Research and analysis

Work and Health Programme evaluation: synthesis report

Updated 18 January 2024

Applies to England and Wales

DWP research report no. 1044

A report of research carried out by Kantar Public, ICF and BPSR 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 website.

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 October 2023.

ISBN 978-1-78659-579-9

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

Voluntary statement of compliance with the Code of Practice for Statistics

The Code of Practice for Statistics (the Code) is built around three main concepts, or pillars, trustworthiness, quality and value:

• trustworthiness – is about having confidence in the people and organisations that publish statistics
• quality – is about using data and methods that produce assured statistics
• value – is about publishing statistics that support society’s needs for information

The following explains how we have applied the pillars of the Code in a proportionate way.

Trustworthiness

• Kantar Public, an independent social research agency, conducted this research on behalf of DWP.
• Kantar Public are bound by the Market Research Society’s and ESOMAR code of conduct. These codes ensure professional standards are maintained, work is impartial and done in an ethical way.

Quality

• The survey was carried out using established statistical methods.
• This report meets the ISO 9001, ISO 20252 and ISO 27001 international quality standards for market research, and Kantar’s internal quality checking processes.

Value

• This research provides information on the delivery, process and experience of the Work and Health Programme.
• The data contained within this report have informed the ongoing development of policy decisions regarding the running of the Work and Health Programme and the future of employment support programmes for disabled people and key priority groups.

Executive summary

In November 2017 the Government announced a new employment support programme, the Work and Health Programme (WHP), as part of a wide package of support for disabled people^{[footnote 1]}. The WHP aims to help people find permanent work and is available on a voluntary basis for people who are disabled as well as the long-term unemployed or those in the Early Access group. Prior to the introduction of the Restart scheme in 2020, participation was mandatory for Jobseeker’s Allowance or Universal Credit claimants who had been unemployed for two years. As of May 2023, 410,000^{[footnote 2]} people had been referred to the programme.

Between 2018 and March 2020, the WHP was run as a randomised control trial (RCT). Eligible participants who agreed, or were mandated to join the WHP, were then randomly allocated to the programme or to a control group where they continued to receive the package of support they would normally receive via Job Centre Plus (JCP). The WHP was procured and is being managed nationally across six Contract Package Areas (CPAs), with five service providers delivering across these six CPAs in England and Wales. Within each of the national CPAs in England and Wales, a smaller geography has been designated as a Devolved Deal Area (DDA). These are City regions or combined authorities who worked in partnership with DWP to ensure that priority groups within their specific labour markets would benefit from the programme.

WHP delivery in London and Greater Manchester is managed by Local Government Partnerships (LGPs). These local authorities have selected their own providers. In four local authorities, Public Sector Comparators (PSCs) Jobcentre Plus (JCP) offices were funded to provide an equivalent service.

In 2018 Kantar Public and ICF were commissioned to carry out a programme of research as part of the evaluation of the WHP. The evaluation took a mixed-method approach, using quantitative surveys and qualitative, deep dive research with stakeholders, JCP staff, service providers and programme participants. It was conducted across two waves. This report includes findings from the Wave One and Wave Two provider surveys, the Wave One and Wave Two participant surveys and the Wave One and Wave Two deep dive qualitative interviews. It also includes findings from an ad hoc participant survey which was conducted to understand the impact of the COVID-19 pandemic on the implementation of the programme.

Key Findings

Programme set up

In the CPAs, DDAs and LGPs, the WHP was delivered by third-party providers that were commissioned through a competitive tender process. Providers were paid a service delivery fee and outcome-related payments based on employment outcomes.

In each of the WHP areas represented in the deep dive research sample, there was a single ‘prime provider’ leading the overall delivery of WHP support. Most of the prime providers also had a supply chain of subcontracted providers that contributed to the delivery of their WHP support. In the PSCs, senior JCP staff were responsible for determining the design of their local PSC ‘offer’.

Programme design and delivery

Evidence from the provider surveys and deep dives considers how the programme was put together and delivered against its mandate.

Providers considered the programme design to be effective in supporting participants to find work. The programme’s flexible nature, provision of a dedicated contact and long-term assistance facilitated this success. Whilst the pandemic required shifts of delivery strategy, the system proved sufficiently robust over time.

Programme commissioning

Both commissioners and providers felt that the commissioning process had functioned well, increasing quality and value for money of bids and supporting a local focus. The involvement of local authorities in the DDA commissioning was perceived to have encouraged bidders to tailor their offer to the characteristics of the local area. In the majority of cases, delivery was through a prime contractor using a ‘supply chain’ of subcontractors, and, if necessary, supported by an informal network of other suppliers. Over time, the development of these networks and their utilisation adapted to support delivery.

WHP referrals

The referral process was not perceived to be difficult, but appropriate, and willing (voluntary) referral was perceived as supporting successful outcomes. Work coaches reported some inappropriate referrals and that the initial RCT design made introducing the programme to vulnerable participants more difficult.

Initial referral volumes were lower than expected but rose after the first six months. Following the introduction of lockdown restrictions to combat COVID-19, referral volumes fell again. They returned to pre-pandemic levels when lockdown restrictions were lifted.

Referral volumes were improved with increased JCP work coach awareness of the programme and diversified routes to referral.

Clear communication between work coaches and key workers facilitated effective hand over of participants between Jobcentre offices and provider organisations.

Role of key workers

The role of key workers was a key element in successful programme delivery. Under the WHP programme, participants were allocated a dedicated person to focus on them and their needs. Key workers regularly met with participants and provided support for up to 15-months. This allowed time for deeper understanding of complex customer needs and building trust. This type of focussed one-on-one support was seen as a positive condition for success and necessary to help those furthest from the job market by both participants and providers. Additional training for key workers helped them to work effectively with those with complex needs.

Support provided

Types of support provided varied widely from help with skills and job-related tasks, to help with health issues, and financial support for training or practical items. Key workers were able to tailor support to the needs of the participant. They continued to provide support to those who entered work.

Provider key workers were able to refer participants to specialists addressing specific needs delivered through in-house and external teams. In this way, providers were generally able to access sufficient and suitable services to support their participants, though there were challenges providing specialist support in certain areas including health provision, addiction services and bereavement support. A lack of available resource and long waiting lists were noted for mental health issues.

Providers felt that understanding the needs and barriers of the participants at, or before, the initial meeting, was critical to success and that accurate and up-to-date information and communication was necessary to ensure the programme’s overall effectiveness. Above all, having the time to maximise understanding of potentially complex, multi-faceted participant needs and implement appropriate responses was seen to be one of the keys to success.

Remote delivery

The pivot to remote service provision driven by the COVID-19 pandemic, caused disruption to the WHP (as it did in the economy more widely). In some cases, participants felt more comfortable taking part remotely and engaged more fully than they might have done in person. However, participants lacking basic IT skills struggled to engage during periods of remote delivery.

Remote service delivery also meant that specialist support was no longer limited to particular geographic areas and was accessible to a wider cross-section of participants.

Participant characteristics and attitudes to work

Three main groups were included in the participant surveys. The voluntary group included disabled people and certain key priority groups, known as the Early Access Group (EAG). The mandatory group included those who were required to participate as they had claimed Jobseeker’s Allowance or Universal Credit and had reached 24 months of unemployment. Given the difference in composition between the voluntary and mandatory groups, it is not possible to attribute variations in experience or outcomes between them to either the conditionality, or the group composition in isolation. The Public Sector Comparator group included those who were referred to support provided through JCPs in the four local authorities where JCPs were funded to provide an equivalent service. As the number of PSC survey respondents was small (reflecting the relatively small volume of participants in the PSCs) sub-group analysis focussed on differences between voluntary and mandatory participants.

The characteristics of the participant survey respondents were largely reflective of those of the programme as a whole. Characteristics across the three main groups included in the surveys (voluntary, mandatory and PSC) were broadly similar to one another with relatively little difference in terms of gender, age profile, education level and ethnicity.

On balance, participants were more likely to be male than female. They covered the full age range, but with relatively more aged 50-59. Most were educated to Level 2 or below, although three in ten had qualifications at Level 3 or higher. The over-50s and those out of work for 10 years or more were relatively more likely to have lower-level qualifications, and so may have required greater support with basic skills.

The research found that long-term health conditions and disabilities were common among all WHP participants, not just those who were referred to the Disability group. Reflecting this, the most common self-reported barrier to finding work was a physical or mental health condition or disability. Post-pandemic, challenges due to COVID-19, including supporting those who had been self-isolating for a long period or who felt unsafe returning to a workplace, were also reported as a barrier. Long-term health conditions were somewhat higher among voluntary participants driven by those with disabilities being more likely to be in the voluntary, rather than the mandatory group. Perhaps as a consequence of this, voluntary participants tended to be more negative about the state of their health than mandatory participants and negativity about state of health was linked with less positive opinions about employability and success in the workplace.

More than eight in ten participants who took part in the Wave Two survey had been in employment at some point before referral to the programme, with little difference between the voluntary and mandatory referral groups. Physical or mental health was considered the greatest barrier to re-joining the workforce, along with age and long periods away from the workplace. Complex, multi-faceted needs, combining several barriers, were perceived by participants to reduce the chance of finding work and/or restrict the nature of work they felt able to consider.

Evidence of participant outcomes

The survey (particularly Wave Two) and deep dive fieldwork explored the outcomes achieved by participants to date from the participant and Work Coach/key worker perspectives^{[footnote 3]}. The report uses a mixture of impact analysis, descriptive data and additional material collated over time to seek to measure the extent to which the WHP lead to better or worse outcomes than JCP’s ‘business as usual’.

The overall findings indicated a range of positive outcomes for all WHP participants. Impact analysis based on survey responses shows that outcomes were generally more positive for voluntary participants than mandatory participants.

Participant satisfaction

Overall participant satisfaction was high for both voluntary and mandatory participants (more than eight in ten satisfied) with the control group slightly less satisfied with the business-as-usual JCP support they received. Participants identified several positive aspects of the programme including the positive, friendly delivery environment, flexibility of delivery, programme duration and additional support provided over and above the focus on finding work.

In addition, participants were generally satisfied with the methods and frequency of contact with key workers as well as the extent to which support was tailored to the individual. The pandemic appeared to have little impact on overall satisfaction levels with only a minority expressing dissatisfaction with the service provided during the pandemic.

Employment outcomes

Voluntary WHP participants were statistically significantly more likely than their control group to have done some work since starting the programme (27% compared to 22%)^{[footnote 4]} and to be in work at the Wave Two survey (19% compared to 16%)^{[footnote 5]}. Voluntary participants also reported longer hours than mandatory participants, with more employed full time compared to mandatory participants. Mandatory participants were slightly more likely to have finished participation and then been referred back to JCP support than voluntary participants. While there were also positive differences between the mandatory WHP participants and their control group, these did not reach statistical significance.

Both mandatory and voluntary participants who were in work at the time of interview, said that the support they had received through the WHP had helped them to get their current job. Voluntary WHP participants who were in work at the Wave Two survey were far more likely than their control group to feel that the support they received had helped them get that work (68% compared to 48%).

Employment outcomes varied by demographics: men were slightly more likely to be employed full time and women part time, people over 50 less likely to report having been employed at all compared with younger people, those with lower-level qualifications less likely to be employed than those with higher qualifications and those with no health conditions more likely to be employed or seeking employment than those with health conditions. These sub-group differences were consistent across both the treatment and the control groups and can’t be attributed to the WHP. These differences are broadly consistent with broader labour market trends.

Improved employability / increased likelihood of moving into work

The WHP appeared to have a positive impact on participants’ levels of job search activity compared to business-as-usual JCP support. In both voluntary and mandatory groups, one in five reported applying for more jobs than before (indicating greater activity than their control groups). Among both voluntary and mandatory participants, six in ten also thought the support received had increased their chances of moving into work.

Whilst voluntary and mandatory participants were equally positive that the support had increased their chances of moving into work, only voluntary WHP participants were more positive compared with those in the control group.

Skills, confidence and motivation

Among both voluntary and mandatory participants, there was no statistically significant evidence to suggest that the WHP had an additional impact on participants’ interpersonal or job application skills compared to business-as-usual JCP support. However, there was evidence that the WHP had a positive impact on self-reported confidence in finding a job and increased motivation to find work among voluntary participants.

PSC participants

Like other groups, PSC participants were positive about the support received and agreed it had increased their chances of moving into work, though this group were less likely to be able to access some of the more specialist support offered through the contracted WHP. At the time of the Wave 2 survey, 23% had started work at any time since being referred to the programme, and 14% were currently employed. This was consistent with the proportion of WHP participants who were in work at the time of the survey or who had been employed since being referred to the programme.

