to all respondents who had not opted out from primary data collection (8,945) and achieved a total of 4,087 interviews, a response rate of 46%. It was conducted primarily as a web (n=1,768) or telephone (n=2,138) survey; respondents who requested to do so were able to take part in a face-to-face interview (n=181)^{[footnote 3]}. The 4-month survey, capturing interim outcomes for respondents, is reported separately in the implementation and 4-months outcome report.

1.2. Notes to text and tables

1. The data tables accompanying this report are provided in the appendices A and B. The relevant tables are referenced throughout this report. The questionnaire is provided in the accompanying technical report.

2. The data used in the report have been weighted to account for non-response bias. The weights ensure the final survey respondents’ profile closely matches that of the trial population, controlling for factors such as health, demographics, and geographic factors like index of multiple deprivation rank^{[footnote 4]}. All percentages are based on the weighted data. Only the unweighted sample sizes are shown at the foot of each table. Details of the weighting are provided in the separate technical report.

3. Unless otherwise stated, where comparisons are made in the text between different population groups or variables, only those differences found to be statistically significant at the 95% level are reported. In other words, differences as large as those reported have no more than a 5% probability of occurring by chance. The term ‘significant’ refers to statistical significance (at the 95% level) and is not intended to imply substantive importance. P-values that are below or equal to 0.05 are significant.

4. The following conventions have been used in tables:

• 0 indicates no observations (zero value)
• *indicates non-zero values of less than 0.5%

5. Owing to rounding, column percentages may not add exactly to 100%. For some questions respondents could select more than one response; in these cases, the percentages will add up to more than 100%.

6. ‘Missing values’ occur for several reasons, including refusal or inability to answer a particular question or section and cases where the question is not applicable to the respondent.

7. Where a table contains more than one variable, the bases may not be the same. This is indicated in the table’s base.

2. Demographics

This section summarises the demographic characteristics of the final survey respondents, based on their responses at the baseline survey.

Gender:

• in SCR IW, 56% of respondents were women, compared with 44% and 46% respectively in the SCR OOW and WMCA groups (Appendix A: Table 1).

Age:

• all age groups were represented in the trial, with SCR IW respondents more likely to be older than those in the other two trial groups (Appendix A: Table 1).

Figure 2.1: Age at final survey, by trial group

Age	All %	WMCA %	SCR OOW %	SCR IW %
17 to 24	16	17	18	13
25 to 34	21	20	23	21
35 to 44	19	19	19	20
45 to 54	24	24	23	27
55+	19	20	17	19

Base: all final survey respondents (Unweighted base size: All: 4,087, SCR IW: 1,181, SCR OOW: 1,488, WMCA: 1,418).

Ethnicity:

• the majority of respondents in the SCR IW and SCR OOW groups were from a white ethnic background (89% and 86%), compared with 64% of respondents in the WMCA group. Compared with both SCR groups, the WMCA group was more ethnically diverse (Appendix A: Table 1).

Figure 2.2: Ethnicity at final survey, by trial group

Ethnicity	WMCA %	SCR OOW %	SCR IW %
White	64	86	89
Mixed/multiple ethnicgroups	4	2	2
Asian/Asian British	17	6	4
Black/African/Caribbean/Black British	13	4	3
Other ethnic group	2	2	1

Base: all final survey respondents (Unweighted base size: All: 4,071, SCR IW: 1,179, SCR OOW: 1,483, WMCA: 1,409).

Marital status:

• a third (33%) of SCR IW respondents were married or in a civil partnership, compared with 16% of SCR OOW respondents and 18% of WMCA respondents.
• WMCA and SCR OOW respondents were more likely to be single (61% in both groups) when compared with 41% among SCR IW respondents (Appendix A: Table 1).

Dependent children:

• there was a similar distribution of children living with respondents across the trial groups, with between 24% and 26% of respondents living with dependent children (Appendix A: Table 1).

Education (highest qualification held):

• the SCR IW respondent group were more likely to hold higher qualifications than those from the other trial groups. A quarter (27%) of SCR IW respondents held a degree level qualification or higher, compared with 13% in WMCA and SCR OOW groups. WMCA and SCR OOW respondents’ highest qualifications were more likely to be either GCSEs at D-G grade (or equivalent) or no formal qualifications (Appendix A: Table 1).

Figure 2.3: Education at final survey, by trial group

Education	All %	WMCA %	SCR OOW %	SCR IW %
No formal qualifications	12	16	55	17
GCSE grades D-G;NVQ/SVQ level 1	15	17	55	13
GCSEs (A-C) or A-Levels; NVQ/SVQ level2-3	13	19	54	13
Degrees and equivalentsor higher; NVQ/SVQlevel 4-5	6	11	56	27

Base: all final survey respondents (Unweighted base size: All: 4,026, SCR IW: 1,164, SCR OOW: 1,466, WMCA: 1,396).

Tenure:

• respondents in all trial groups were most likely to be renting. However, compared with SCR OOW and WMCA respondents, SCR IW respondents were more likely to own their home outright or have bought their home with a mortgage or loan (Appendix A: Table 1).

Figure 2.4: Tenure at final survey, by trial group

Tenure	WMCA %	SCR OOW %	SCR IW %
Own it outright	10	11	15
Buying it with the help of a mortgage or loan	8	9	30
Part own and part rent(shared ownership)	1	0	1
Rent it	54	55	40
Live there rent-free(including living with parents)	27	24	15

Base: all final survey respondents (Unweighted base size: All: 4,051, SCR IW: 1,172, SCR OOW: 1,474, WMCA: 1,405).

3. Employment

This section describes respondents’ employment situation, work history during the 12 months of the trial, and their satisfaction with their job.

3.1. Employment status at final survey

Figure 3.1 shows the employment status of respondents in each of the trial groups at the final survey. Owing to the composition of the groups at the start of the trial, those in the SCR OOW and WMCA groups were more likely to be unemployed and seeking work than those in the SCR IW group. This was reported by around half of SCR OOW and WMCA respondents (48% and 53% respectively), compared with 12% of the SCR IW group. Another fifth of respondents in the SCR OOW and WMCA groups (22% and 21% respectively) were unemployed and not seeking work, compared with 6% in the SCR IW group.

Correspondingly, the SCR IW group were more likely to be employed full-time (30 hours a week or more) or part-time (16-29 hours a week). Nearly half (47%) of SCR IW respondents were employed full-time, compared with 11% of respondents in the SCR OOW group and 8% of respondents in the WMCA group. In the SCR IW group, nearly a fifth (19%) of respondents were reported to be working part-time (16 to 29 hours a week), compared with 6% of the SCR OOW group and 6% of the WMCA group (Appendix A: Table 2, Figure 3.1).

Figure 3.1: Employment status at final survey, by trial group

Employment status	All %	WMCA %	SCR OOW %	SCR IW %
Employed full-time(30 hours a weekor more)	20	8	11	47
Employed part-time(16 to 29 hours aweek)	9	6	6	19
Employed part-time(less than 16 hoursa week)	6	4	6	6
Unemployedseeking work	40	53	48	12
Unemployed notseeking work	17	21	22	6
Other	18	19	19	16

Base: all final survey respondents. Note: ‘Other’ includes the following question responses: Training, education or apprenticeship (full or part-time); Volunteering or work experience; Self-employed (full or part-time); Carer (adult or child); Other activity (Unweighted base size: All: 4,087, SCR IW: 1,181, SCR OOW: 1,488, WMCA: 1,418).

3.2. Employment history

Respondents who were in paid work at the time of the final survey were asked whether at any point in the last 12 months they had not been in employment, the results of which can be seen in Figure 3.2. Any time off work on sick leave while employed was not included. Those who were in work in the SCR OOW and WMCA groups were more likely to have spent time not in employment in the last 12 months compared with the SCR IW group. Among those in the SCR OOW group and the WMCA group, 73% and 71% of respondents respectively reported being not in employment at some point in the last 12 months compared with 45% of the SCR IW group. However, it is clear that stable employment was a challenge across all groups, as overall, 56% of all respondents had at some point in the last 12 months not been in employment (Appendix A: Table 6, Figure 3.2).

Figure 3.2: Proportion of respondents in work at the time of the final survey, who had been out of work at any point during the last 12 months

Trial group	Percentage
All	56
WMCA	71
SCR OOW	73
SCR IW	45

Base: all final survey respondents in work at the time of the final survey (Unweighted base size: All: 1,553, SCR IW: 920, SCR OOW: 368, WMCA: 265).

Respondents who were not in paid work at the time of the final survey were asked whether they had undertaken any paid work in the past 12 months, the results of which can be seen in Figure 3.3. Those in the SCR IW group were more likely to have completed any paid work (63%), compared with those in the SCR OOW group (18%) and the WMCA group (14%) (Appendix A: Table 8, Figure 3.3). This is likely due to the high proportion of people in the SCR IW group who started the trial from a position of employment, with 73% of people in this group in work at the time of the final survey (Appendix A: Table 2, Figure 3.3)^{[footnote 5]}. Having spent the whole of the trial period without working, the barriers to gaining employment may be long-term and challenging.

Figure 3.3: Proportion of respondents not working at the time of the final survey who had been in work at any point in the last 12 months, by trial group

Trial group	Percentage
All	21
WMCA	14
SCR OOW	18
SCR IW	63

Base: all final survey respondents not in work at the time of the final survey (Unweighted base size: All: 2,475, SCR IW: 261, SCR OOW: 1,094, WMCA: 1,120).

Weeks worked over the last 12 months

Respondents who said they had been in employment at any time during the trial were also asked to indicate the number of weeks that they had worked during the past 12 months (Appendix A: Table 10, Figure 3.4). This shows how stable their employment was during the trial period.

Similar to the findings on employment history, these results show a stronger experience of employment in the SCR IW group. A majority of respondents who had worked at all during the trial had worked 40 to 52 weeks in the past 12 months; 67% compared with 24% in the SCR OOW group and 17% in the WMCA group.

In the SCR OOW and WMCA groups, it was much more common to have worked for shorter periods. Of those in employment at any point during the trial, around 1 in 4 had worked only 1 to 13 weeks (37% and 41% respectively among SCR OOW and WMCA respondents), with a further quarter working 14-26 weeks (23% and 25% respectively among SCR OOW and WMCA respondents). By contrast, in the SCR IW group, only 11% of respondents had worked 1 to 13 weeks, and a further 11% had worked 14 to 26 weeks (Appendix A: Table 10, Figure 3.4).

Figure 3.4: Number of weeks worked over the last 12 months, by trial group

Number of weeks worked	All %	WMCA %	SCR OOW %	SCR IW %
40 to 52 weeks	44	13	17	1
27 to 39 weeks	17	17	25	0
14 to 26 weeks	24	14	23	2
None	67	10	11	1

Base: all final survey respondents in work at any point during the last 12 months (Unweighted base size: All: 1,967, SCR IW: 1,026, SCR OOW: 540, WMCA: 401).

3.3. Job quality

The questions related to job quality came from the Workplace Employment Relations Study (WERS), commissioned by the Department for Business, Energy and Industrial Strategy. Job quality is measured by asking respondents about how much influence they have over the following statements: the tasks they do in their job; the pace at which they work; how they do their work; the order in which they carry out tasks; and the time they start or finish their working day.

Results from the final survey consistently suggested that there were no significant differences between trial groups in terms of how much influence they had over tasks in their job (Appendix A: Table 120), pace of work (Appendix A: Table 122), how work is done (Appendix A: Table 124), and order of tasks (Appendix A: Table 126). Similar patterns were observed across these outcomes. The proportions of respondents reporting ‘a lot’, ‘some’, ‘little’ and ‘no’ influence across these questions ranged between 33 to 40%, 30 to 31%, 16 to 18% and 8 to 14%, respectively.

WMCA group respondents were significantly more likely to feel they had ‘a lot’ (29%) and ‘some’ (22%) influence over number of hours worked, compared with those in the SCR IW (26% and 20%, respectively) and SCR OOW (20% and 19%, respectively) groups. Conversely, a greater proportion of respondents in the SCR OOW group reported that they had no influence over hours worked, 38% compared with 32% of those in the WMCA group (Table 3.1).

There were no significant differences between the treatment and control groups in terms of how much influence they had over hours worked.

Table 3.1: How much influence respondents had over hours worked, by trial group

Level of influence	SCR IW %	SCR OOW %	WMCA %	Total %	p-value
A lot	26	20	29	25	0.023*
Some	20	19	22	20	0.023*
A little	16	18	13	16	–
None	37	38	32	36	–
Don’t know	2	4	4	3	–
Unweighted base	921	368	266	1,555	–

Base: all final survey respondents who were in employment at the final survey (Unweighted base size: All: 1,555, SCR IW: 921, SCR OOW: 368, WMCA: 266). * P-value was tested comparing combined ‘a lot’ and ‘some’ categories with the rest of the sample.

3.4. Job match

This question assessed the extent to which the job matched the skills and interests of the respondents. Overall, just less than a third (31%) of the respondents reported that their job matched ‘a lot’ of their skills and interests, 33% reported their job ‘somewhat’ matched their skills and interests, and 34% reported their job matched their skills and interests either ‘a little’ (20%) or ‘not at all’ (14%) (Table 3.2).

Table 3.2: How much respondent’s job matches their skills and interests, by trial group

Level of match to skills and interests	SCR IW %	SCR OOW %	WMCA %	Total %	p-value
A lot	31	32	33	31	0.897*
Somewhat	33	33	31	33	0.897*
A little	20	19	22	20	–
Not at all	15	13	11	14	–
Don’t know	2	2	4	2	–
Unweighted base	921	368	267	1,556	–

Base: all final survey respondents who were in employment at the final survey (Unweighted base size: All: 1,556, SCR IW: 921, SCR OOW: 368, WMCA: 267) * P-value was tested comparing combined ‘a lot’ and ‘somewhat’ categories with the rest of the sample.

In most cases, the treatment group across all trial groups was more likely to report their job matched their skills and interests ‘a lot’ or ‘somewhat’ (Appendix A: Table 131, Figure 3.5). However, the difference was not significant except within the SCR IW group, where respondents in the control group (18%) were more likely to report their current job matched their skills and interests ‘not at all’, compared with those in the treatment group (12%).

Figure 3.5: Extent to which respondents’ job matches their skills and interests, among SCR IW respondents, comparing treatment and control groups

Level of match to skills and interests	Treatment %	Control %
A lot	31	30
Somewhat	36	30
A little	19	20
Not at all	12	18
Don’t know	2	2

Base: all final survey respondents among SCR IW group who were in employment at the final survey (Unweighted base size: All: 921, Treatment: 506, Control: 415).

3.5. Overall job satisfaction

Satisfaction with current job varied significantly between the trial groups in the survey. Just over half (52%) of respondents in the SCR IW group were either ‘quite satisfied’ or ‘very satisfied’ with their current job. This was lower compared with the SCR OOW (74%) and WMCA (73%) group respondents. Similarly, 16% of SCR IW respondents were ‘very satisfied’ with their current job, compared with 26% and 27% in the SCR OOW and WMCA respondents, respectively. In contrast, 24% of respondents in SCR IW and 11% of respondents in both SCR OOW and WMCA groups were either ‘very dissatisfied’ or ‘quite dissatisfied’ with their current job (Appendix A: Table 128, Figure 3.6). Finally, 24% of SCR IW respondents were ‘neither satisfied nor dissatisfied’ with their current job, compared with 15% of the SCR OOW group and 16% of the WMCA group respondents.

Figure 3.6: Satisfaction with current job, by trial group

Level of satisfaction	All %	WMCA %	SCR OOW %	SCR IW %
Very satisfied	20	27	26	16
Quite satisfied	41	46	48	36
Neither satisfied nor dissatisfied	20	16	15	24
Quite dissatisfied	13	8	8	16
Very dissatisfied	6	3	3	8

Base: all final survey respondents who were in employment at the final survey (Unweighted base size: All: 1,549, SCR IW: 917, SCR OOW: 366, WMCA: 266).

Within the SCR IW group, there were significant differences between treatment and control groups (Appendix A: Table 129, Figure 3.7), with respondents in the treatment group (56%) more likely to feel ‘quite satisfied’ or ‘very satisfied’ with their current job compared with those in the control group (47%). Conversely, respondents in the treatment group (22%) were less likely to feel ‘quite dissatisfied’ or ‘very dissatisfied’ with their current job, compared with those in the control group (28%).

Figure 3.7: Satisfaction with current job among SCR IW respondents, comparing treatment and control groups

Level of satisfaction	Treatment %	Control %
Very satisfied	17	14
Quite satisfied	39	33
Neither satisfied nor dissatisfied	23	24
Quite dissatisfied	15	18
Very dissatisfied	7	10

Base: all final survey respondents among SCR IW group who were in employment at the final survey (Unweighted base size: All: 917, Treatment: 504, Control: 413).

4. Work search activities and self-efficacy

This chapter describes the steps taken by respondents who were not in employment at the time of the final survey to look for work, as well as their perceptions of their ability to achieve their goals, in particular their ability to successfully carry out work-related tasks.

4.1. Work search activities

Respondents who were not working at the time of the final survey were asked to indicate the number of hours a week (on average) they spent searching for a job (Appendix A: Table 114, Figure 4.1). Trial groups showed a similar pattern and there were no differences between them. Within all groups, the largest proportion had spent 1 to 10 hours a week searching for a job. Among the SCR IW group, 43% spent 1 to 10 hours searching for a job, compared with 37% of the SCR OOW group and 38% of the WMCA group. Across groups, a fifth of respondents (20%) spent no time each week searching for a job. There was a similar proportion of respondents reporting spending over 21 hours a week searching for a job, with 11% reporting spending 21 to 30 hours and 11% spending 31 hours or more (Appendix A: Table 114, Figure 4.1).