Glossary

Glossary term	Definition
Allocation Group	Allocation of participants to either the WHP, PSC or control group.
Contract Package Area (CPA)	WHP is delivered by five providers across six areas known as Contract Package Areas. These areas were the first to deliver the WHP in the phased rollout.
Control Group (also see Randomised Control Trial)	A group similar to the ‘treatment group’ (in this case, those receiving business as usual support from Jobcentre Plus), used to compare how an intervention works.
Devolution Deal Area (DDA)	Areas or city regions that have Devolution Deals and were involved in co-designing the WHP in their area, including being involved in the evaluation of provider bids and receiving ongoing performance management information on delivery in their DDA area.
Disability Employment Advisor (DEA)	DEAs support Work Coaches by providing expert knowledge on how to support disabled customers, influencing employer engagement in local communities, and by directly supporting customers with a health condition or disability where additional support can benefit the customer.
Early Access Group (EAG)	The early access group is voluntary and aimed at DWP customers who may need additional support to move into employment and are in one of a number of priority groups.
Disability group	Disability participant group. Disabled people as defined by the Equality Act 2010 and who have been identified as being capable of finding work within a year of starting the programme.
Eligibility group	Which WHP group participants have been allocated to (Disability group, EAG, LTU).
Jobcentre Plus (JCP)	Jobcentre Plus is a government-funded employment organisation and social security office that can be found in most localities, whose aim is to help people of working age find employment in the UK.
JCP Gatekeeper	Gatekeepers manage allocations to the WHP, PSC and control group and administer the random allocation process.
Key worker	Key workers are employed by service providers. They advise participants once they join the programme.
Local Government Partnerships (LGPs)	These local authorities are allocated funding under devolution deals via a quarterly grant. LGPs commission and contract manage their own providers.
Local partners	Not to be confused with LGPs. WHP providers form partnerships with local partners to deliver WHP in their areas. Local partners may include charities, health and wellbeing professionals, employers and other organisations providing a range of support designed to help people into work.
Long Term Unemployed (LTU)	The LTU group on the Work and Health Programme consists of claimants receiving Universal Credit Intensive Work Search or Job Seekers’ Allowance (JSA) who have been unemployed for 24 months+.
Participants	Individuals who have received support through the programme.
Providers/ prime providers	Contracted providers delivering the WHP. One prime provider per CPA, although may also deliver through sub-contractors.
Public Sector Comparator (PSC)	In four local authorities JCPs were provided with additional funding to run an equivalent programme of additional support for eligible participants themselves. The PSC support is similar to the WHP.
Randomised Control Trial (RCT)	A study in which a number of similar people are randomly assigned to two or more groups to test an intervention.
Work Coach	JCP adviser who works with DWP customers to support them into work. This includes, where appropriate, Work Coaches referring participants to employment programmes including the WHP.

1. Introduction

1.1. Background

The Work and Health Programme was first announced in the 2015 Spending Review as an employment support programme succeeding the Work Programme and Work Choice schemes, as part of a wide support package for disabled people^{[footnote 6]}. In 2017, the Government published ‘Improving Lives: The Work, Health and Disability Green Paper, which confirmed the Work and Health Programme would ‘offer a more personalised, local approach to supporting disabled people to overcome barriers to employment’^{[footnote 7]}. The WHP aimed to support people to find and keep a job and was available to the following three eligibility groups:

1. Disability group. WHP participation was voluntary for disabled people as defined in the Equality Act (2010).

2. Early Access Group (EAG). The EAG was aimed at people who might need support to move into employment and were in key priority groups (for example homeless, ex-armed forces, care leavers, refugees). EAG participants took part on a voluntary basis.

3. Long-term Unemployed group (LTU). Participation was mandatory for Jobseeker’s Allowance (JSA) or Universal Credit (UC) claimants who had been unemployed for 24 months or longer.

1.1.1. Programme outline

The WHP was rolled out in England and Wales between November 2017 and April 2018. It was procured and was managed nationally across six Contract Package Areas (CPAs), with five service providers delivering across these six CPAs in England and Wales.

Devolved delivery of the WHP

Within each of the national CPAs in England and Wales, a smaller geography was designated as a Devolved Deal Area (DDA). These were city regions or combined authorities who worked in partnership with DWP to ensure that priority groups within their specific labour markets would benefit from the programme. They retained an interest and influence in the ongoing monitoring of the performance of the programme in their areas.

WHP delivery in London and Greater Manchester was procured and managed by Local Government Partnerships (LGPs). These local authorities were allocated funding (paid as a quarterly grant) under devolution deals and procured their own providers. Provision in Manchester was launched in January 2018. The London programme launched between January and March 2018 and was delivered on a sub-regional basis with groups of boroughs working together through four partnerships.

Referral to the WHP

Potential participants were identified by Jobcentre Plus (JCP) Work Coaches who explained the programme to them. Eligible participants who agreed, or were mandated, to join WHP, were then allocated to the programme on a random basis. To facilitate the impact analysis, a proportion of participants were allocated randomly to a control group and continued to receive the package of support they would normally receive via JCP. The random control design aimed to allow for comparison of outcomes and experiences of those participating in the WHP, those participating in the PSC, and those in the control group. This was not possible in all cases due to changes to services and research design as a result of the COVID-19 pandemic.

Work Coaches referred participants who were randomly selected to join WHP to the local WHP provider, for them to make initial contact with the participant^{[footnote 8]}. The original programme design anticipated that 270,000 people would be referred to the WHP programme, 5,300 would be assigned to the PSC and 60,000 assigned to the control group.

WHP providers

Participants were assigned to a WHP provider key worker who worked with them for up to 15 months to provide appropriate support. During the first couple of appointments, providers assessed participants’ needs in a one-to-one discussion. These may have been guided by the use of a system-based diagnostic tool or an online resource, depending on the provider. These diagnostics determined what the participant’s needs were and which needed to be addressed most urgently. Key workers also helped the participant create an action plan with activities that would help them to gain stable employment^{[footnote 9]}. During this time claimants still needed to fulfil their benefit requirements with JCP and were required to attend Work Focused Interviews where Work Coaches could see how the participant was progressing.

Providers were responsible for overall contract management and the end-to-end delivery of non-specialist WHP support and partnered with sub-contractors to provide specialist services. Providers were paid a service delivery fee and outcome-related payments (known as payment by result) when a person reached either a specified level of earnings once in employment, or six months of self-employment. The earnings threshold for a job outcome for the National WHP and the majority of LGPs was 16 hours per week for 26 weeks at the National Living Wage, however the threshold for the West London Alliance was the same number of hours but at the London Living Wage and the Greater Manchester Combined Authority at the Real Living Wage.

If a participant transitioned to work, the providers continued to provide in-work support for up to six months.

Public Sector Comparator

In addition to the WHP, DWP established four Public Sector Comparators (PSC). In these areas JCPs were provided with additional funding and resources to themselves run an equivalent programme of additional support for eligible participants.

1.1.2. Adaptations due to COVID-19

Due to increased delivery pressures in Jobcentre Plus and changes to priorities as a result of COVID-19, referrals to the PSC and the Randomised Control Trial allocation groups were paused in March 2020. All participants referred to the WHP since then were referred to the programme.

During the first UK-wide lockdown, face-to-face delivery of the WHP was suspended. Instead, providers employed alternative routes to service delivery including using digital options and providing a telephone service to ensure support continued. Support provided was broadened to cover general wellbeing and those referred to the LTU group were no longer sanctioned if they chose not to participate.

To protect the capacity and capability necessary to support the expected increase in unemployment that arose from COVID-19, the contracts between DWP and Providers were temporarily transitioned to a cost-plus basis^{[footnote 10]}. It was planned that this would revert to payment by result once the economy and operating environment had stabilised. All Providers managed under DWP contracts transitioned back to payment by results by December 2021. LGPs were responsible for managing the transition of payment models for their own contracts with Providers.

1.1.3. Early outcomes

As of May 2023, 410,000 individuals had been referred to the WHP with 280,000 having started on the programme. Of the people referred, 316,000, were in the Disability group. Around 34,000 participants were Long-term Unemployed with 57,000 participants in the Early Access Group^{[footnote 11]}.

68% of individuals referred by May 2023 started on the programme.

Of those participants starting on the programme in May 2021 or earlier, 45% achieved the first earnings threshold within 24 months, and 30% achieved a job outcome within 24 months^{[footnote 12]}.

1.2. Evaluation aims

• To provide rich, contextual data on how the programme operated to help people into work and understand why outcomes were achieved / not achieved.
• To provide understanding of the WHP processes in order to improve live running and inform future programme design.
• To provide evidence on new features of the WHP not used in other employment support programmes or used on a different scale or with different groups.
• To contribute to the impact evaluation by providing data and analysis of employment and health outcomes and distance travelled towards employment. Also, to support the impact evaluation by developing a picture of the type of support individuals in each group receive.
• To provide evidence on differences between subgroups of interest, in particular by participant type (voluntary Disability group and EAG participants and mandatory LTU participants), and, in order to measure impact more accurately, programme versus control group, (where sample sizes allow).

1.3. Methodology

1.3.1. Original design

The evaluation, which was conducted across two waves, took a mixed-method approach, using quantitative surveys and qualitative, deep dive research (see Figure 1.1 at the end of this chapter).

Quantitative surveys

The quantitative work included two strands:

Strand 1: Participant Surveys

The first wave of the participant survey was conducted with 2,982 participants between February 2019 and March 2020. This included participants who were referred to the WHP between November 2018 and December 2019. Interviews were conducted via Computer-Assisted Telephone Interviewing.

Participants were sampled in monthly batches according to the date they were referred to the programme, aiming to carry out interviews around three months after the initial referral. The early batches of sample (those interviewed up to around May 2019) were slightly larger than subsequent batches, with increased numbers of core WHP participants (excluding those in LGP areas) and control cases. This was to maximise the chance of achieving the target number of interviews in these groups and to guard against any later shortfalls in referral numbers^{[footnote 13]}.

Wave Two participant survey interviews were conducted with 5,655 participants between September 2020 and June 2021. Participants included 1,144 who had completed the Wave One survey as well as 4,511 participants who had not participated in previous waves of the fieldwork.

The participant surveys had four aims:

1. To collect information on participant outcomes and other data not available through DWP administrative data. This included demographics, type of disability or health condition, health and wellbeing outcomes, self-employment earnings, and ‘distance travelled’ (progress made) towards employment.

2. To compare health and employment outcomes between a number of subgroups (for example, WHP participant / PSC participant / control group).

3. To provide evidence on elements of the WHP which were new or there was little evidence on: referral process, service integration and partnership working, devolution and localism, and the PSC.

4. To provide evidence on participant experience and support received by each allocation group (WHP, PSC or control) and referral (Disability group, EAG or LTU). The majority of participant data charts show data from the Wave Two survey which provide the most up to date information about participants’ outcomes and experiences (including employment outcomes). A small number of references to the Wave One survey data are made where relevant. The data source is specified on each chart.

Strand 2: Provider Surveys

The provider surveys aimed to provide insights on how the WHP was delivered on the ground by key workers, to understand what support was offered and how it was tailored to participants’ needs. They also aimed to assess provider attitudes towards the programme and how they thought it was working for participants.

The first wave was completed online between April and July 2019 by 272 key workers who were employed by WHP providers.

The second wave was completed online by 394 key workers between July and September 2020. Because the total number of completed interviews for the provider surveys was relatively low, comparisons between the two waves should be treated as indicative.

Qualitative deep dives

The deep dives provided qualitative insights and evidence on the delivery of the WHP in different areas, exploring challenges and what was working well, and differences between areas and commissioning approaches.

Two waves of deep dive research were conducted over the course of the evaluation: the first between March-December 2020 and the second between September-December 2021.

The first wave was conducted in a sample of 12 areas, comprised of:

• Three of the six Contract Package Areas (CPAs)
• Three of the six Devolution Deal Areas (DDAs)
• Two of the five Local Government Partners (LGPs)
• All four of the Public Sector Comparators (PSCs).

The sample was selected to represent the different commissioning approaches, providers, and supply chain models evident across the areas in which the WHP was being delivered.

The smaller-scale second wave was conducted in the same sample of CPAs, DDAs and LGPs as in the first wave but did not include the (by then discontinued) PSCs.

Within each deep dive and in each wave, interviews were conducted with local JCP staff (including district leads, JCP managers, Work Coaches and DEAs), provider staff (including senior managers and key workers), stakeholders (local authorities and other public or third sector organisations), and participants (individuals who had received support through the programme). Table 1 shows the number of interviews conducted in each wave.