Figure 4.1: Number of hours a week (on average) respondents spent searching for a job, by trial group

Number of hours	All %	WMCA %	SCR OOW %	SCR IW %
31+ hours	11	12	10	12
21 to 30 hours	11	12	12	6
11 to 20 hours	19	20	19	16
1 to 10 hours	38	38	37	43
No time	20	19	21	23

Base: all final survey respondents not in work at the final survey (Unweighted base size: All: 2,299, SCR IW: 220, SCR OOW: 1,035, WMCA: 1,044).

On the other hand, in both SCR groups differences were found when comparing the treatment and control groups. Within the SCW IW group, the control group were more likely to have spent ‘no time searching for a job’, with 29% of the control group respondents spending no time each week searching for a job, compared with 17% of the treatment group. There was also a difference among the SCR OOW group, with 26% of the control group reporting no time each week searching for a job, compared with 15% of the treatment group (Appendix A: Table 115).

Number of jobs a week (on average) respondents applied for

Respondents who were not working at the time of the final survey were asked to indicate the number of jobs a week (on average) they applied for. Those in the SCR IW group were slightly more likely to have not applied for any jobs (34%), compared with those in the SCR OOW group (30%) and the WMCA group (25%). Within all groups, the largest proportion applied for 1 to 10 jobs a week. Among the SCR IW group, 49% applied for 1 to 10 jobs a week (on average), compared to 50% of the SCR OOW group and 50% of the WMCA group.

Figure 4.2: Number of jobs a week (on average) respondents applied for, by trial group

Number of jobs applied for	All %	WMCA %	SCR OOW %	SCR IW %
31+	6	8	5	5
21 to 30	4	5	4	4
11 to 20	11	12	11	9
1 to 10	50	50	50	49
None	28	25	30	34

Base: all final survey respondents not in work at the final-month survey (Unweighted base size: All: 2,311, SCR IW: 218, SCR OOW: 1,030, WMCA: 1,063).

There were differences between the treatment and control group across all trial groups when looking at the proportion of respondents who had not applied for any jobs. The control groups were more likely to have not applied for any jobs compared with the treatment groups. Among the SCR IW group, 43% of the control group had not applied for any jobs, compared with 24% of the treatment group. Among the SCR OOW group, 38% of the control group had not applied for any jobs compared with 22% of the treatment group and for the WMCA group the figures were 29% compared with 21% (Appendix A: Table 117, Figure 4.3). These findings align with the intervention level theory of change which expected that the trial should increase the number of jobs that respondents apply for.

Figure 4.3: Proportion of respondents who had not applied for any jobs, comparing treatment and control groups

Group	All %	WMCA %	SCR OOW %	SCR IW %
Treatment	22	21	22	24
Control	35	29	38	43

Base: All final survey respondents not in work at the final-month survey (Unweighted base size: All: 2,311, SCR IW: 218, SCR OOW: 1,030, WMCA OOW: 1,063).

4.2. Self-efficacy

4.2.1. Job search self-efficacy

The Job Search Self-Efficacy Scale (JSSE) assesses perceived level of self-efficacy related to work-search. The JSSE Scale captures respondents’ level of confidence in their ability to understand a variety of work-search tasks and achieve intended results. The scale is composed of two dimensions (job-search self-efficacy behaviour; job-search self-efficacy outcomes) consisting of 10 items each. All of the 20 items are rated on a 5-point Likert scale from 1 (not at all confident) to 5 (totally confident).

The JSSE Score was found to be significantly higher in the SCR IW group, when compared with those in the WMCA and SCR OOW groups (3.3, 3.2 and 3.18, respectively) (Table 4.1).

Table 4.1 JSSE mean score, by trial group

Base: all final survey respondents	SCR IW	SCR OOW	WMCA	Total	P-value
JSSE score	3.298	3.179	3.202	3.219	0.015
Unweighted base	1,135	1,405	1,342	3,882	–

When looking across individual statements or items associated with the JSSE (Appendix A: Table 28), respondents in the SCR IW group (27%) were more likely to feel ‘very confident’ in making a good list of all the skills they had, and which could be used to find a job, compared with those in the SCR OOW (22%) and WMCA groups (22%). Similarly, those in the SCR IW group were more likely to feel ‘very confident’ in searching and applying for jobs online: 41% of respondents in the SCR IW group reported that they were ‘very confident’ in searching for jobs online, compared with 35 and 37% of respondents in the WMCA and SCR OOW groups, respectively.

4.2.2. General Self-Efficacy

The General Self-Efficacy (GSE) Scale is designed to assess a general sense of perceived self-efficacy. The aim is to predict coping with daily hassles as well as adaptation after experiencing stressful life events. Studies have found that higher scores on the GSE scale are associated with positive emotion, optimism and work satisfaction. At the other end of the scale, lower scores are associated with depression, stress, health complaints, burnout and anxiety^{[footnote 6]}. The GSE scale is self-administered and consists of a total of 10 items. Responses are made on a 4-point scale: ‘not at all true’, ‘hardly true’, ‘moderately true’ or ‘exactly true’. The total GSE score is calculated by finding the sum of all 10 items. The total score ranges between 10 and 40, with a higher score indicating more self-efficacy.

The highest mean score was 28.0, among the SCR IW group respondents, which was higher than the SCR OOW group respondents and the WMCA group respondents (27.1 and 27.4, respectively) (Table 4.2). Although there were some significant differences between the trial groups for a small number of items associated with the GSE scale, overall, the trial groups responded similarly.

Table 4.2: General self-efficacy mean score, by trial group

General self-efficacy	SCR IW	SCR OOW	WMCA	Total	P-value
GSE score: mean	28.032	27.094	27.393	27.459	0.003
Unweighted base	1,121	1,350	1,265	3,736	–

Base: all final survey respondents.

5. Health conditions

At the time they were recruited to the trial, all respondents had a mild or moderate health problem or an impairment, which made it difficult for them to either find or stay in paid work. This chapter outlines the health status of respondents at the time of the final survey, one year on from joining the trial. It describes any health conditions and impairments they were living with and the impact of those on their daily activities including productivity at work.

5.1.1 Health conditions and impairments

A majority of respondents reported having one or more health conditions o r impairments in the final survey. No significant differences were found when comparing the SCR IW group, the SCR OOW group, and the WMCA group. A greater proportion of respondents in the WMCA group and SCR OOW group reported having one or more health conditions or impairments (91% and 90% respectively), compared with the SCR IW group (88%).

When comparing treatment and control groups across the trial groups, no significant differences were observed.

5.1.2. How much respondents’ health affects their daily activities

Among respondents who reported a health condition at the time of the final survey, around one-third reported that their health affected their ability to carry out everyday activities ‘a great deal’ across all trial groups (ranging from 29 to 33%). In addition, 41% of respondents in the SCR IW and WMCA groups and 39% of those in the SCR OOW group reported that their health ‘to some extent’ affected their daily activities (Appendix A: Table 16).

In the SCR OOW group, respondents in the treatment group were less likely to report that their health affected their daily activities ‘a great deal’, compared with respondents in the control group (29% versus 36%, respectively). No significant differences were found between the respondents in the treatment and control groups among the SCR IW and WMCA groups (Appendix A: Table 17).

5.1.3. Impact of health condition on productivity at work

To assess the impact of their health condition on productivity at work, respondents with a health condition who were working (not on sick leave) were asked two questions in the final survey. One centred on the number of hours missed from work due to health problems in the last 7 days, while the other asked to what extent their health problems affected their productivity.

Number of hours missed from work due to health problems

A majority of respondents across all three trial groups (ranging from 83-89%) did not miss any time from work due to health problems in the last 7 days (Appendix A: Table 18). Only a small percentage of respondents reported that 1 to 3 hours were missed from work due to health problems in the last 7 days (4% in the SCR IW group, 6% in the SCR OOW group, and 2% in the WMCA group). The mean number of hours missed from work due to health problems in the last 7days were 2, 2 and 1 in the SCR OOW group, WMCA group and SCR IW group, respectively.

No significant difference was found between the respondents in the treatment and control groups across all trial groups (Appendix A: Table 19).

How much health problems affect productivity

Respondents in employment at the time of the final survey were asked the extent to which their health conditions had an effect on their productivity at work during the last 7 days. Responses were invited on a scale ranging from 0 to 10: where 0 meant health problems had no effect on your work; and 10 meant health problems completely prevented you from working. Around a fifth (19%) of respondents reported that their health had no effect on their productivity at work, nearly a third (31%) that it had a minimal effect (1 to 3), 33% that it had some effect (4 to 6), and 16% that it had a substantial effect (7 to 9). Only 1% of respondents selected 10, indicating that it completely prevented them from working.

When interpreting the responses above, it should be noted that these respondents were those in employment and they were less likely to have health problems which made it impossible for them to work at all. When comparing the trial groups, the SCR OOW and WMCA groups were more likely to report less serious effects on their productivity, with around a quarter of respondents in these groups saying that their health had no effect on their work compared with 14% of the SCR IW group. This should be seen in the context that WMCA and SCR OOW respondents were less likely to be in employment at the time of the final survey, and those who were in work could be those with the least serious health problems, potentially allowing them to move more easily into work (Table 5.1).