Table 1: Overview of the deep dive research: number of interviews

Wave	JCP staff	Provider staff	Participants	Stakeholders
Wave One	98	95	113	20
Wave Two	37	59	81	10

Alongside the deep dive fieldwork in each area, DWP staff who had been involved in the commissioning of WHP provision were also consulted in Wave One to gain their reflections on the process.

Impact analysis

The impact of the WHP was estimated by comparing the outcomes of those offered support through the WHP (‘WHP participants’) against the outcomes of those randomly allocated to the standard offer of support provided by JCP (‘control group’). The analysis measured the extent to which the WHP led to better or worse outcomes than JCP’s ‘business as usual’. The impact estimates were based on the Wave Two survey data collected from WHP participants and the control group 18-24 months after the point of allocation.

It is reasonable to anticipate that the impacts of the WHP may differ between those entering the programme via different routes. Disability group participants and the Early Access Group (EAG) voluntarily used the service, while those who were long-term unemployed for 24+ months were mandated to do so. The control group included participants from the disability group and EAG and long term unemployed. Separate impact estimates were therefore produced for ‘voluntary’ and ‘mandatory’ participants.

As the EAG made up only –10% of the WHP sample, the number of EAG participants in the survey was too small to produce comprehensive, robust impact estimates. However, there were a small number of areas where the impact of the WHP on Disability group participants differed so markedly from the EAG group as to meet the criteria for statistical significance (such cases are noted in a footnote).

Excluding those with missing data on key variables^{[footnote 14]}, the analysis is based on the numbers of interviews shown in Table 2:

Table 2: Sample structure for impact analysis

Groups	Voluntary	Disability	Early Access Group	Mandatory (LTU)
WHP participants	2,037	1,691	346	1,348
Control group	1,600	1,538	62	354

Analysis has also been produced that looks at the impacts of WHP for particular sub-groups within the voluntary and mandatory groups, looking by gender and across age, health conditions and length of unemployment.

Estimates of impact that draw on other sources, such as administrative data, or relate to other time periods, could produce different results. The merits of different estimates should be considered when making conclusions about the impact of WHP.

1.3.2. Adaptations due to COVID-19

Quantitative surveys

Fieldwork for the second wave of the participant survey was scheduled to interview participants around 18 months after they were referred to the programme. However, fieldwork was paused on all strands of the evaluation following the introduction of national restrictions to combat the COVID-19 pandemic.

As a result, the start of the Wave Two participant survey was delayed from May 2020 to September 2020. This meant that around half of the survey respondents participated more than 18 months after they were first referred to the programme. The survey questions were adapted to include COVID-19 appropriate response options and some pandemic specific questions were included.

To understand the impact that COVID-19, and the resulting restrictions, had on the programme, the evaluation was redesigned to include a third ad hoc survey strand. This survey aimed to understand:

• If and how participants had received support since March 2020
• The frequency of meetings with their key worker
• How meetings between participants and key workers took place
• Participants’ outlook on the job market at the point of interview
• The impact of COVID-19 on their employment situation
• Whether there had been a change in wellbeing in comparison to Wave One responses.

This ad hoc COVID-19 survey was conducted via telephone with 300 participants on the WHP programme between September and November 2020. Participants were sampled from those who were referred to the WHP between September 2019 and March 2020. The sample used the same CPAs as the Wave One survey sample, however neither PSC nor control participants were included. The survey was administered between six and 12 months after referral. None of the ad hoc COVID-19 survey participants took part in the main Wave One or Wave Two participant surveys. Respondents from this survey will be referred to as ‘ad-hoc participants’ on relevant charts.

The Wave Two provider survey was not delayed by the COVID-19 restrictions; however, the questionnaire was adapted to gain an understanding of key workers’ perceptions of how the programme had been affected by the pandemic.

Qualitative deep dives

The fieldwork for the Wave One deep dive research initially started in March 2020, was paused due to the pandemic in April 2020, then restarted from August 2020 and was completed in December 2020.

This meant some discontinuity and elapsed time between when different respondents in the deep dive areas were interviewed. A small number of JCP offices did not feel able to participate in the research after it restarted in August due to ongoing workload pressures linked to the pandemic. This did not significantly impact on the overall numbers of interviews conducted across the deep dives, and the elongated timeframe over which research was conducted provided scope to explore how the pandemic had impacted on WHP delivery over time.

1.4. Caveats and limitations

This report outlines findings from qualitative and quantitative interviews with programme providers and participants. As such the outcomes were based on the perceptions of interviewees which might differ from the outcomes recorded in administrative data. DWP is undertaking additional work to evaluate the Work and Health Programme using administrative data.

Differences in findings between this report and other studies might also be driven by differences in the population included, and the time at which they experienced support on the WHP. The quantitative work conducted by Kantar Public only included a sample of participants recruited to the programme between November 2018 and January 2020.

1.5. Understanding the data

Three main groups were included in the participant surveys. The voluntary group included disabled people and certain key priority groups, known as the Early Access Group (EAG). The mandatory group included those who were required to participate as they had claimed Jobseeker’s Allowance or Universal Credit and had reached 24 months of unemployment. The Public Sector Comparator group included those who were referred to support provided through one of the four local authorities funded to provide an equivalent service.

Unless otherwise stated, all survey sub-group analysis focused on differences that were statistically significant at a 95 per cent confidence level. This means you would only expect to see the result caused by chance one in 20 times.

As the EAG base sizes were typically too small to include in sub-group analysis, reporting concentrates on the differences between voluntary and mandatory participants. Statistically significant differences between the EAG and Disability groups have been included where appropriate base sizes were available.

figure 1.1 Evaluation fieldwork timeline

Year	Participant surveys: wave 1	Participant surveys: wave 2	Provider surveys: wave 1	Provider surveys: wave 2	Deep dives: wave 1	Deep dives: wave 2
2018	Questionnaire design and Cognitive testing	–	Questionnaire design	–	–	–
2019	Pilot and Survey fieldwork	Questionnaire design and cognitive testing	Cognitive testing and survey fieldwork 1, 2 and 3	–	–	–
2020	Survey fieldwork continued from 2019	Pilot, fieldwork delayed due COVID-19 restrictions and survey fieldwork	–	Questionnaire design and survey fieldwork	Topic guide design and fieldwork (extended due to COVID-19 restrictions)	–
2021	–	Survey fieldwork continued from 2020	–	–	–	Topic guide and fieldwork

2. Programme design and delivery model

Summary

• The volume of referrals from JCP to the WHP varied throughout the programme. Factors which helped to improve the number of referrals included:
  • Reminders and training about WHP for Work Coaches
  • Removal of randomisation
  • Increased referrals through alternative routes
• Key workers felt that the quality of the information provided at referral was not always accurate or up to date. A large majority said that they did not receive enough information about participants to understand their needs and barriers to work before their first appointment. Improved communication between providers and JCP, including face-to-face meetings and updates on participant progress, could help key workers to feel prepared when meeting with participants for the first time.
• By necessity, there was a wholesale transition from face-to-face programme delivery to remote contact following the onset of the pandemic. Provider staff and participants felt that this transition was successful.
• Providers considered the programme to be an effective way of supporting participants to find work. They reported that it was effective due to its flexibility, the longevity of the support offered, its voluntary nature and the dedicated support of a key worker.

Evidence in this chapter was taken primarily from the deep dive research among providers and from the Wave One and Wave Two provider surveys, supplemented by data from the ad-hoc participant survey to help understand the impact of COVID-19 on delivery.

2.1. Programme commissioning

The WHP was delivered by commissioned third-party providers in different areas of England and Wales, designated as either a Contract Package Area (CPA), Devolution Deal Area (DDA) or Local Government Partnership (LGP). Prior to the pandemic, equivalent support was also being delivered in-house by JCP in four Public Sector Comparator (PSC) areas.

DWP was responsible for commissioning the WHP providers for CPA areas. Within CPA areas, the WHP was designed in consultation with representatives from local authorities in the local areas or city regions. Responsibility for commissioning in LGPs was devolved to the local combined authorities in Greater Manchester and London.

Overall reflections on the commissioning process were very positive in the qualitative deep dives. DWP and the LGPs reportedly implemented a similar, iterative approach. This included a substantial commercial dialogue element, where initial written proposals submitted by bidders were discussed, challenged, and refined through face-to-face sessions.

Commissioners felt this process had been effective in increasing the quality and value for money of the bids received and helped to raise the performance targets that bidders committed to deliver if selected. Providers felt it provided a better forum for them to demonstrate their capabilities than more conventional paper-based commissioning and had pushed them to refine and improve their initial bid.

‘It allowed us to show that we knew what we were talking about.’ Provider manager

The involvement of local authorities in the DDA commissioning (and even more so the devolution of the LGP commissioning to the combined authorities) was perceived to have promoted localism and encouraged bidders to partly tailor their offer to the characteristics of the relevant area.

‘It made us think about our footprint in certain areas. We hadn’t done much in [the DDA area] for a number of years, so it encouraged us to learn about the local area.’ Provider manager

The only reservations expressed about the commissioning came from a few of the providers concerned after programme delivery had started. From the commissioners’ perspective it was perceived that, with the benefit of hindsight, the commissioning process had encouraged them to set over-ambitious performance targets in their bids (which they had subsequently struggled to meet).

Some also queried the benefits of designating contract areas the size of the CPAs. CPAs were the largest type of WHP contract area and contained both urban centres and dispersed rural localities. The rationale for this was to avoid the potential inefficiencies (for both commissioners and providers) associated with having too many small contract areas. However, some CPA providers reported challenges in providing geographical coverage and operationalising the localism ethos of the programme across such large and diverse territories.

2.1.1. Contractual structures

In each of the areas represented in the deep dive research sample, there was a single ‘prime provider’ that led on the overall delivery of WHP support. Most, but not all, prime providers also had a supply chain of subcontracted providers that contributed to the delivery of their WHP support in the area.

Prime providers were responsible for overall contract management and the end-to-end delivery of non-specialist WHP support (regular ongoing support for participants by a key worker, and other centralised functions such as work placements and in-work support) across most, or all, of the area.

Some subcontractors were responsible for the end-to-end delivery of non-specialist WHP support in parts of the contract area not covered by the prime provider. Other subcontractors were responsible for delivering more specialist WHP support (to help participants overcome specific barriers to work and/or tailored to specific participant types, such as those with a visual impairment) that key workers could refer participants on to.

The exception to this supply chain model was in one of the CPA areas in the deep dive research sample. Here one prime provider was responsible for the end-to-end delivery of non-specialist support across the contract area. There were no formal subcontracted providers. Instead, the prime provider established a community network partnership and invited specialist providers to join this on a non-contractual basis. Providers registered on the partnership were reported to offer a similar range of specialist support to the subcontracted specialist providers in other contracted WHP areas. Key workers could refer participants onto any provider in the partnership. Specialist providers were either paid on a cost basis for each participant referred or they provided support using other public funding they already receive.

This suggested two very different models, but in practice there was evidence of some convergence between them once programme delivery had started. Prime providers who had adopted the ‘prime+subs’ model said over time they had built up databases or informal frameworks that included their subcontracted specialist providers and other local organisations that could potentially take on referrals at no charge or that the prime provider could remunerate on a spot-purchasing cost basis.

2.2. Referral process

2.2.1. How participants are referred

The original design of the programme was based on individuals being referred through two main routes: by solely JCP Work Coaches; and signposting to JCP Work Coaches by other local organisations. Up to the start of the pandemic almost all referrals onto the programme were made by JCP Work Coaches, but this changed subsequently, with an increasing number of referrals through other routes.

Referrals by JCP Work Coaches

The process for identifying potential beneficiaries of the programme and making referrals was designed by DWP to support the randomised control trial (RCT) design.

When Work Coaches identified people on their caseload in the EAG and Disability groups they thought could benefit from the programme, they were first required to introduce the possibility of being referred onto WHP to the claimant. If the claimant voluntarily agreed to potentially being referred onto the programme, the work coach recorded their characteristics against set criteria on their internal IT system. This information was reviewed by a designated JCP ‘gatekeeper’ and individuals deemed eligible were randomly allocated to either a control group who received business as usual support or a treatment group who were referred to WHP support. The aim of this was to compare the two groups and test the impact of the WHP. The only variation in the design of the process was for LTU claimants, for whom referrals were mandatory.

For individuals randomly or mandatorily allocated to receive support through the programme, information about them was passed on to the local WHP contracted provider.

This was how the process operated up to the start of the pandemic in March 2020. At that point the randomisation element was suspended, and all potential beneficiaries identified by Work Coaches were subsequently referred onto the programme. From the start of the pandemic, participation in the programme by LTU claimants was also changed from being mandatory to voluntary.