No significant difference was found between the respondents in the treatment and control groups across all trial groups.

Table 5.1: Impact of health condition on productivity at work, measured by how much health problems affect productivity, by trial group

During the past 7 days, how much did your health problems affect your productivity	SCR IW %	SCR OOW %	WMCA %	Total %
0 – health problems had no effect on your work	14	26	25	19
1 to 3	32	34	27	31
4 to 6	36	26	34	33
7 to 9	18	13	13	16
10 – health problems completely prevented you from working	1	1	2	1
Unweighted base	605	233	177	1,015

Base: all final survey respondents with a health condition and in work (not on sick leave).

5.1.4. Musculoskeletal health

A majority of respondents in all trial groups reported having poor musculoskeletal health, 33% among SCR IW respondents, 29% in the SCR OOW group, and 28% in the WMCA (Appendix A: Table 22). To be described as having poor musculoskeletal health, respondents had to report having problems with either joints, back, neck, bone or muscles. These could be aches, pains or stiffness. They also had to report that in the last two weeks these symptoms had bothered them either ‘very much’ or ‘extremely’. There was a difference among SCR IW respondents with 70% of the treatment group reporting good musculoskeletal health compared to 64% in the control group (Appendix A: Table 23).

5.1.5. Self-management of symptoms

When asked how successfully they were able to manage their health problems, the most common response, reported by a majority of respondents in all groups, was that they were able to manage them ‘to some extent’, reported by 60% of SCR IW respondents, 55% of SCR OOW respondents, and 54% of WMCA respondents. Slightly under a third of respondents reported they were able to manage them ‘very well’ (29% in the SCR IW group, 32% in the SCR OOW group, and 31% among WMCA respondents). However, there was a minority of respondents who reported they were not able to manage their symptoms ‘at all’ (11% in the SCR IW group, 13% in the SCR OOW group, and 15% among WMCA respondents).

5.1.6. Visits to GP

Figure 5.1 below shows the number of visits made by respondents to a GP in the last 12 months (before they completed the final survey), excluding visits to the hospital.

This shows a wide range in number of visits made, with slightly less than 1 in 5 respondents making more than 10 visits to a GP, while others did not visit a GP at all in the same period. SCR IW respondents were slightly less likely to not have visited a GP, reported by 10% of respondents in this group compared with 15% of SCR OOW respondents and 17% of WMCA respondents.

Figure 5.1: Number of visits to a GP in the last 12 months, by trial group

Number of GP visits	All %	WMCA %	SCR OOW %	SCR IW %
None	14	17	15	10
One or two	23	22	23	23
Three to five	27	25	25	30
Six to ten	18	16	18	20
More than ten	19	19	19	18

Base: all final survey respondents (Unweighted base size: All: 3,998, SCR IW: 1,168, SCR OOW: 1,457, WMCA: 1,373).

5.2. Health and wellbeing

5.2.1. EQ5D5L dimensions

The EuroQol-5D-5L (EQ5D5L) is a standardised measure of health-related quality of life across five different dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. The EQ5D5L measures each of these across five levels, for example: whether a respondent has no problems with their mobility, slight problems, moderate problems, severe problems, or is unable to walk about at all. The respondent is asked to choose the statement which best reflects their health state for each dimension.

Figure 5.2 below summarises respondents’ responses in the areas of self-care, mobility and their ability to carry out their usual activities. Broadly these findings show respondents were least likely to have problems when asked about self-care, with 68% reporting no problems with washing or dressing and less than 5% reporting either severe problems or being unable to do so. This was followed by mobility, with around half of respondents reporting no problems walking, compared with 9% with severe problems and 1% reporting that they were unable to walk about at all.

The area that respondents were most likely to have problems with was their ‘usual activities’; 34% of respondents reported no problems with their usual activities compared with 10% who reported they had severe problems with them and 3% that they were unable to carry them out. In this area there were also differences between the trial groups. Respondents in the WMCA group were most likely to have no problems with their everyday activities (reported by 38% of WMCA respondents, compared with 33% in the SCR OOW group and 31% in the SCR IW group) (Appendix A: Table 40, 46, and 48).

Figure 5.2: How far respondents experienced problems with self-care, mobility and their usual activities, by trial group

Problems with walking about

Level of problem	All %	WMCA %	SCR OOW %	SCR IW %
No problems	51	49	51	53
Slight problems	21	20	21	21
Moderate problems	19	19	19	18
Severe problems	9	10	8	8
Unable to do carry out activity	1	1	1	0

Problems with doing usual activities

Level of problem	All %	WMCA %	SCR OOW %	SCR IW %
No problems	34	38	33	31
Slight problems	29	27	30	30
Moderate problems	24	23	25	25
Severe problems	10	9	9	11
Unable to do carry out activity	3	4	3	2

Problems with washing or dressing

Level of problem	All %	WMCA %	SCR OOW %	SCR IW %
No problems	68	68	68	68
Slight problems	16	15	16	18
Moderate problems	12	12	12	11
Severe problems	3	4	3	2
Unable to do carry out activity	1	1	1	0

Base: all final survey respondents (Unweighted base size: All: 4,057 to 4,076, SCR IW: 1,178 to 1,180, SCR OOW: 1,475 to 1,485, WMCA: 1,404 to 1,413).

The other two domains of the EQ5D5L measure how far people are experiencing anxiety or depression and discomfort or pain, summarised in Figure 5.3 below. Among survey respondents in all three trial groups, nearly a quarter reported either extreme or severe anxiety or depression. At the other end of the scale, around 45% of respondents reported either slight anxiety or depression or none at all (Appendix A: Table 42). When looking at pain or discomfort, the proportion reporting extreme or severe problems was slightly lower (less than one-fifth in all trial groups), while nearly 1 in 6 respondents reported either only slight pain or discomfort or none at all (Appendix A: Table 44).

Figure 5.3: How much respondents experienced pain/discomfort and anxiety/depression, by trial group

How much pain or discomfort the respondent reports

Level of pain/discomfort and anxiety/depression	All %	WMCA %	SCR OOW %	SCR IW %
None	31	32	32	29
Slightly/slight	28	26	28	30
Moderately/moderate	25	25	24	25
Severely/Severe	12	11	12	12
Extremely/Extreme	5	7	5	4

How anxious or depressed respondent reports they are

Level of anxiousness or depression	All %	WMCA %	SCR OOW %	SCR IW %
None	19	22	18	15
Slightly/slight	26	25	25	30
Moderately/moderate	31	30	32	33
Severely/Severe	15	14	16	14
Extremely/Extreme	9	8	10	9

Base: all final survey respondents (Unweighted base size: All: 4,058 to 4,067, SCR IW: 1,175 to 1,180, SCR OOW: 1,478 to 1,479, WMCA: 1,405 to 1,408).

EQ5D5L Index Scores

The utility index scores of the EQ5D5L were calculated^{[footnote 7]}. These range from zero (theoretical ‘death’) to one (theoretical ‘perfect health’). The mean EQ5D5L index scores were similar across all groups, with a mean score of 0.62 among SCR IW respondents, and 0.63 among both SCR OOW respondents and WMCA respondents (Appendix A: Table 36). This can be equated to approximately 60% of perfect health. No comparator population data is yet available for EQ5D5L. However, looking at the EQ5D3L (an earlier version of the score), for the UK population the index value was 0.856 in 2014^{[footnote 8]}. This suggests the scores reported by trial respondents are likely to be lower than overall population levels.

EQVAS Score

The EQ5D5L also incorporates a visual analogue scale (EQVAS). This is represented as a thermometer ranging from zero (worst imaginable health state) to 100 (best imaginable health state), and respondents are asked to indicate on the scale how good or bad they perceived their health state to be on the day that they completed (or were supported to complete) the final survey. The highest mean score was 59.8, among the SCR IW group respondents, which was significantly higher than the score of 56.3 in the SCR OOW group and 57.8 among WMCA respondents (Appendix A: Table 38). It is not possible to determine from these data whether those who were able to find work were healthier, or whether work itself supports increased health.

5.2.2. Wellbeing

The respondents to the final survey were asked a series of questions drawn from the Short Warwick Edinburgh Mental Wellbeing Scale (SWEMWBS)^{[footnote 9]}. The scale uses seven statements about thoughts and feelings; respondents were asked to describe their experience related to, for example, feeling optimistic about the future, feeling relaxed, feeling useful and dealing with problems well. Each of the seven statements has five response categories ranging from ‘none of the time’ to ‘all of the time’, in order to monitor mental wellbeing. The total score is obtained by summing the score for each of the statements, producing a score of between 7 and 35. The total raw scores are then transformed into metric scores. Answers to these questions were used to create three groups: people with ‘high’ mental wellbeing whose total score was above 28, those with ‘average’ mental wellbeing (scores between 21 and 27), and those with ‘low’ mental wellbeing – a score of 20 or less. When comparing SWEMWBS scores with measures of depression, falling into the low mental wellbeing group is considered to be indicative of possible depression for scores between 18 and 20, and probable depression for scores of 17 or less^{[footnote 10]}.