Referrals through other routes

From the start of the programme there was provision for other local organisations to signpost individuals to Work Coaches to be considered for referral to the programme. Once a local organisation identified someone they thought could benefit from the programme, they could forward their details to JCP or advise the individual to ‘self-refer’ themselves at a local JCP office. This was so that JCP staff could complete basic eligibility checks, randomly allocate individuals to the treatment or control group^{[footnote 15]} and make referrals to the local WHP provider.

As an interim response to the pandemic and the significant drop off in JCP referrals, DWP also made provision for WHP providers to engage with potential participants directly, checking eligibility and submitting a form to a centralised DWP processing team to conduct some limited checks.

2.2.2. Volume of referrals made

Figure 2.1 shows the total number of referrals made to the programme each month since its start in November 2018 to May 2023 when the latest data was available.

Figure 2.1: Number of referrals to the WHP, by month

Source: Work and Health Programme statistics to May 2023

Pre-pandemic referral volumes

The deep dive research indicated almost all the referrals up to the onset of the pandemic were ones made by JCP Work Coaches. Provider staff interviewed in the areas in which the deep dive research was conducted also reported that the volume of referrals in the first six months of the programme was generally lower than they had expected, and that even beyond this point they continued to be quite variable between different JCP offices in their area and from one month to the next.

The main reasons identified in the research for this were:

• Attitudes to randomisation. There were misgivings with the RCT design of the programme across Work Coaches interviewed in the Wave One deep dive research. They said they felt that all potential beneficiaries should be able to benefit from the programme rather than only a randomly determined proportion. They also disliked the requirement to introduce the possibility of being referred onto the programme but not being able to guarantee to individuals that they would be referred. This, they said, made introducing the programme more difficult and they expressed concerns about how potentially vulnerable individuals could respond to not being allocated onto it. Similar concerns also reportedly contributed to very few referrals being made to WHP by local organisations in the pre-pandemic period.
• Work Coach awareness and consideration of WHP. Work Coaches reported they had received information about the programme from the local WHP provider and through briefings from their JCP management at the start of the programme. However, Work Coaches highlighted that WHP was just one amongst a range of local and national provision they could consider referring individuals to.

‘There is so much thrown at you, you don’t know where you are, so what you refer to depends on what is on your mind – if someone is prodding me, ‘don’t forget WHP’, I am likely to make a referral.’ Work Coach

Large caseloads and short appointments with claimants were flagged as underlying constraints on Work Coach time, and their ability to make informed referral decisions. In the pre-pandemic period, JCP managers said they had made periodic attempts to encourage their Work Coaches to make more referrals to the programme. This was reportedly effective in increasing referrals in the weeks immediately afterwards, but then often fell back again, possibly explaining the peaks and troughs in referrals from JCP offices reported by WHP providers.

• Communications between JCP offices and local WHP providers. Provider staff in some of the areas in which the deep dive research was conducted, had an ongoing presence in the local JCP offices prior to the pandemic. Both sides perceived benefits to this. It kept WHP in Work Coaches’ minds and allowed them to do a ‘warm handover’ with participants to the provider staff onsite. In other cases, provider staff said they regularly attended JCP team meetings. However, examples of this type were not universal across the areas in which the deep dive research was conducted.

Feedback from WHP providers about participants after the point of referral, was described as minimal. It was widely thought that more feedback, for example in the form of ‘good news stories’ about individual participants, would motivate Work Coaches to make referrals and help them do this more effectively.

‘People like to know what’s in it for them. We need more feedback to be able to sell it.’ Work Coach

In terms of good practice, the WHP provider in one of the areas in which the deep dive research was conducted had decided to send individual JCP offices weekly figures on the outcomes achieved by individuals referred to them. This was perceived to be helpful in illustrating the value of the provision to Work Coaches.

Referral volumes from the start of the pandemic onwards

The sharp decline in WHP referrals at the onset of the pandemic was a consequence of the unprecedented number of new claims JCP received in this period, and the need to focus Work Coach time on processing these. A few different factors appear to have contributed to the increasing volumes of referrals after this point:

• Backlogs and increased inflows of potential beneficiaries. Once the initial peak of new claims to process had passed, Work Coaches had a substantial accumulation of pre-existing and new individuals on their caseload who could potentially benefit from WHP, and more time to consider referral options for them.
• Removal of randomisation. This was viewed positively by Work Coaches and made them reportedly more comfortable with considering and making referrals to the programme. In theory it also removed the main previously reported barrier to local organisations referring into the programme. However, in Wave Two deep dive research, awareness of the removal of randomisation appeared to be low amongst local organisations (who were reportedly heavily focused on managing their own organisational responses to the pandemic).
• Increased referrals through alternative routes. The deep dive evidence indicated there were higher rates of referrals at Wave Two through alternative routes than prior to the pandemic. Wave Two of the provider survey found that 49% of key workers had one or more participants on their caseload who had been referred through an alternative route. Despite the upward trajectory in WHP referral volumes after the initial onset of the pandemic, there were also factors that were reportedly acting as a constraint (without which volumes were likely to have been even higher):
• The removal of mandation for LTU claimants. Prior to the onset of the pandemic, LTU referrals accounted for 18% of all referrals to the programme, but in November 2021 they represented only 5% of referrals^{[footnote 16]}. Once mandation had been removed in response to the pandemic, Work Coaches could still make LTU referrals on a voluntary basis. However, Work Coaches interviewed in the Wave Two deep dives indicated they were not routinely attempting this because they considered most LTU individuals unlikely to voluntarily agree to being referred.
• Interaction with other DWP programmes. As part of its comprehensive response to the pandemic, DWP introduced three new programmes (JETS, Kickstart and Restart) to provide additional employment support. Several interviewees perceived an overlap, or at least a grey area, between WHP and Restart. Both were based on regular engagement with a key worker and access to more specialist support over a long period (Restart 12 months; WHP 15 months). It was also reported in a few areas that local Restart providers had built in a health component to the support that could be provided to participants. This was perceived to have further blurred distinctions between Restart and WHP.

In addition, Restart was reportedly being heavily promoted to Work Coaches as a referral option by JCP managers and local providers.

‘It’s about what is flavour of the month from up above… if there is any grey area the Work Coach will choose Restart.’ Disability Employment Advisor

There was, consequently, a belief amongst many JCP and provider staff that some claimants who would have potentially benefited from WHP support were instead being referred to Restart.

• Intake of new Work Coaches. The pandemic necessitated JCP significantly increasing its headcount of Work Coaches. One implication of this was that referral decisions were often made by Work Coaches who had not been in post when information dissemination and promotion of WHP took place, before the pandemic.

New Work Coaches interviewed in the deep dive research said their initial training had not touched on what referral options were potentially at their disposal, with the expectation being that they would develop this knowledge on the job (i.e., through internal comms, information from providers, and conversations with colleagues – something that had been constrained during periods when staff had been working remotely rather than together in the JCP office environment). This was thought to have resulted in low levels of awareness and understanding of WHP amongst newer Work Coaches, and consequently fewer referrals.

• Ongoing communications between JCP staff and local WHP providers. These communications had reportedly been severely constrained by the pandemic, as JCP staff focused on new claims and providers focused on adapting their provision to online delivery. The pandemic also created practical barriers to provider staff being present in JCP offices. At the time of the Wave Two interviews in late 2021, some providers talked about making recent renewed efforts with local JCP offices (e.g., through virtual presentations) to re-promote WHP as a referral option.

2.2.3. The appropriateness of referrals

In the Wave One deep dives, provider staff reported receiving referrals for individuals who they didn’t perceive to meet the eligibility criteria for the programme, and as a consequence, could do little to engage or help. The Wave Two provider survey asked key workers if any participants on their caseload met specific criteria making them ineligible for the programme (see Figure 2.2). The primary reason (for 59%) was that some participants were not eligible because they were unable to achieve the goal of finding employment within one year. Being in paid work or taking part in other schemes were secondary reasons, as well as failing on basic EAG, Disability group or LTU criteria.

Figure 2.2: Whether key workers’ caseloads included any participants ineligible for the following reasons (Wave Two provider survey)

Reason	Percentage
Unable to achieve the goal of finding employment within 1 year	59%
Already in paid work	34%
Participating in another DWP employment support programme	27%
Didn’t have a disability or health condition, not in EAG and not unemployed for 24 months	23%
Had previously participated in the work and health programme	14%

(Q_Suitable) Have you had any participants who were not eligible for the Work and Health programme because….

Base: All Provider Wave Two respondents (394)

The main factors influencing the perceived appropriateness of referrals across both deep dive waves overlapped with those influencing referral volumes in several key areas:

• Work Coach understanding of WHP. Provider staff and other JCP staff such as DEAs perceived issues with the appropriateness of referrals partly stemmed from an imprecise or inaccurate understanding of the programme amongst some Work Coaches (particularly newly recruited coaches (Wave Two)).
• Communications between JCP staff and local WHP providers. Where these communications were good, it was generally perceived to aid Work Coach understanding of the programme and help reduce the proportion of inappropriate referrals being made.

‘The key ingredient is relationships between us and JCP. If you’ve got that, everything else falls into place.’ Provider manager

WHP providers were asked not to contact jobcentres for a period at the start of the pandemic due to the resourcing pressures they were under as they responded to unprecedented numbers of new claims. However, from late 2021 onwards, providers were reportedly starting to re-engage with JCP staff. Some reported giving presentations at local JCP offices combing information on a variety of programmes they were delivering. This was perceived to be effective in conveying to Work Coaches the differences between each programme (including WHP) and the claimant types each was best suited to. There was also support from JCP interviewees to resource the continuation of this practice.

• Disability Employment Advisor (DEA) involvement in the referral process. DEAs were reported to be positively supporting the WHP referral process, to varying degrees, in all the areas in which the deep dive research was conducted in Wave One. This was also the case in Wave Two and there was additional evidence of the added value of more intensive DEA involvement. In some areas, local DEAs had taken the lead in maintaining links between JCP and local provider staff during the pandemic (at a time when most other JCP staff were being diverted to other priorities) and assumed responsibility for judging eligibility and making referrals to the programme. Given the skillset of DEAs and the pressures on Work Coaches, this was perceived to be an effective approach that had a positive impact on the appropriateness of referrals to WHP locally.
• Increased referrals through alternative routes. Provider staff perceived that a high proportion of individuals referred through these alternative routes (rather than through JCP) were appropriate for the programme. Information sharing at the point of referral.

The data that JCP shared with providers was covered by general data sharing permissions. It included the minimum details required to administer benefit and provide the contracted services. This included a method to flag additional needs, for instance the need for a hearing loop, required for initial engagement with the participant, but did not cover other health-related information.

Both the deep dive research and provider surveys highlighted issues with the information that providers received from JCP about a participant at the point of referral.

In the Wave One provider survey, 85% of key workers said that they received at least some form of information when participants were referred (Figure 2.3). The type of information most commonly received, was whether the participant’s participation was mandatory or voluntary (72%), followed by information about their health or disability (33%) and their work history (20%).

Figure 2.3: Types of information key workers received from JCP (Wave One provider survey)

Types of information	Percentage
Any information from any on (NET)	85%
Whether the participant’s participation was mandatory or voluntary	72%
Participants’ health or disability condition	33%
Participants’ work history	20%
An assessment of participants’ barriers to work	10%
An assessment of the support they might need	5%

Q_Info1: Which of the following types of information do you receive from Jobcentre Plus about the participants who are referred to you, before they start?

Base: All Wave One provider respondents (272)

Key workers who had received some form of information from JCP were asked about the quality of this information. Attitudes were negative on balance, with 49% of respondents saying that it was not accurate, while 56% said it was not up to date.

Key workers were least positive about the amount of information, with 90% saying that they did not receive enough information about participants to understand their needs and barriers to work before their first appointment. This was consistent with the deep dive findings, that providers felt there was a lack of information provided at the point of referral. Improved communication between providers and JCP could help key workers to feel prepared when meeting with participants for the first time.

2.2.4. Ease of the referral process

Key workers were generally positive at both waves about how easy it was to arrange initial appointments with participants. Consistent with the Wave One provider survey, at Wave Two, 52% of key workers said they found it easy to arrange initial appointments with participants while 13% found it difficult.

2.3. Wave Two provider perceptions of delivery

2.3.1. Overall reflections on delivery

Findings from the provider surveys and the deep dive research showed that the programme was considered an effective way of supporting participants to find work. In the Wave Two survey, 77% of providers agreed that the programme offered a tailored approach, which took account of the individual needs of participants and 88% of staff were confident it would help participants move into work. Key workers with LGP providers appeared to have higher levels of confidence that the programme would help participants move into work (46%) compared with of national WHP providers (31%).

Participants in the deep dive interviews could not identify a single element or ingredient of WHP that made the programme effective. Rather, there was a common belief that, where it was working well, it was delivering a combination of support that no other DWP-commissioned provision could offer.

The main, impactful, elements of WHP were perceived to be:

• Flexibility to tailor the conversation, as well as the wider help different participants could benefit from in addressing their individual needs. In the provider surveys there was an increase in agreement that the programme allowed them to be flexible in meeting participant needs, from 64% of provider staff at Wave One to 73% at Wave Two, suggesting greater flexibility over time.
• Having a dedicated person (i.e. a key worker) to focus on participants and listen. In comparison, JCP Work Coaches acknowledged they were rarely able to give individuals the same attention and focus that a key worker could, due to time constraints on appointments.

The provider survey showed an increase in the share who felt they had sufficient time to spend with participants from 43% in Wave One to 60% at Wave Two, suggesting this improved substantially as the programme bedded in.

• Its voluntary nature, which differentiated/disassociated it from a participant’s preceding mandatory engagement with JCP and created the basis for genuine conversations about their feelings towards, and readiness to, work.
• Longevity. At 15 months, WHP was longer than newer programmes like Restart and there was a strong belief amongst provider staff that this was justified if DWP wanted to achieve positive employment outcomes with those further from the labour market (an, several argued for an even longer timeframe for WHP support in future).

2.3.2. Key challenges to and facilitators of delivery

Key worker recruitment, training and retention

In the deep dives, the key worker was perceived by all parties to be central to the delivery of the programme. The role was seen to require a similar but broader skillset than equivalent roles on other programmes. This included welfare to work^{[footnote 17]} knowledge, case management skills and the ability to work with people who had health and other, potentially complex, support needs. Since programme delivery started, some providers also said they increasingly saw value in key workers having the ability to network and build relationships with external organisations and employers.

One prime provider reported initial difficulties in recruiting individuals to fill the key worker role and had not had a full complement of staff when the programme first started, which they acknowledged had impacted on early delivery. However, by the time of the Wave One deep dive research in 2020, all reported having sufficient staff in place.

Providers reported running training for key workers in preparation for programme delivery and after then as well. This included training to give staff the understanding to work with people who had a physical or mental health condition or impairment. Interviews with key workers conducted for the deep dive research and the provider surveys, both indicated that training had been effective.

Overall, there was a high degree of confidence among key workers in their ability to work with participants. In the Wave Two provider survey, 88% agreed that they had the necessary skills to deal with the participants they worked with (see Figure 2.4). Similarly, 85% of key workers agreed that they felt confident dealing with the range of participants and the issues they present. Both measures remained stable between waves one and two of the provider survey.

Figure 2.4: Key worker self-reported level of skill and confidence when dealing with participants, by wave (Wave One and Wave Two provider surveys)

	Strongly agree	Slightly agree	Neither agree nor disagree	Slightly disagree	Strongly disagree	Net agreee
I feel confident dealing with the range of participants and issues they present: wave 1	51 %	31%	9%	7%	1%	82%
I feel confident dealing with the range of participants and issues they present: wave 2	52%	33%	6%	6%	3%	85%
I have the necessary skills to deal with paricipants I work with: wave 1	54%	29%	8%	6%	2%	84%
I have the necessary skills to deal with paricipants I work with: wave 2	58%	30%	5%	5%	2%	88%

(Q_skills) How much do you agree or disagree with the following statements?

Base: All provider respondents Wave One (272), Wave Two (394).

One challenge reported by some providers in the deep dives was finding and retaining their key worker staff. This was linked to the introduction of new DWP programmes in response to the pandemic (Restart, JETS and Kickstart) which had reportedly created an extra demand for individuals with a similar skillset to a WHP key worker.

Establishing and maintaining participant contact

From the start of the programme there were a proportion of individuals who were referred to WHP but did not subsequently engage with the local WHP provider when they attempted to contact them to arrange an initial appointment, or who had initially engaged but then subsequently disengaged. The first participant survey, conducted roughly three months after referral to the programme, asked whether participants were still receiving WHP support at this point. Around 10% of participants (8% mandatory, 11% voluntary) were not receiving WHP support.

To an extent this was viewed by JCP and provider interviewees to be a fact of life with any third-party provider referrals, and partly a result of unavoidable factors such as poor health or other life crises preventing individuals from engaging, or individuals moving to another area or finding work. Equally, most provider staff interviewed in the deep dive research perceived that difficulties engaging participants on WHP were also partly linked to the appropriateness of referrals. In particular, provider staff understood that individuals should be willing to engage with support and have some prospect of progressing towards work in order to be eligible and perceived that some individuals did not meet these criteria. Communication between JCP and provider staff after the point of referral – for example, to try to resolve cases where an individual had started on the programme, but then disengaged – were variable across the areas in which the deep dive research was conducted. In some localities, key workers said they felt they could easily call or email the relevant JCP office to ask for their help in re-engaging with an individual on their caseload, but in others this open line of communication had not reportedly been established. Differences in practice in different areas appeared to reflect pre-existing relationships between JCP staff and providers locally and/or the willingness of each side to invest staff time in building this over time.

Engagement with local service providers

Provider staff in both waves of the deep dive research said that the effectiveness of their support had been enhanced by linkages they had made with local public and voluntary service providers in their area. These gave them a wider network of potential support they could signpost or refer participants to beyond their contracted supply chain.

The reported extent of this varied between providers and areas. Where increased linkages had been forged, this was generally at a localised scale through individual key workers developing relationships with local organisations mostly in the third sector. Linkages with local public services were less widely reported and provider staff highlighted challenges in identifying key contacts and getting their buy in.

‘You have a lead in a council, but councils are so huge, and there are seven different councils, so there haven’t been the intermediary figures to link in with housing, mental health…’ Provider manager

The provider in the deep dive sample that appeared to have made most progress in this respect was in one of the LGP areas. At the instigation of the combined authority that commissioned them, the provider had employed 10 integration co-ordinators, corresponding to the 10 local authorities in the contract area. Each local authority had an integration board, which the relevant co-ordinator attended, and which met regularly to forge links between their local WHP provision and relevant external services. The combined authority was also perceived to have played a positive role in brokering these contacts between the provider and local services.

At the time the Wave One deep dive research was conducted in 2020, providers in other areas had not devoted the same dedicated staff resource to local partnership working, but there appeared to be increasing recognition among all that making further progress with this would potentially benefit the effectiveness of their provision. One had recently established new local integration boards in some of its contract area. A manager in another prime provider acknowledged that they “could have done more” towards engagement with other local services at a strategic level and said they were considering investing more staff resource in this in the future.

At the time of the Wave Two deep dive research, it was unclear whether such plans had fully come to fruition. Providers generally reported similar levels of engagement with local services as they had in the previous wave. The pandemic had reportedly also posed some additional challenges to engagement – as other local services focused on their own responses to the pandemic – which may explain this.

Employer engagement

Providers did not generally cite this as an overwhelming barrier to achieving positive outcomes for participants. All had experience through previous DWP programmes of engaging employers as a means to identifying work experience and/or job opportunities for participants and indicated they had been able to adapt and apply approaches from this on WHP.

‘It’s about being honest with employers about what participants can do, focus on the positives and be frank about the challenges they face and how they need to be supported to stay in work.’ Provider manager

Facilitators and instances of good practice that providers had developed on WHP were also highlighted. Some had invested increasingly in employing dedicated staff (including, for example, individuals from a recruitment agency background) to complement the employer engagement that key workers themselves performed as part of their role. This was perceived to be effective in identifying increased placement and job opportunities for participants.

Tapping into larger employers’ Corporate Social Responsibility (CSR) agenda and referencing the national Disability Confident scheme^{[footnote 18]} could also reportedly be an effective lever for getting buy-in. In one of the LGP areas in the second wave of deep dive research, the combined authority was credited with having brokered links between the local WHP provider and major local employers, with potential to provide placement and job opportunities to substantial numbers of participants on the programme.

Further challenges are discussed in Chapter Four in relation to aspects of delivery of programme support, while the specific challenges created by COVID-19 are covered in the next section of this chapter.

2.4. Implications of COVID-19 for delivery

2.4.1. Mode of contact

The most immediate impact of the pandemic was a wholesale transition by providers from face-to-face to remote telephone and online contact – for both key worker support and the more specialist support participants could be referred to.

Provider staff in every deep dive area presented a positive narrative about how effectively they had adapted their programme support in the pandemic. They generally felt they had managed the transition to telephone and online contact successfully, and cited benefits to this both from a delivery and participant perspective. These included: increased internal efficiency and flexibility (with staff and subcontracted specialist provision no longer being tied to one geographical location and able to support participants across the entirety of a contract area); and increased quality of engagement with participants (with many said to be more comfortable with opening up in a remote rather than face-to-face format).

The ad-hoc COVID-19 survey of participants conducted in the same period also provided further evidence of the move to remote communication in terms of how participants engaged with their key worker. Participants cited telephone calls as the most frequently mentioned and used form of communication in the early months of the pandemic (69% mandatory and 74% voluntary (see Figure 2.5)).

Wave Two deep dive research with participants conducted during the pandemic identified that the majority of participants indicated they were comfortable with using remote communications and some even expressed a preference for it.

‘I found that [talking on the phone] was easier because I get nervous when I go and meet new people.’ Participant

There were some participants who said they would have preferred to meet with their key worker, and potentially access other more specialist support, face-to-face.

‘It wasn’t ideal, but I understood that we couldn’t meet face to face at the time because we were coming in and out of lockdowns. I struggled really because I’m one of those people that likes to put a face to the person I’m speaking to.’ Participant

There was also a concern that a minority of participants (generally older ones, living on their own) without basic IT skills, may have had access to quite limited support through the programme. Providers talked about having provided tablets to participants without IT equipment, but participants still needed basic IT skills to be able to use them to access support beyond telephone contact with their key worker.

‘The COVID-19 crisis has highlighted the lack of IT skills within my caseload. I would like to see a higher importance put on IT training to ensure all participants are able to use a computer.’ Key worker

Text messages (45% mandatory, 44% voluntary) and emails (45% voluntary, 43% mandatory) were also commonly mentioned, but few said they were the main form of communication, suggesting these supplemented phone calls. Despite the frequent use of video calls in many other settings during COVID-19, these were rarely reportedly used for key worker contact. Evidence from the deep dive research suggested video calls were being more widely used to access specialist non-key worker support such as training, with this typically adapted to online delivery using Teams, Zoom, or similar applications.

Figure 2.5: Communication mode with key worker since the start of the COVID-19 restrictions in March 2020 (Ad-hoc participants survey)

	Mandatory participants: main mode	Mandatory participants: any mention	Voluntary participants: main mode	Voluntary participants: any mention
By phone call	50%	69%	65%	74%
Email	7%	43%	7%	44%
Text	13%	45%	6%	44%
In a face-to-face meeting	2%	8%	2%	8%
Video calls (eg Zoom, Whatsapp or Skype)	0%	0%	0%	4%
Other	2%	4%	0%	2%
None of the above	25%	25%	19%	19%

CVID_MODE1: Since the start of COVID-19, have you received support in any of the following ways?

CVID_MODE2: And which is the main way you have you received support?

Base: All ad-hoc survey participants: Mandatory (55*), Voluntary (203) *CAUTION: Number of respondents below 100

In the ad-hoc COVID-19 survey the majority of participants (85% mandatory, 86% voluntary) were satisfied with the range of ways they were able to contact a key worker. This was broadly in line with the Wave One findings.

At the time of the Wave Two deep dive research in late 2021, providers said they were in the process of transitioning at least some of their support (notably key worker support) back to face-to-face. It was unknown whether the subsequent emergence of Omicron might have curtailed or delayed this. Longer-term, senior staff in all the providers in the areas in which the deep dive research was conducted indicated they planned to adopt a blended model of delivery - combining face-to-face and remote contact with participants - for the remainder of the programme.

2.4.2. Frequency of contact and availability of key workers

Providers in the deep dives believed that the partial reduction in frequency of key worker contact had been appropriate to the circumstances – especially at the start of the pandemic when there were limits to what they could practically do to progress participants into work.

Evidence from JCP staff suggested a slightly more mixed picture. There were indications that the frequency of key worker-participant contact might have dropped very significantly early on in the pandemic in certain areas (possibly as the local WHP provider was dealing with their own organisational challenges due to the pandemic).

For example, a Work Coach team leader in one of the areas in which the deep dive research was conducted said there were some participants they were aware of locally who had “largely been left to fend for themselves” for an extended period. However, this was not reported more widely across other parts of the contract area.