Low mental wellbeing was reported by 60% of those in SCR IW, 62% of those in SCR OOW, and 59% in WMCA. Approximately one-third of respondents reported average mental wellbeing, and less than 1 in 10 respondents reported high mental wellbeing, with no significant differences present between the trial groups (Appendix A: Table 34; Figure 5.4).

Figure 5.4: Short Warwick Edinburgh Mental Wellbeing Score, by trial group

Mental Wellbeing Score	All %	WMCA %	SCR OOW %	SCR IW %
Low mental wellbeing	61	59	62	60
Average mental wellbeing	32	32	31	34
High mental wellbeing	7	9	7	6

Base: all final survey respondents (Unweighted base size: All: 4,002, SCR IW: 1,168, SCR OOW: 1,449, WMCA: 1,385).

Satisfaction with life

Life satisfaction was measured using the first of the Office for National Statistics (ONS) personal wellbeing questions which asked ‘Overall, how satisfied are you with your life nowadays?’ on a scale of 0 to 10, where 0 is ‘not at all’ and 10 is ‘completely’^{[footnote 11]}. Life satisfaction varied significantly between the different groups of respondents: 22% of respondents in the SCR IW group and 25% of those in WMCA had no or little life satisfaction, compared with 30% of SCR OOW respondents. Slightly over one-third of SCR IW (40%), SCR OOW (36%) and WMCA (39%) respondents had neutral life satisfaction. General or complete life satisfaction was reported by 38% of SCR IW respondents, 34% of respondents in the SCR OOW group and 36% of those in WMCA (Appendix A: Table 30, Figure 5.5).

Figure 5.5: Life satisfaction, by trial group

Level of life satisfaction	All %	WMCA %	SCR OOW %	SCR IW %
Low life satisfaction (score 0-3)	26	25	30	22
Medium life satisfaction (score 4-6)	38	39	36	40
High life satisfaction (score 7-10)	36	36	34	38

Base: all final survey respondents (Unweighted base size: All: 4,038, SCR IW: 1,170, SCR OOW: 1,463, WMCA: 1,405).

GAD-7 score

The GAD-7 is designed to measure or assess the severity of generalised anxiety disorder (GAD)^{[footnote 12]}. It is comprised of 7 items, each of which asks the individual to rate how often they experienced their symptoms over the past two weeks. These questions include how often people feel ‘nervous, anxious or on edge’, how often they have ‘trouble relaxing’, whether they become irritable easily, find themselves restless or unable to sit still, or are unable to ‘stop or control worrying’. The GAD-7 score is calculated by assigning scores of 0, 1, 2, and 3 to these response categories, and adding together the total score for the 7 questions, with a high score indicating greater anxiety. Scored out of a total of 21, scores of 5, 10, and 15 are taken as the cut-off points for mild, moderate and severe anxiety, respectively. When used as a screening tool, further evaluation is recommended when the score is 10 or greater.

There were no significant differences between the average scores of the trial groups, with a mean score of 9.2 among the SCR IW group, 9.4 among the SCR OOW group, and 9.1 among the WMCA group (Appendix A: Table 134), suggesting that the three trial groups experienced similar levels of anxiety overall, which was, on average, likely to be mild to moderate. However, this still meant around a quarter (26%) were suffering severe anxiety and another fifth (19%) moderate anxiety.

PHQ-9 score

The Patient Health Questionnaire (PHQ-9) is a multipurpose instrument for screening, monitoring and measuring the severity of depression^{[footnote 13]}. Questions include how often people feel ‘down, depressed or hopeless’, feel ‘tired or have little energy’ or feel bad about themselves – that they are a failure and have let themselves or their family down. As with the GAD-7, a total score is calculated by assigning scores of 0, 1, 2, and 3, to the response categories of ‘not at all’, ‘several days’, ‘more than half the days’, and ‘nearly every day’. Out of a possible total of 27, scores of 5, 10, 15 and 20 represent mild, moderate, moderately severe and severe depression respectively.

There were no significant differences between the average scores of the trial groups, nor were there significant differences when comparing the average scores of the control and treatment groups. The mean score for all trial groups was just above the cut-off point for moderate depression, with an average score of 10.7 among the SCR IW group, 10.8 among the SCR OOW group, and 10.5 among the WMCA group. However, similar to the GAD-7 score, this measure suggests there is a substantial subset of people experiencing more severe mental health problems, with 13% of respondents classified as having severe depression and a further 17% moderately severe depression (Appendix A: Table 132).

Social contact

Strong social relationships are a vital part of an individual’s wellbeing and quality of life. As a result, the frequency of someone’s social interactions acts as an indicator of their quality of life and can give a broader perspective on overall wellbeing. Final survey respondents were asked how often they are in touch with family or friends, either face-to-face, by phone, electronically or by any other means. Nearly half of respondents reported having daily social contact with their friends or family with no significant differences between trial groups: 48% of the SCR OOW and WMCA groups reporting daily social contact, and 49% of the SCR IW group (Appendix A: Table 136). There were no significant differences, in any of the trial groups, between the treatment and control groups in terms of the number of respondents reporting daily social contact.

In order to examine the possible effect of the COVID-19 pandemic on social contact, respondents were split into three cohorts: those who responded to the survey pre-COVID-19 (June 2019 – February 2020); those who responded to the survey during COVID-19, but whose 12 months in the intervention would have been finished pre-COVID-19 (March – June 2020); and those whose 12 months in the intervention would have fallen during COVID-19, as well as their completion of the final survey (July – November 2020). Respondents across the three cohorts reported similar levels of social contact^{[footnote 14]}, with the number reporting daily social contact highest for those who responded during the pandemic, but for whom the intervention finished pre-COVID-19 (50%) (Appendix A: Table 138).

Figure 5.6: Proportion of respondents reporting daily social contact, by interview date

Interview date	Percentage
June 2019 to February 2020	47
March to June 2020	50
July to November 2020	49

Base: all final survey respondents (Unweighted base size: All: 4,067).

6. Barriers to employment

This section describes how far the respondents felt ready to return to work at the time of the final survey, 12 months after joining the trial, and any barriers they faced in moving into work. Most reported their health, either physical or mental, was a barrier to work. However, there was also a range of other factors that served as barriers which respondents highlighted and which they often described as more important than their health.

6.1. Readiness for work

All respondents who were not in paid work at the time of the final survey were asked how they felt their health affected their ability to return to work. The most common response in all trial groups was that ‘on some days’ they could think about returning to paid employment, reported by 40% of the SCR IW group, 45% of the SCR OOW group and 44% of the WMCA group (Appendix A: Table 50). The SCR IW group respondents were most likely to say their health problem ‘ruled out work’, reported by more than a quarter (27%), compared with 24% of the SCR OOW group and 21% of the WMCA group. This should be seen in the context that a higher proportion of the SCR IW trial group were in paid work at the final survey, so those not in paid work may be those with more severe health problems that have caused them to leave paid employment since the beginning of the trial. Among all the different trial groups, approximately a third (33% of the SCR IW group, 32% of the SCR OOW group, and 35% of WMCA) said that, despite their health problems, they could ‘consider returning to work right now’.

When comparing the treatment and control groups among the three trial groups, the treatment group among the WMCA and SCR OOW respondents were less likely to feel that their health would prevent them returning to work. In the SCR IW group, 30% of control group respondents felt their health ‘rules out work’ compared with 23% in the treatment group. There was a similar pattern among SCR OOW respondents, where in the control group 26% felt their health ruled out a return to work compared with 21% in the treatment group. Finally, among the WMCA group 24% of the control group felt their health condition ruled out work compared with 19% of the treatment group (Appendix A: Table 51). This may reflect the achievement of a number of intermediary outcomes in the treatment group. This could be in terms of preparing for work, such as advising them on how to disclose their health condition at work, signposting them to support for their health condition, or helping them identify suitable jobs for their situation. The experience of receiving in-work support could have improved participant’s management of their health condition in the workplace, potentially making them more optimistic about returning to work in the future. If they found work during the trial, IPS specialists may also have negotiated them better working conditions, which also could have left them with a more positive attitude towards working.

6.2. Barriers to finding work

All respondents to the final survey were asked whether they currently experience, or have previously experienced, a range of different barriers to finding paid work. The majority reported facing more than one barrier, with over half of respondents in all trial groups reporting difficulty finding a suitable job, mental health problems, and a lack of confidence in their abilities and skills.

The most common barrier selected by people in the SCR OOW and WMCA groups was difficulty finding a suitable job, reported by roughly three-quarters compared with 55% of people in the SCR IW group (Figure 6.1). The next three most prevalent barriers overall were a lack of confidence in abilities and skills, mental health conditions, and lack of qualifications or experience. The differences in the proportion of respondents across the trial groups reporting mental health and a lack of qualifications and experience as barriers were significant. Mental health was reported as a barrier by nearly two-thirds of SCR OOW (62%) compared with 55% of SCR IW respondents and 51% of those in WMCA group. A lack of qualifications and experience was more common among the OOW groups, reported by 57% in the WMCA group and 50% in the SCR OOW group, compared with 40% of the SCR IW group.