Ad-hoc survey participants were asked how often they were in contact with their key worker both before and after the start of the COVID-19 pandemic. Among both mandatory and voluntary participants, the proportion meeting with their key worker at least once a month fell in the first few months of the COVID-19 pandemic (from 85% to 61% for mandatory, and from 84% to 56% for voluntary participants - see Figure 2.6).

Figure 2.6: Frequency of contact with key workers, before and after the COVID-19 outbreak (Ad-hoc participants survey)

…before the COVID-19 outbreak

Frequency	Mandatory (LTU)	Voluntary (H & D/EAG)
At least once a month	85%	84%
At least once a week	23%	32%
Less often, but at least once a month	62%	53%
Once every 3 months	1%	1%
Less often than every 3 months	2%	3%
It varied	4%	4%
Don’t know	6%	7%
Prefer not to say	–	–

JCPOFFER_FMEETINGFREQ1: On average, how often did you meet or have contact with your key worker before the COVID-19 outbreak? JCPOFFER_FMEETINGFRECVD: On average, how often did you meet or have contact with your key worker since the COVID-19 outbreak?

Base: All ad-hoc participants: Mandatory (55*), Voluntary (203) *CAUTION: Number of respondents below 100

In the Wave Two participants’ survey, which was conducted between September 2020 and June 2021, just over half of participants reported meeting with their key worker at least monthly (56% mandatory, 52% voluntary), suggesting that frequency of contact had not increased at this point in the pandemic. However, the frequency of meetings reported varied according to when participants were recruited to the programme. Participants who were allocated to the programme between November 2018 and January 2020 were less likely to report meeting with their key worker at least monthly (mandatory 56%, voluntary 48%) than those allocated a year later (mandatory 75%, voluntary 73%).

Despite the fall in frequency of key worker meetings, 87% of mandatory and 81% of voluntary participants in the Ad-hoc COVID-19 survey were satisfied with frequency of contact with their key worker. While satisfaction had remained high, it had fallen slightly for voluntary participants since the Wave One participant survey (91%). As will be seen in Chapter Five, levels of satisfaction recovered a little by Wave Two for voluntary respondents to 88%, even though frequency of contact had not increased.

Similarly, participants’ satisfaction with key worker availability remained high but was slightly lower than at Wave One in the ad-hoc COVID-19 survey – see Figure 2.7. Among mandatory participants, 83% were satisfied with how available a key worker was when they needed them (compared with 90% at Wave One). Among voluntary participants 79% were satisfied (88% at Wave One).

Figure 2.7: Participant satisfaction with key worker contact (Ad-hoc participants survey)

Frequency of contact with key worker/Work coach

Frequency	Mandatory (LTU)	Voluntary (DG/EAG)
Net satisfied	87%	81%
Very satisfied	47%	47%
Fairly satisfied	40%	34%
Fairly dissatisfied	0%	8%
Very dissatisfied	5%	8%
Don’t know	9%	3%

How available a key worker/Work coach is when you need them

Satisfaction	Mandatory (LTU)	Voluntary (DG/EAG)
Net satisfied	83%	79%
Very satisfied	60%	54%
Fairly satisfied	23%	25%
Fairly dissatisfied	3%	7%
Very dissatisfied	4%	7%
Don’t know	10%	6%

CVID_COACHSAT1: And over that period, have you been satisfied or dissatisfied with: The frequency of contact respondent has had with their key worker/Work Coach; How available a key worker/Work Coach is when you need them.

Base: Ad-hoc participants: Mandatory (55*), Voluntary (203) *CAUTION: Number of respondents below 100

2.4.3. Impact of COVID-19 on challenges and facilitators of delivery

While general challenges and facilitators of delivery revealed by the deep dives were discussed earlier in this chapter, COVID-19 brought new challenges and facilitators as discussed below.

Participant engagement

As seen above, there were already reported challenges with establishing and maintaining participant engagement in the programme prior to the pandemic. Several providers indicated that the proportion of participants failing to start the programme after being referred had risen at the start of the pandemic.

Various explanations were suggested for this. Some participants were thought to be less likely to answer initial calls from a provider than they were to attend a face-to-face appointment. Participants may have seen limited prospect of a positive outcome from engaging in the programme, especially early in the pandemic when the labour market was contracting. The pandemic also created barriers to provider staff having a physical presence in JCP offices and being able to encourage engagement through warm handovers.

By late 2021, providers generally reported an improvement in start rates and were optimistic that levels of engagement were returning to pre-pandemic levels. This is supported by the referral data discussed earlier in this chapter and summarised in Figure 2.1. In terms of maintaining engagement with participants after they had started on the programme, provider staff did not perceive the pandemic to have had an overwhelming impact.

The results from the Wave Two participant survey suggested a similar conclusion. When asked the main reasons for no longer receiving support, the vast majority gave reasons not directly linked to COVID-19 (see Figure 4.1 for details). A minority spontaneously cited COVID-19 as the reason, with no specific detail (6% mandatory, 4% voluntary participants). An even smaller proportion gave more specific COVID-19 related responses such as having stopped engaging when they could no longer meet their key worker face to face (1% of both mandatory and voluntary participants) and being in one of the COVID-19 shielding groups (1% of voluntary participants).

Provider staffing

Providers reported initial challenges associated with their staff working remotely, in terms of maintaining their productivity and ensuring their well-being. As the pandemic progressed, they indicated these had been effectively addressed. Managers were reportedly in close contact with their staff through online team meetings and one-to-one check-ups. At least one provider said they had also given their staff access to some of the new wellbeing resources they had introduced for their WHP participants.

Introduction of the cost-plus model

As described in Chapter One, the WHP was introduced on a payment by results basis but from the start of the pandemic this was changed to a ‘cost plus model’, whereby contracted WHP providers received payments to cover their agreed costs to deliver the provision plus an agreed profit margin. All of the providers in the areas in which the deep dive research was conducted were appreciative of this change. They said it gave them a stable, continued source of revenue through the programme to meet ongoing delivery costs during a very challenging period. Some also indicated that it had afforded them a valuable opportunity to take stock and invest in some improvements to their provision (in addition to specific enhancements they made in response to the pandemic). For example, one talked about building new links with local education and skills providers to deliver training for their participants. Others mentioned new or enhanced internal databases of subcontracted and external support they had implemented that key workers could tap into more easily.

Cross-programme pooling resources

Providers who were delivering other DWP- or LGP-commissioned provision in their local area introduced since the start of the COVID-19 pandemic, as well as WHP, highlighted some spill-over benefits and economies of scale arising from this. For example, one had capitalised on new links with local support providers that had initially been developed by colleagues working on the JETS programme. Another reported that, rather than having separate (and potentially competing) employer engagement teams for each of the programmes they were delivering, a combined team had been created to identify work placement and job opportunities that could potentially benefit participants on any of the programmes.

3. Participant attitudes and characteristics

Chapter Three provides an overview of WHP participants’ demographic characteristics, health and work history. Findings are primarily taken from the Wave Two participants’ survey and supplemented with additional information from the deep dive research.

Summary

• Just under three-quarters of participants were voluntary referrals (Disability group or Early Access Group). The remainder were referred on a mandatory basis because they had been unemployed for at least 24 months.
  
• The profiles of mandatory and voluntary WHP participants were very similar: The majority were male (61% mandatory / 57% voluntary). And, while participants covered a wide age range, a relatively large proportion were aged 50-59 (29% mandatory / 30% voluntary). Around a fifth of participants had a Level 1 qualification or lower (17% mandatory / 22% voluntary).
  
• Reflecting the random allocation process to WHP, the profile of control group participants was consistent with those referred to the WHP.
  
• Work histories of WHP participants varied considerably, although more than eight in ten who took part in the Wave Two survey had been in employment at some point before being referred to the programme.
  
• Physical or mental health conditions, disabilities or illnesses expected to last at least six months were common in all groups of WHP participants. This included mandatory participants who had not been referred for health-related reasons.
  
• Participants reported a range of perceived barriers to finding work, the most common being physical or mental health conditions or disabilities. Challenges due to COVID-19 were also commonly reported.

3.1. Participant characteristics

The profile of participants sampled for the survey reflected the profile of WHP participants as a whole (for example by type of allocation and eligibility group). Accordingly, among surveyed Wave Two participants (Figure 3.1), 84% were allocated to the WHP, 15% were allocated to the randomised control group and one percent to the PSC. Just under three-quarters (73%) of surveyed participants were voluntary referrals, comprised of Disability group referrals (68%) and Early Access Group (EAG) referrals (5%). Mandatory referrals (27% of surveyed participants) were all Long Term Unemployed (LTU).

Figure 3.1: Breakdown of Wave Two participant sample by mandatory / voluntary participation, eligibility group, and allocation group (Wave Two participants survey)

	Mandatory (LTU)	Voluntary (DG/EAG	Eligibility group	Allocation group
Referral type	27%	73%	–	–
Disability Group	–	–	68%	–
Early Access Group	–	–	5%	–
Long-term Unemployed	–	–	27%	–
Work and Health Programme	–	–	–	84%
Randomised Control Group	–	–	–	15%
Public Sector Comparator	–	–	–	1%

Base: All respondents Wave Two (5655) WHP Wave Two (3525), Control Wave Two (2042), PSC Wave Two2 (88*)

Please note data on this slide combines responses collected at Wave One from longitudinal participants who participated at Wave Two and retrospective responses collected at Wave Two from fresh respondents. *CAUTION: Number of respondents below 100

This chapter looks at participant characteristics for the voluntary group, mandatory group and the control group^{[footnote 19]}. The commentary focuses on the mandatory and voluntary WHP participants and, where comparison was relevant, control group data also. The control group comprises both the voluntary and mandatory groups. However, as the voluntary group is significantly larger than the mandatory group, the control group is more similar to the voluntary group than the mandatory group.

The demographic profile of all three groups were similar: there was very little difference between mandatory and voluntary WHP participants, and the control participants in terms of gender, age profile, education level and ethnicity (see Table 5 Appendix Three).

On balance, participants were more likely to be male than female. They covered the full age range, but with relatively more aged 50-59. Most were educated to Level 2 or below, although three in ten had qualifications at Level 3 or higher. The over-50s and those out of work for 10 years or more were relatively more likely to have lower-level qualifications, and so may require greater support with basic skills.

The characteristics of the participants interviewed in both waves of the deep dive research were also similar those of the Wave Two survey sample. Two-thirds of participants were in the Disability group, there were more male than female participants on balance, and distributed over the full age range (see Table 6, Appendix Three for details).

3.2. Participant health and wellbeing

Overall, physical or mental health conditions, disabilities or illnesses expected to last at least six months were common among WHP participants. Voluntary participants were more likely to report a health condition (72%) than mandatory participants (55% see Figure 3.2). The higher level among voluntary participants was driven by Disability group participants (74%) with lower levels among EAG participants (47%). The participants interviewed in the deep dive research showed a similar pattern. This suggested that any impacts of ill health on employment may be common across all groups who were eligible for the programme, not just those who were referred because of their health or disability.

The reason not all participants in the Disability group identified as having a long-term health condition or disability in the survey is likely to be either a change in their health status, or because the participant’s understanding or definition of a long-term health condition was different to that used during recruitment.

At Wave Two, among those reporting any condition, the types of health condition were similar for all three groups. The most commonly reported conditions were mental health conditions, musculo-skeletal conditions or physical injuries, and chronic or system conditions (see Figure 3.2).

Figure 3.2: Participant health conditions (Wave Two participants survey)

Whether respondent has any health conditions

Response	Mandatory (LTU)	Voluntary (H and D/EAG	Control
Yes	55%	72%	66%
No	42%	25%	31%
Don’t know	2%	2%	2%
Prefer not to say	1%	1%	1%

Which health conditions respondents have (those with any)

Health condition	Mandatory (LTU)	Voluntary (HD/EAG	Control
Mental health	43%	50%	48%
Musculo-skeletal/physical injury	41%	46%	41%
Chronic/systemic	38%	36%	38%
Learning difficulties/cognitive	9%	12%	12%
Sensory impairment/communication	9%	9%	10%
Other condition	16%	16%	18%

WPK23: Can I check, do you have any physical or mental health conditions, disabilities or illnesses lasting or expected to last for six months or more?

WPK25: Could you tell me what your illness, health condition or disability is? Base: Wave Two participants who were asked: Mandatory (1,348), Voluntary (2,037), Control (1,954); Wave Two participants who were asked and have a health condition: Mandatory (749), Voluntary (1,436), Control (1,358)

Reflecting this, when asked to rate their own health on a scale of 0 to 10 where 0 was very poor and 10 was very good, mandatory participants gave themselves a relatively higher average score of 6.3, compared with an average of 5.5 for voluntary participants and 5.7 for the control group. Similar patterns of difference were seen in relation to the frequency of GP and hospital appointments.

3.2.1. Perceived impact of health on employment prospects

In the Wave Two participants survey, participants were asked to rate the impact of their physical and mental health on their ability to remain in a job. On the scale, where 0 was not at all and 10 was a lot, responses were very similar for mandatory and voluntary participants. The reported impact on ability to stay in a job was greater for physical health, with 66% of mandatory and 65% of voluntary participants of those with a physical health condition giving a score of five or more. It was slightly lower in terms of mental health, with 56% of mandatory and 57% of voluntary participants with a mental health condition giving a score of five or more.

Participants with any health condition (physical, mental or both) were asked to rate their agreement to three further questions on the impact of their health on work and employment (on the same scale). Agreement was classified as a score of six or more.

• 55% of mandatory and 56% of voluntary participants agreed that they were worried that people would not employ them because of their health condition
• 56% of mandatory and 59% of voluntary participants agreed that they were worried that it would prevent them from staying in a job
• of 45% of mandatory and 44% of voluntary participants agreed that the idea of working making them anxious

While responses were very similar between mandatory and voluntary participants, for all three of these concerns, agreement was higher in both groups among those with poorer self-assessed health. Those with both physical and mental health conditions were relatively more likely to give a high score in terms of worrying it would stop them getting or keeping a job. However, those with a mental health condition were more likely than those with a physical health condition to say the idea of working made them anxious.

Levels of concern were also generally higher for voluntary participants among women than men, and among those currently out of work.

Participants taking part in the deep dive research expressed similar concerns regarding the influence of their health conditions or disabilities on finding work, which represented the most commonly reported barrier for many of the interviewees, particularly, but not exclusively among those in the Disability group.

3.3. Participant work history

More than eight in ten participants who took part in the Wave Two survey had been in employment at some point before referral to the programme, with little difference between the groups (see Figure 3.3). This was usually being mostly in and out of employment (58% mandatory, 61% voluntary) before being referred. In both groups, men were relatively more likely than women to have been mostly in or out of work, with the under-30s relatively more likely to say they had never been employed.

Figure 3.3: Participant work history, before referral to the WHP (Wave Two participants survey)

Employment status	Mandatory (LTU)	Voluntary (HD/EAG	Control
Mostly employed	58%	61%	57%
I have mostly been in paid employment	24%	30%	31%
I have been in and out of paid employment	34%	31%	26%
I have mostly been out of paid employment	27%	25%	26%
I have never been in paid employment	13%	12%	15%
Don’t know	2%	2%	2%

ESASUPP_PAIDWORK1: At that time, which of the following statements best describes you work history?

Base: Wave Two participants who were asked: Mandatory (1,348), Voluntary (2,037), Control (1,954)

Mandatory participants who had ever worked were more likely to report having been out of work for at least a year (77%) than voluntary participants (66%), reflecting the eligibility criteria for this group (see Figure 3.4). Being out of work for 10 years or more was relatively more common for participants with a physical health condition, and for women in the voluntary group.

Figure 3.4: Length of time since respondent had been in paid work (Wave Two participants who have ever been in work)

Length of time	Mandatory (LTU)	Voluntary (HD/EAG	Control
10 years or more	19%	16%	16%
3 years to less than 10 years	31%	26%	28%
1 year to less than 3 years	27%	24%	26%
Less than 1 year	13%	26%	21%
Don’t know	10%	8%	9%

ESASUPP_LENGTHWORK1: How long was it since you had paid work before …? Base: Wave Two participants who were previously in paid employment and were asked: Mandatory (1,178), Voluntary (1,801), Control (1,676)

3.4. Barriers to finding work

Among participants in the Wave Two survey, barriers to finding work were assessed in a two-stage process. Participants were initially asked an unprompted question about the barriers they faced finding work; they were then asked to identify the single main barrier they faced from among these issues (see Figure 3.5^{[footnote 20]}). A shorter prompted list was then used to assess further issues experienced by participants that could also potentially act as barriers.

Figure 3.5: Barriers to finding work (Wave Two participants survey)

Barrier to finding work	Mandatory participants: main barrier	Mandatory participants: any barrier	Voluntary participants: main barrier	Voluntary participants: any barrier
Physical/mental health/disabilities	25%	34%	43%	52%
COVID-19 related	16%	26%	11%	20%
Lack of vacancies/too much competition	8%	20%	6%	17%
Age related factors	6%	14%	4%	11%
Your own self-confidence	3%	10%	3%	12%
Not having right skills for jobs interested in	3%	12%	3%	9%
Length of time out of work/between jobs	2%	11%	1%	8%
Lack of work experience	3%	10%	2%	7%
Lack of transport	2%	10%	2%	9%
Personal or domestic issues (excl. health)	4%	8%	4%	7%
Lack of childcare	4%	6%	2%	4%
Lack of flexible working (not childcare)	1%	6%	0%	4%
Lack of motivation	1%	6%	0%	6%
Lack of adequate references	0%	5%	0%	4%

JCPOFFER_BARRIERS1: What would you say [is / was] preventing you from finding work [at the moment / before you found your current job]?

JCPOFFER_BARRIERS2: And what do you feel [is/was] the main barrier preventing you from finding work [before you found your current job]?

Base: All Wave Two participants: Mandatory (1,414), Voluntary (2,111)

The deep dive research with participants also explored their attitudes towards work and perceived barriers to this. The majority described joining the programme to find work, suggesting a widespread willingness to return to or find work for the first time. However, the expectation of actually securing a job as a result of the programme varied considerably among the interviewees, with a range of barriers being reported.

The deep dive participants had most commonly reported that underlying physical or mental health conditions or disabilities were their main barriers to work, most acutely for those in the Disability group, but also evidenced across the EAG and LTU groups. This was confirmed in the Wave Two survey (Figure 3.5). For voluntary participants, by far the main barrier came from physical or mental health issues and disabilities (52% a barrier, 43% the main barrier, and with this a greater barrier for Disability group participants (54%) than for those in the EAG (30%). While this was also the top barrier for mandatory participants, this was selected by relatively fewer participants (34% as a barrier, 25% as the main barrier).

COVID-19 was mentioned spontaneously by both groups as the second ranked barrier overall, slightly more so by mandatory participants (26%) than by voluntary participants (20%), with mandatory participants also more likely to say this was their main barrier to finding work (16%) than were voluntary participants (11%).

When a comparable question about barriers was asked on the Ad-hoc COVID-19 survey, very similar barriers were reported, suggesting there was no significant change later on in the pandemic.

There was little difference between the two groups at Wave Two in terms of how frequently other barriers were mentioned, including lack of vacancies, length of time out of work and a range of personal barriers.

Health-related barriers, age and sustained unemployment

In the deep dives, where ongoing health conditions were reported as the main barrier to work, these were often combined with lengthy periods away from work and age-related barriers. These barriers, collectively, were perceived to reduce the chance of finding work and/or restrict the nature of work they felt able to consider.

Similar links between health conditions and sustained unemployment and age were reported by participants in the Wave Two survey. For example, among mandatory participants, those aged 50 and over and those out of work for 10 years or more were relatively more likely to report health barriers. This reflected the nature of those out of work for longer, who tended to be older or have health problems.

Further, when responding to a prompted question asking about work barriers, around six in ten said that they had experienced being out of the workplace for eighteen months or more (60% mandatory, 58% voluntary), much lower than the proportion citing time out of work spontaneously as a barrier. This suggested that, while widely experienced, being unemployed for 18+ months was not necessarily seen as a major barrier to finding work.

Other barriers cited spontaneously

The deep dives found that, for others, notably those without/with less severe health conditions and those away from work for 12 months or less, the main barriers to work cited included the limited availability of suitable local vacancies while caring responsibilities, personal commitments and attending regular medical appointments limited their availability for work.

In the Wave Two survey, lack of vacancies was cited as a barrier by 19% of mandatory participants and 23% of voluntary. Those out of work for three years or less were relatively more likely to cite a lack of vacancies as a barrier in both groups, although health remained a top issue, even for these participants.

While overall few mentioned family and caring issues as barriers, women were more likely than men to cite issues such as lack of childcare (13% vs. 2%), lack of flexible working (9% vs. 4%) and personal or domestic issues (11% vs. 6%).

Furthermore, lack of self-confidence was cited as a barrier by 7% of mandatory and 6% of voluntary participants. This was relatively more likely to be a barrier at all for the under-30s, those in employment and for those with a mental health condition in both groups.

Findings in this section highlighted the importance of offering a tailored package of support via the programme, based on the individual needs and circumstances of each participant.

3.5. Participant needs

A majority of key workers who took part in the provider surveys said that the needs of the participants they worked with were more complex than those anticipated. However, this perception appeared to decrease as the programme progressed. At Wave One more than half of providers reported that participant needs were a lot more complex than anticipated (54%), this had fallen to 38% by the Wave Two interview (Figure 3.6).

Figure 3.6: Provider perception of participant needs (Wave One and Wave Two provider survey)

Perception of needs	All providers W1	All providers W2
More complex	80%	76%
A lot more complex than anticipated	54%	38%
A little more complex	26%	38%
As complex as anticipated	18%	24%
A little less complex	1%	1%
A lot less complex	–	–
Don’t know	–	–

Q_COMPLEXITY: Thinking of the participants that you work with would you say that their needs are generally…?

Base: Wave One Providers: (268); Wave Two providers: (394)

In the deep dive research, key workers interviewed also often referred to having participants with “complex” or “more complex” needs. Key workers said these were participants who had, in addition to barriers directly associated with work and health, other personal challenges that may need to be addressed in order for them to enter employment. Such challenges included alcohol or drug abuse, serious debt problems, and homelessness.

‘People have more complex issues. I have one person who’s made themselves homeless. They won’t speak to the jobcentre and won’t speak to the council about rehousing them.’ Key worker

4. Programme support and satisfaction

This chapter provides an overview of the support provided by the WHP including the role of key workers and specialist support provided. It also provides an overview of participant satisfaction with the Work and Health Programme, drawn from a number of measures of satisfaction with, and perceptions of, the impact of support received in Wave Two of the Participants Survey.

Summary

• every participant on the WHP programme had an assigned key worker. Regular meetings between participants and key workers were arranged for up to 15-months. In many cases the key worker was assigned to participant for the full duration of the programme.
• Key workers typically provided some forms of support themselves to participants which were mostly related to job-search activities. Other forms of support included: support or advice for setting up a business or becoming self-employed; help managing health conditions; financial management/debt advice; training, and provision of volunteering or work experience opportunities.
• WHP participants tended to receive a wider range of support compared with the control group. With the exception of training courses, participants in the control group were less likely than WHP participants to report all other types of support.
• Key workers could refer participants to specialist support to address specific needs. This support was primarily provided through supply chains arranged by the providers. In some cases, specialist support was also provided by in-house teams or through informal partnerships with local organisations.
• Providers generally seem to have had access to most services that participants needed via their supply chains. Where gaps in provision existed, these most often related to health provision (including mental health, physical disabilities, mental disabilities), addiction services and bereavement support, support for housing and homelessness, and learning English as a second language. The most widely perceived gap or weak spot highlighted by providers was around mental health
• Adaptations to the programme offering were made in light of changing participant needs due to the COVID-19 pandemic. These were largely related to mental health.
• Overall satisfaction was high for both voluntary and mandatory participants. Participants in the control group were slightly less likely to be satisfied with the support they received than WHP participants.
• Participants identified several positive aspects of the programme including the positive, friendly delivery environment, flexibility of delivery, programme duration and additional support provided over and above the focus on finding work.
• Both mandatory and voluntary participants were similarly satisfied with the range of ways they could get in contact with their key worker as well as the frequency of contact and extent to which the support was tailored to their needs.
• The pandemic appeared to have little impact on overall satisfaction levels with only a minority expressing dissatisfaction with the service provided during the pandemic

4.1. Overview of key worker and specialist support

The providers in the contracted CPA, DDA and LGP deep dives all described a similar model of delivery based on providing key types of support to participants:

• key worker and specialist support
• a range of types of support, including work experience, entry and in-work support.

All parties viewed key workers as a central aspect of WHP delivery. Every participant on the programme had an assigned key worker who they would speak with regularly for up to 15 months (followed by up to six months of in-work support). They also agreed an action plan with participants that would help them back into work.

4.1.1. Key worker support

Every participant on the WHP programme had an assigned key worker. Regular appointments were arranged between key workers and participants for up to a 15-month period. Elements of the engagement key workers had with participants, for example its timing and duration, were prescribed by DWP service standards. Within two days of receiving a new referral, key workers were required to contact the participant to arrange an initial meeting or call and they would then be in contact with them at regular pre-determined intervals while they remained on the programme.