Figure 6.1: Barriers to work, by trial group

Barriers to work	WMCA %	SCR OOW %	SCR IW %
Difficulty finding a suitable job	77	75	55
Lack of confidence in abilities or skills	57	59	54
Lack of qualifications or experience	57	55	40
Physical health condition	52	49	41
Mental health condition	51	62	55
Availability or cost of transport to work	45	43	26
Being financially worse off	35	36	38
Caring for a child, or an elderly or disabled family member	15	16	13
Availability or cost of childcare	10	9	9

Base: all final survey respondents (Unweighted base size: All: 4,057, SCR IW: 1,174, SCR OOW: 1,481, WMCA: 1,402).

There were further significant differences between trial groups with regard to the proportion of respondents reporting physical health and transport as barriers to work. Having a physical health condition was more common in the OOW groups (52% of WMCA respondents and 49% of SCR OOW respondents) than in the SCR IW respondent group (41%). The availability and cost of transport to work was reported as a barrier by nearly half (45%) of WMCA respondents and 43% of SCR OOW respondents, compared with only 26% of SCR IW respondents. There were no significant differences between the treatment and control groups in terms of how likely they were to report the different barriers to work.

Most barriers had stayed fairly constant through the lifetime of the trial, with similar proportions of respondents reporting each barrier at the final survey as they did in the baseline. Areas where there were some changes included difficulties finding a suitable job, which fell across all trial groups from 77% to 70%. In the two OOW trial groups, being financially worse off in work also became more common, with 36% in the SCR OOW group reporting this barrier at the final survey compared with 27% in the baseline survey. There was a very similar shift in the WMCA group (Appendix A: Table 140).

Overall, the majority of respondents experienced more than one barrier to employment and the barriers experienced across the trial groups were very diverse. For example, a substantial number of respondents reported experiencing barriers that were not directly related to work, such as caring responsibilities, health conditions and transport. This highlights the importance of addressing a variety of barriers, not simply those that relate directly to work, alongside the provision of employment support – an assumption that underpins the IPS approach.

6.3. The most important barrier to working

With most respondents reporting multiple barriers to finding paid work, it is useful to know which of the barriers they faced were felt to be the most important. To address this, when in the final survey respondents selected multiple barriers to work, they were asked to identify their ‘most important’ barrier. The ‘most important’ barriers respondents selected are presented below, combined with the responses of those who only selected one barrier to finding work in the first questions.

The barrier most commonly selected as the most important in the SCR IW and SCR OOW trial groups was mental health, reported by 24% of SCR OOW respondents and 23% of SCR IW respondents (Appendix A: Table 56). Comparatively, the number of respondents in the WMCA group selecting mental health as the most important barrier (19%) was significantly lower. The second most important barrier among the SCR OOW and SCR IW respondents was physical health, identified by 18% of respondents in both groups. In the WMCA group, physical health was the most commonly reported barrier with 20% of respondents selecting it as the most important barrier. This suggests that addressing the health problems of trial recruits is crucial to moving many into paid work. As noted in section 6.1, when asked how they felt about returning to work given their health problems, 21% of WMCA respondents, 27% of SCR IW respondents and 24% of SCR OOW respondents said that their health ruled out paid work as an option.

The next ‘most important’ barrier identified for the SCR OOW and WMCA groups was difficulty finding a suitable job (reported by 15% and 16% respectively). For the WMCA group, the next most important barrier was lack of qualifications and experience (15%), differing significantly from the 11% of respondents in both the SCR IW and SCR OOW groups selecting this as the most important barrier. A lack of confidence in their skills and abilities was more common among the SCR IW respondent group, 15% of whom reported this as their main barrier to working. This was different from the WMCA and SCR OOW groups, where 9% of respondents in both groups reported this as their main barrier to work.

Figure 6.2: Most important barrier to working, by trial group

Most important barrier to working	WMCA %	SCR OOW %	SCR IW %
Mental health condition	19	24	23
Physical health condition	20	18	18
Lack of confidence in abilities or skills	9	9	15
Lack of qualifications or experience	15	11	11
Availability or cost of transport to work	4	4	3
Difficulty finding a suitable job	16	15	12

Base: all final survey respondents (Unweighted base size: All: 3,665, SCR IW: 984, SCR OOW: 1,383, WMCA: 1,298).

The control and treatment groups were largely similar in terms of the main barriers to work they selected. The only difference was in the SCR OOW group, where the proportion of respondents reporting physical health as the most important barrier to work was lower in the treatment group than the control group (15% compared with 21%) (Appendix A: Table 58). This difference could be a result of the intervention successfully addressing fears and misconceptions about employment negatively affecting health, and therefore those who received the support being less likely to see physical health as the most significant barrier to employment. However, this result is not consistent across the trial groups and so we cannot be certain that this is what is driving the difference.

6.4. Perceptions of how health will be affected by work

Respondents who were working at the time of the final survey and those who were not working at time of the final survey were asked separately about how their ability to manage their health will be affected by starting work.

Among respondents who were not in work the most commonly recorded response across all trial groups was that work would make it harder to manage their health, with the majority of respondents in the SCR IW (61%) and SCR OOW (53%) groups, and just under half of WMCA respondents (49%) selecting this response (Appendix A: Table 60). The second most recorded response was that work would make it easier to manage health, with the least popular response being that work would have no effect on health management. When comparing the treatment and control groups there were no differences among the SCR OOW and WMCA groups. However, in the SCR IW group the control group were more likely to say working made it harder to manage their health, with 71% reporting this in the control group compared with 51% in the treatment group (Appendix A: Table 61).

The responses of those in employment at the time of the final survey when asked how their ability to manage their health is affected by work were significantly different. The majority of SCR IW respondents (56%) reported that work makes it harder to manage their health, compared with 35% of SCR OOW respondents and 27% of the WMCA group. The most commonly recorded response among the SCR OOW and WMCA groups was that work made it easier to manage their health (42 and 40% respectively), compared with just 26% of the SCR IW group (Appendix A: Table 62).

Figure 6.3: Proportion of in-work respondents reporting that work makes it harder to manage health conditions, by trial group

Trial group	Percentage
All	46
WMCA	27
SCR OOW	35
SCR IW	56

Base: all final survey respondents in work at the time of the final survey (Unweighted base size: All: 1,316, SCR IW: 788, SCR OOW: 307, WMCA: 221).

When comparing the treatment and control groups for the working respondents, a significant difference was found between those in the SCR OOW group. The most commonly reported response within the control group was that work made it harder to manage their health (44% compared with 27% in the treatment group). In comparison, the most commonly reported response within the treatment group was that work made it easier to manage their health (47% compared with 37% in the control group) (Appendix A: Table 63). Across all trial groups, the treatment group was more likely than the control group to report that work made managing health conditions easier. This supports the causal pathway outlined in the intervention level theory of change that through being in work and through IPS physical and mental health will be improved. This finding suggests that the support goes some way to address fears and misconceptions about the potential negative impacts of employment.

Figure 6.4: Proportion of SCR OOW respondents reporting that work makes it harder to manage health conditions, by treatment group

Treatment group	Percentage
All	35
Treatment	27
Control	44

Base: all SCR OOW respondents in work at the time of the final survey (Unweighted base size: All: 307, Treatment: 136, Control: 171).

7. Perceptions of the intervention support

This chapter reports on the perceptions of respondents in the treatment group about the support they received during the trial. This includes their attitudes to the employment specialist who delivered their support. It also covers how helpful they found the support in managing their health (in general and in work-related contexts) and in preparing them for work, for example, in terms of increased confidence or motivation to work.

7.1. Perceptions of the employment specialist

Respondents in the treatment group were asked how well their employment specialist understood ‘their needs’ and whether they had the ‘right skills and expertise’ to help them. They were asked how far they agreed with these statements with response categories: ‘a lot’, ‘a little’ or ‘not at all’. The chart below shows a majority of respondents felt ‘a lot’ that the employment specialist understood their needs and had the right skills and expertise. It also shows this positive perception was more common among SCR respondents (Appendix A: Tables 64 and 69).

Figure 7.1: Perceptions of the employment specialist, by trial group

Perceptions	All %	WMCA %	SCR OOW %	SCR IW %
Respondents who felt ‘a lot’the employment specialist had right skills and expertise	69	58	73	76
Respondents who felt the employment specialist understood their needs ‘a lot’	68	56	72	76

Base: all treatment group respondents (Unweighted base size: All: 2,102 to 2,078, SCR IW: 619 to 625, SCR OOW: 741 to 749, WMCA: 718 to 728).

However, a substantial minority of respondents in the trial groups had quite negative attitudes towards their employment specialist. For example, in the WMCA group slightly more than 4 in 10 felt their needs were understood either ‘not at all’ or only ‘a little’. Similarly, around 4 in 10 WMCA respondents felt the employment specialist had the right skills and expertise only ‘a little’ or ‘not at all’. Around 1 in 4 respondents in the SCR groups also reported these less favourable attitudes.

7.1.1. Which respondents were more likely to have a positive view of the employment specialist?

To assess whether some respondents in the treatment group might have benefitted more in terms of the support they received, their attitudes to the support were analysed by their health, mental wellbeing, life satisfaction and educational level. Overall, this showed that respondents in the treatment group with better health, life satisfaction and greater mental wellbeing had more positive views of the employment specialist, although this was not the case in all areas.

In this section, respondent health is measured by how much their health condition or impairment limited their everyday activities, either ‘a great deal’, ‘to some extent’ or ‘a little/not at all’. The indicator for life satisfaction is based on the ONS scale of how satisfied people are with their life, with responses grouped into low, medium and high life satisfaction. Mental wellbeing is assessed using the SWEMWS, grouping people into low, neutral, or high mental wellbeing. Finally, educational level is categorised based on the highest qualification each person holds, grouping them into those with:

• no formal qualifications
• qualifications up to GCSEs of A to C grade or equivalent
• qualifications up to A-levels or equivalent
• degree level qualifications and above

Looking first at how well the employment specialist understood their needs, respondents in the SCR OOW treatment group healthier people were more likely to report the specialist understood their needs ‘a lot’. There was a similar pattern among SCR IW and WMCA treatment group respondents, but this difference was not significant. In the SCR OOW and WMCA groups, people with higher life satisfaction were more likely to say the adviser understood their needs ‘a lot’, although this was not significantly different among the SCR IW group. In all trial groups those with better mental wellbeing were more likely to say the adviser understood their needs ‘a lot’, compared with those with low mental wellbeing. There were no differences by education.

Turning to how far respondents in the treatment group felt the employment specialist had the right skills and expertise, again among SCR OOW treatment group respondents only, healthier people were more likely to report feeling the adviser had the right skills ‘a lot’. Overall, respondents in the treatment group with high life satisfaction were more likely to report feeling ‘a lot’ that the adviser had the right skills, and those with better mental wellbeing were also more likely to say they agreed ‘a lot’ that the adviser had the right skills and expertise. Finally, in the WMCA group, higher levels of education were associated with being more likely to agree ‘a lot’ that the adviser had the right skills and expertise (Appendix A: Tables 65 to 8 and 70 to 3).

7.2. Health-related support

Respondents in the treatment group were also asked about how helpful the IPS service was in terms of their health across three different areas.

In the first of these, making referrals to other support for their health, a majority of respondents reported the trial support had been helpful in this area. However, WMCA respondents were again less likely than those in the SCR groups to have found it helpful (58% in the WMCA group compared with 73% and 76% among SCR OOW and SCR IW respondents respectively).

Respondents were also asked about how helpful the support was in identifying adaptations which could be made in their work that might help them manage their health better. These adjustments would then allow them to manage their health better at work and might include physical adjustments or changes to their work duties or hours. Again, a majority of respondents in the treatment group found the support helpful in this area. SCR IW respondents were most likely to have found it helpful (68% reporting this), followed by SCR OOW respondents (61%) and WMCA respondents (55%).

Finally, respondents in the treatment group were asked how far the support had been helpful for their own ability to manage their health condition/impairment. This links to the causal pathway outlined in the intervention level theory of changed that through being in work and through IPS physical and mental health will be improved. SCR IW respondents were most likely to have found it helpful (74%), followed by SCR OOW respondents (69%), while those in the WMCA group were least likely to have found it helpful (59%) (Appendix A: Tables 74, 79 and 84). These results are summarised in Figure 7.2 below.

Figure 7.2: Proportion of respondents who found the support helpful in health-related areas, by trial group

Health related area	All %	WMCA %	SCR OOW %	SCR IW %
Referring them to other health support	68	58	73	76
Identifying work adapations to better manage health at work	61	55	61	68
For their ability to manage their health condition/disability	67	59	69	74

Base: all treatment group respondents (Unweighted base size: All: 2,108 to 2,121, SCR IW: 619 to 631, SCR OOW: 725 to 750, WMCA: 691 to 738).

7.2.1. Which respondents had more positive perceptions of the intervention’s health-related support?

Respondents’ perceptions of how effective the support was in helping them in health-related areas were analysed using the same four variables as above, specifically: health, life satisfaction, mental wellbeing, and education. Overall, this analysis showed a similar pattern, with respondents in the treatment group who reported better life satisfaction, physical health and mental wellbeing generally reporting more favourable attitudes towards the support they received.

Looking first at how helpful IPS had been in referring treatment group respondents to other sources of health-related support, those with greater life satisfaction in all trial groups were more likely to have found it helpful. In SCR OOW and WMCA treatment groups, respondents with better health were also more likely to have found it helpful. Additionally, SCR IW and WMCA treatment group respondents with average or high levels of mental wellbeing were also more likely to have found the support helpful in referring them to other support. There was a similar pattern among SCR OOW respondents, although it was not statistically significant (Appendix A: Tables 85-8, Figure 7.3).

Figure 7.3: Proportion of respondents who found the support very helpful in referring them to other sources of health-related support, by life satisfaction

Level of life satisfaction	All %	WMCA %	SCR OOW %	SCR IW %
Low life satisfaction (score 0 to 3)	26	17	28	37
Medium life satisfaction (score 4 to 6)	34	19	38	46
High life satisfaction (score 7 to 10)	46	35	50	52

Base: all final survey respondents in treatment group (Unweighted base size: All: 2,108, SCR IW:631, SCR OOW: 750, WMCA OOW: 727).

Turning to the helpfulness of the support in identifying workplace adaptations, again respondents in the treatment group who did not have low mental wellbeing were more likely to have found the support helpful across trial groups. In addition, in SCR OOW and WMCA groups, respondents with better health and life satisfaction were also more likely to have found the support helpful in this regard (Appendix A: Tables 80-3).

Finally, in terms of their ability to manage their own health, across all trial groups respondents in the treatment group who were healthier, reported higher life satisfaction and who did not have low mental wellbeing, were more likely to report that the support had been helpful (Appendix A: Tables 75 to 8).

7.3. Readiness for work

How far respondents in the treatment group felt the support they received had helped their readiness for work was measured in terms of how helpful it had been for their confidence in their skills and their ability to decide what type of job they want. As Figure 7.4 shows, a majority of treatment group respondents in all trial groups found the support helpful in these areas, although respondents in the WMCA group were less likely to report this than SCR respondents. Where treatment group respondents did not find the support helpful, they were most likely to give neutral responses (neither helpful nor unhelpful), and only around 1 in 10 found the support unhelpful. For example, the proportions who found the support unhelpful for their confidence in their skills were, 13% of WMCA respondents, 8% of SCR OOW respondents and 6% of SCR IW respondents (Appendix A: Tables 89 and 94).

Figure 7.4: Helpfulness of the support in preparing people for work

Type of helpfulness	All %	WMCA %	SCR OOW %	SCR IW %
Helpful in deciding what type of job they want to do	69	64	73	70
Helpful for their confidence in their skills	71	66	73	75

Base: all treatment group respondents (Unweighted base size: All: 2,098 to 2,107, SCR IW: 625 to 631, SCR OOW: 745 to 747, WMCA: 726 to 731).

7.3.1. Which respondents had more positive views of the intervention’s work-related support?

The views of respondents in the treatment group on the effectiveness of the intervention in work-related areas showed a similar pattern by health, life satisfaction and mental wellbeing.

Looking first at how helpful the support had been in deciding what type of job to do, respondents in the treatment group with better health and life satisfaction, and who did not report low mental wellbeing, were all more likely to have found the support helpful. There was no association with education on this for SCR IW and WMCA respondents although for SCR OOW treatment group respondents the results showed that those with degree level qualifications or above were less likely to have found the support helpful than those with lower levels of education, although even in this group 62% reported the support had been helpful in deciding what type of job they wanted to do (Appendix A: Tables 95 to 8).

A similar pattern was seen in respondents’ perceptions of how helpful the support was for confidence in their skills. Respondents in the treatment group who had higher levels of health and life satisfaction and either average or high levels of mental wellbeing more likely to have found the support helpful. There was no association with education (Appendix A: Tables 90 to 3).

7.4. Motivation to return to work

Overall, the responses indicate a positive effect on treatment group respondents’ motivation to work, with a majority of treatment group respondents in all trial groups reporting their motivation to work had increased. Motivation to work was asked about separately for responding not in work at the time of the final survey, those who were in work, and those who were on sick leave.

Motivation to return to work among those not in work

The first analysis explored treatment group respondents who were not in work at the time of the final survey. This group was asked about how their motivation to find work had been affected by the support they received during the trial. Respondents in the SCR OOW treatment group were most likely to report that their motivation to work had increased, with nearly three-quarters (73%) reporting this, compared with 66% among SCR IW respondents and 59% in the WMCA respondents. At the other end of the scale, 6% of respondents in both SCR groups and 16% of WMCA respondents reported their motivation to work had been decreased by the support (Appendix A: Table 99).

Motivation to work among those in employment

Among those in work at the time of the final survey again the majority reported the support had increased their motivation to work when they were most recently not working^{[footnote 15]}. The SCR IW group were the least likely to say their motivation to work had increased, despite being the most positive about support in other areas. However, even in the SCR IW group 74% said their motivation had increased. This compared with 85% in the SCR OOW group and 81% among WMCA respondents. Very few respondents who were not working at the time of the final survey reported their motivation to work had fallen due to the support (5% or less in all groups) (Appendix A: Table 109, Figure 7.5).

Motivation among those on sick leave

Finally, among those respondents in the treatment group who were on sick leave at the time of the final survey, only the SCR IW group had a large enough base size to report on. Among these, again a majority (66%) said the support had increased their motivation to return to work. Only 4% reported it had decreased their motivation (Appendix A: Table 104).

Figure 7.5: How much the support increased or decreased motivation to return to work among those out of work, by trial group

Series	4	3	2	1
Series 1	19%	14%	15%	25%
Series 2	53%	53%	58%	50%

Base: All final survey respondents in the treatment group who were in employment at the twelve-month survey but who have been out of work since the start of the trial (Unweighted base size: All: 462, SCR IW: 210, SCR OOW: 143, WMCA OOW: 109).

7.4.1. Which respondents were more likely to report an increase in motivation to work?

For treatment group respondents on sick leave and those in work at the time of the final survey, base sizes were too small to report on how motivation to work varied by health and wellbeing characteristics. For respondents in the treatment group who were not in work at the time of the final survey, those with better health and higher life satisfaction and mental wellbeing tended to be more likely to report positive attitudes towards the support, although differences were not statistically significant for all trial groups. In the SCR OOW group, healthier respondents were more likely to report their motivation to work had increased, while in both the SCR OOW and WMCA groups, respondents with better life satisfaction and mental wellbeing were more likely to report increased motivation to return to work (Appendix A: Tables 100 to 103, 105 to 108, 110 to 113).

8. Discussion

When comparing trial groups in the final survey data, the SCR IW group were most likely to have been in employment during the trial, to have had stable employment, and also being less likely to report some of the work-related barriers to work. They were less likely to report difficulty finding a suitable job and a lack of qualifications or experience as barriers to finding work. Their demographic and baseline data support this. They reported higher levels of education, being more likely to own their own homes, and having stronger work histories before the trial started. In terms of their health, however, the trial groups were fairly similar and respondents in all trial groups reported substantial effects from health conditions and impairments, which continued to pose a barrier to their returning to work. In every trial group either physical or mental health were the most common barriers preventing participants from returning to work.

Overall, most respondents in the treatment group reported positive views of the support they had received in the trial. Where they did not indicate a positive perception, more indicated a neutral rather than negative view. Those in the SCR IW treatment group were most likely to report favourable attitudes towards the IPS service and WMCA respondents were the least likely. However, even among WMCA respondents, a majority still reported positive views. The difference between SCR IW respondents and the other trial groups may be affected by their different work histories before the trial. In contrast to the SCR OOW and WMCA groups, the SCR IW group were struggling in their current job and unlikely to be receiving employment support which may have led to their more favourable views.

Treatment group respondents whose health problems interfered less with their day-to-day activities, who had better mental wellbeing and who reported higher satisfaction with their lives were also more likely to report the positive attitudes towards the support they received. This suggests that people with more severe health problems or impairments, whether these affect them physically or mentally, were less likely to benefit from the IPS services as configured for the trials. However, IPS has more commonly focused on groups with higher needs. It is possible the needs of this group were greater than envisaged when the trials were designed. This needs consideration when targeting IPS services to the groups recruited to the trials.

Health remained an important problem for the majority of respondents and was a key barrier they faced to work. Around 9 in 10 respondents had at least one health condition or impairment at the final survey, and around 7 in 10 reported that this continued to affect their day-to-day activities ‘to some extent’ or ‘a great deal’.

It was an important criterion of the trial that, while respondents were likely to continue experiencing health problems, these would not be so serious that it was impossible for them to move into work. The theory of change underlying the trial highlighted that the IPS support would increase the treatment group’s knowledge of job opportunities, their self-efficacy in work search, help them establish job goals linked to their strengths and capabilities, and potentially improve their health-related outcomes and that through this they would feel more able to return to work.

One year after randomisation, the final survey shows that, while around a third respondents did feel able to work, a significant proportion stated that their health ‘rules work out as an option’ (between 21 and 27% across trial groups). This may suggest that many respondents had serious health problems, which were challenging for the IPS services to help them overcome. The employment histories of all trial groups illustrate that respondents struggled to stay in employment, with only around 1 in 5 WMCA and SCR OOW respondents in work at the time of the final survey. Even in the SCR IW group, where recruits joined the trial while in work, slightly less than 1 in 5 respondents had become unemployed by the time of the final survey.

The second causal pathway of the theory of change assumed that by being in work, alongside the trial support, the treatment group’s health would be improved. Respondents’ attitudes towards this were mixed and some did agree that being in work had helped them manage their health, but overall, a majority felt it had either no effect or actually made it more difficult for them to do so.

Turning to experiences of being in employment during the trial, the first causal pathway of the theory of change argues that with a better understanding of their skills and interests, and individualised support from their employment specialists, recruits to the treatment group would be able to find a job which fitted their abilities and which they found satisfying. Achieving a good quality of job match would be vital in helping them sustain their employment, while also finding their work fulfilling. The final survey suggests that respondents’ experience on these dimensions was mixed. Around a third of respondents in all groups who were in work reported the job matched their skills and interests ‘a lot’, but another third reported it matched them either ‘not at all’ or only ‘a little’. When looking at their overall satisfaction the SCR IW group was least satisfied (52% saying they were satisfied compared with 74% in SCR OOW and WMCA groups). It is not clear whether this was because they remained in the same job and had not been able to achieve new adjustments or had moved jobs but still found their situation unsatisfactory.

Overall, the results show a positive perception of the support received by respondents, nonetheless respondents continued to face major barriers to finding work. These included persistent health problems, which continued to be challenging in a work context and which respondents did not always feel were helped by being in work. These challenges were particularly serious for people with worse health problems, lower mental wellbeing or lower life satisfaction.

1. The area has since rebranded as South Yorkshire Mayoral Combined Authority ↩

2. Findings related to differences between control and treatment groups that are not presented in this report and will be discussed in the Impact Report include: Employment at the time of the final survey, EQ5D5L, SWEMWBS score, Working 16 hours a week or more, Number of weeks worked since randomisation, Number of weeks worked 16 hours a week or more since randomisation, Job Search Self-Efficacy (JSSE), Musculoskeletal health, GAD-7, PHQ-8, DDA health definition, Life Satisfaction (ONS1) and General Self-Efficacy (GES). ↩

3. Face-to-face interviews were discontinued from batch 11 onwards due to the coronavirus pandemic, and from this point on all respondents took part either by phone or online. ↩

4. An additional factor which may have affected response rates among particular groups was the coronavirus pandemic. For instance, if it had disproportionately affected response rates among people with more serious health conditions, these people would be underrepresented in the results. In practice, there was not a decline in response rates following the beginning of the pandemic and the representativeness of the survey sample was good on variables such as severity of health conditions and impairment. Any differences in response rate related to these characteristics will also have been adjusted for by the non-response weighting. ↩

5. While we would expect all SCR IW respondents to say they had completed some paid work in the last 12 months because they were in employment when they joined the trial, some will have dropped out of work soon after starting on the trial. Furthermore, the 12-month survey data was collected from some a little after the 12-month time point. 41% of respondents to the final survey completed their interview more than 1 year after they were randomised into the trial. Together, these two reasons explain why the percentage that completed some paid work is not higher. ↩

6. Schwarzer, R. and Jerusalem, M. (2021) The General Self-Efficacy Scale (GSE). Available at: http://userpage.fu-berlin.de/health/engscal.htm (Accessed: 11 November 2021). ↩

7. Scores for the five health states can be converted into a utility index score by using scores from value sets (preference weights) elicited from a general population. ↩

8. Szende, A., Janssen, B., Cabases, J., eds. (2014). Self-Reported Population Health: An International Perspective based on EQ-5D. Springer Open: London. Pg. 46. ↩

9. Short Warwick Edinburgh Mental Wellbeing Scale (SWEMWBS), © NHS Health Scotland, University of Warwick and University of Edinburgh, 2008, all rights reserved ↩

10. Warwick Medical School, Collect, score, analyse and interpret WEMWBS (Accessed 25/11/2021). ↩

11. Office for National Statistics. (2018) Surveys using our four personal well-being questions (Accessed: 12 November 2021). ↩

12. Spitzer RL, Kroenke K, Williams JB et al. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006 166(10):1092-7. ↩

13. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001. 16(9):606-13. ↩

14. Social contact included face-to-face contact as well as contact by phone, electronically (video call, email, social media etc) or by any other means. ↩

15. People in work for the duration of the trial were not asked how their motivation to return to work was affected. ↩