In the deep dive research, providers all described similar processes for the initial meetings that key workers had with participants. These were based around explaining the support participants would receive, establishing what their circumstances and potential support needs were, and developing this into an action plan which was subsequently reviewed and updated as things progressed. Key workers stressed the importance of building rapport with participants to ensure they were comfortable talking about their challenges and potential support needs.

‘It is just to listen to them really. Not to be judgemental. There might be valid barriers and reasons for not working. Being patient. Gaining trust.’ Key worker

Key workers interviewed in the deep dive research said they provided some forms of support themselves to participants – mostly related to job-search activities. For other support needs they could also offer general advice and refer participants onto specialist provision.

There was evidence from the deep dive research of some differences in how providers deployed their key workers over time. At the start, most had determined that participants would have the same key worker for the duration of their participation in the programme. The rationale for this was continuity for the participant and being able to capitalise on the rapport built up over time between them and the key worker.

Subsequently, some providers had introduced an alternative model in which different key workers were responsible for working with participants at different stages in their journey towards work. The rationale for this was that different key workers have different capabilities and skills, more suited to one stage in the journey than another. The potential downside was a lack of continuity for participants but by the time of the second wave of deep dive research in late 2021, this appeared to be the model settled upon by most providers interviewed.

4.1.2. Continued receipt of support

Three in ten mandatory participants (30%) and a similar proportion of voluntary participants (27%) reported that they were still receiving support from the Work and Health Programme at the time of the Wave Two interview. In contrast, far more participants (72%) in the control group said that they were still in receipt of support through JCP, which could potentially influence their attitudes.

The main reasons given by both WHP groups for ending support (shown in Figure 4.1) were related to completing the programme or finding work. Mandatory participants were more likely than voluntary to say they had come to the end of their in-work support (36% vs. 27%) with voluntary participants more likely to say they had moved into paid work (21% vs. 12%), suggesting a slightly earlier stage of progress for voluntary participants. Both were equally likely to say they had completed the programme and been referred back to JCP (mandatory 17%, voluntary 15%). Other reasons included changes to health, and lack of provider contact.

Among both mandatory and voluntary groups, participants aged under 50 were relatively more likely to report moving into paid work as a reason, as were those with a higher level of qualification (level two and above for mandatory, and level 1 and above for voluntary), and those with no health condition. In both groups those out of work for less than a year were considerably more likely than average to report moving into paid work (mandatory 24%, voluntary 32%).

Figure 4.1: Reasons for no longer receiving WHP support (Wave Two participants who are no longer receiving support)

Reasons for no longer receiving WHP support	Mandatory (LTU)	Voluntary (HD/EAG)
Completed/came to end of in-work support	36%	27%
Moved into paid work	12%	21%
Completed programme of support and referred back to Jobcentre Plus	17%	15%
Changes to health or health-related reasons – not COVID-19	6%	11%
Was not contacted by provider	5%	5%
Due to COVID-19	6%	4%
Decided the support was not relevant to needs	3%	3%
Personal or domestic issues	3%	2%
Issues with key worker/work coach 1%	2%
Practical issues meant couldn’t attend (not related to health reasons)	2%	1%

MWA_REASONSTOP: What are the reasons you are no longer receiving support through […]? NOTE: Showing reasons given by 2% or more of either group

Base: Wave Two participants who have stopped receiving support: Mandatory (959), Voluntary (1484)

4.1.3. Specialist support

As described in Chapter Two, prime providers in areas in which the deep dive research was conducted, had a supply chain (or equivalent non-contractual framework) of specialist providers that key workers could refer participants onto to address particular support needs. This was supplemented by in-house specialist teams some prime providers had and, as the programme progressed, other external local providers outside their formalised supply chain that key workers could potentially also sign-post or refer participants to.

The types of specialist support delivered to participants on the programme were wide-ranging, but at a high-level, providers in the deep dive research sample said they were delivering variations of the following:

• Employability support. Much of this was reportedly delivered by key workers themselves but several providers also had more intensive employability support (e.g., updating CVs, job searching techniques, motivation and confidence building, and interview coaching) that participants could be referred onto. Additionally, several had some form of dedicated support for self-employment.
• Help with managing health conditions. Including general support and advice that participants with any health condition could benefit from and dedicated support for different types of health condition (e.g., employment support and advice for visually and hearing-impaired participants or those with mild-moderate mental health conditions).
• Financial management/debt advice and support. Reported as a common support need amongst participants and a potential barrier to engaging in support to return to work if not addressed.
• Training. Generally short courses but which could be valuable in enabling participants to meet the basic entry requirements for certain employment sectors.
• Provision of volunteering opportunities. Intended as an alternative or additional option to work experience placements.
• Dedicated support for Early Access Group participants. This included support provided by subcontracted third sector organisations specifically for care leavers, homeless people or ex-armed forces.

The ability of the programme to provide a range of work, health and other related support to participants was widely perceived as one of its strengths, and a key facilitator to achieving positive outcomes with participants.

‘With someone that hasn’t been working for 10 years, there may be areas they need help to address before they can even look at going back into work. The programme gives us the tools to be able to help that participant get through those barriers.’ Key worker

In the Wave Two provider survey more than half (61%) of key workers agreed that their network of partners and suppliers was large enough to deliver all the services that participants needed (see Figure 4.2). This was an improvement on the Wave One survey results. Most key workers (75% at Wave Two) also agreed they were confident about which partners and suppliers to go to for the specific needs of their participants in both waves.

Figure 4.2: Access to specialist support to meet the needs of participants (Wave One and Wave Two providers survey)

Our network partners and suppliers is large enough to eliver all the services that participants need

Level of agreement or disagreement	Wave 1	Wave 2
Net agree	55%	61%
Strongly agree	24%	19%
Slightly agree	31%	43%
Neither agree nor disagree	18%	18%
Slightly disagree	8%	5%
Strongly disagree	17%	14%
Don’t know	–	–

I’m confident about which partners and suppliers to go to for the specific needs of my participants

Level of agreement or disagreement	Wave 1	Wave 2
Net agree	74%	75%
Strongly agree	36%	33%
Slightly agree	38%	41%
Neither agree nor disagree	14%	14%
Slightly disagree	3%	3%
Strongly disagree	8%	8%
Don’t know	–	–

Q_Network: How much do you agree or disagree with the following statements?

Base: All provider respondents Wave One (272), Wave Two (394).

4.1.4. Meeting the breadth of participant needs

Most providers in the Wave One and Wave Two surveys felt they had sufficient access to specialist support and agreed they were confident about which partners and suppliers to go to for the specific needs of their participants in both waves. These survey findings were consistent with the deep dive research among providers, although this did identify some difficulties with sourcing more specialist support across what were very large geographic areas.

A significant minority (45%) of providers in the Wave Two survey said they were unable to find suitable delivery partners for some specialist services, and although this was a high proportion, the survey did not provide any indication of the frequency or quantity of any such gaps in services.

Such gaps were, however, most likely to relate to health provision (including mental health, physical disabilities, mental disabilities), addiction services and bereavement support, support for housing and homelessness, and learning English as a second language^{[footnote 21]}.

The one widely perceived gap or weak spot highlighted by providers in the deep dive research, was around mental health (this is discussed specifically in relation to COVID-19 in section 4.2.7). Providers indicated they had anticipated some participants having support needs in this area and reflected this in the composition of their supply chain or in-house provision. Equally, they emphasised that this was for people with mild-moderate mental health needs and typically had an employment focus – for example training on managing stress or anxiety at job interviews and in the workplace. They did not perceive it to be theirs or the programme’s role to help people overcome more serious mental health conditions.

‘Ultimately it [the WHP] is an employability programme, not a health service or a fast track into the NHS.’ Provider manager

Key workers said they would advise participants with such conditions to consult with their GP and signposted them to local mental health services and charities such as Mind. However, these were widely reported to be difficult for participants to quickly access.

‘The trouble is when we try to refer out for support for people with mental health issues. The waiting list is so long and what do you do while you’re waiting?’ Key worker

4.2. Types of support provided

4.2.1. Support in gaining employment

The WHP had been established for more than two years at the time of the Wave Two provider survey. At this point 98% of the key workers surveyed said they had made efforts to fill job vacancies with employers.

In the deep dive research, it was common for providers to describe doing their own online research and cold calling employers to facilitate this. Senior provider staff reported that alongside this, additional job opportunities were identified corporately through engagement with large national employers and/or inhouse teams responsible for engaging local employers.

Providers also reported providing financial assistance to participants – for example to cover the travel costs to a job interview venue or to buy clothes to wear at it. In addition, participants could be provided with advice and coaching to prepare them for job interviews.

A number of measures at Wave Two of the participants survey offered evidence on the support and job search activity offered by the programme.

Figure 4.3 shows the types of employment support that survey participants reported receiving at Wave Two. Figure 4.5.

Mandatory participants were slightly more likely than voluntary participants to report having received all but two of these types of support, with both groups equally likely to receive the most common form of support, in writing CVs, job applications and interview skills (mandatory 71%, voluntary 68%) and the possibility of taking a training course from a local provider (mandatory 49%, voluntary 46%). The greatest differences were for work experience or voluntary placements (mandatory, 27%, voluntary, 21%) and support in maths, English and IT (mandatory 34%, voluntary 28%) with relatively little difference for other forms of support.

Around six in ten of in each group reported receiving general careers advice with around half having a skills assessment. Each other form of support was reported by, at most, a third of either group.

Participants in the control group were just as likely to report the possibility of training courses as both voluntary and mandatory participants but were less likely than mandatory participants to report all other types of support. In both voluntary and mandatory groups, the following were relatively more likely to be reported by the under-30s: help with CV writing and job applications (80% mandatory, 76% voluntary), work experience (37% mandatory, 26% voluntary), and support from local employers (27% mandatory, 28% voluntary). Among voluntary participants, work experience was also relatively less common among those out of work for three years or less (16%).

Figure 4.3: Whether received any of listed types of support (Wave Two participants who started receiving support)

Types of support	Mandatory (LTU)	Voluntary HD/EAG	Control
Help with writing a CV, job applications or interview skills	71%	68%	59%
General careers advice	61%	56%	51%
An assessment of your skills	51%	46%	37%
The possibility of training course at local college/other provider	49%	46%	47%
A session on motivation or confidence	48%	42%	30%
Support or training in maths, reading, writing/English language, IT	34%	28%	29%
A work experience placement or voluntary work	27%	21%	24%
Support from local employers	22%	19%	16%
Support or advice for setting up business/becoming self-employed	18%	14%	11%

WORKPROG_SUPP1: Have you received any of the following support through the programme? This could be support you would receive from […] or from another organisation that they would refer you to. Base: Wave Two participants who had started receiving support: Mandatory (1,383), Voluntary (2,057), Control (2,042)

In addition to more conventional employment support, almost half of mandatory participants (48%) reported a session on motivation or confidence, with fewer voluntary participants (42%) selecting this and far fewer in the control group (30%).

Mandatory participants who had never been in paid work or had been unemployed for ten years or more were more likely to report such a session (each 53%) than those out of work for less than ten years (45%), while voluntary participants who had never worked more likely to do so (47%) than those out of work for less than a year (37%).

White voluntary participants were also less likely to report a session on motivation and confidence (40%) than participants from all other ethnic groups combined (49%).

4.2.2. Support for self-employment

While 18% of mandatory participants and 14% of voluntary participants had already selected ‘support or advice for setting up a business or becoming self-employed’ from the list of types of support, WHP participants who had not done so were also asked separately if they had received this type of support and, if not, if they would have liked to. These figures have been combined to provide a total prompted figure for receipt of support in setting up a business or becoming self-employed (see Figure 4.4). In total, more mandatory participants (24%) reported having received such support compared with voluntary participants (19%). There was a similar unmet need in both groups, with 14% of mandatory participants and 17% of voluntary participants saying they had not received this support but would have liked to.

Figure 4.4: Whether received and/or wanted help setting up own business or becoming self-employed (Wave Two participants who started receiving support)

Response	Mandatory (LTU)	Voluntary (HD/EAG)
Yes	24%	19%
No – but would have liked to	14%	17%
No – but would not have liked to	59%	62%
Don’t know	3%	3%

WORKPROG_SUPP1: Have you received any of the following support through the programme? JCPOFFER_SELFEMP1: Since you started receiving support from …] have you been offered help with setting up your own business or becoming self-employed? JCPOFFER_SELFEMP2: Would you have liked to discuss help with setting up your own business or becoming self-employed?

Base: Wave Two participants who had started receiving support: Mandatory (1,383), Voluntary (2,057)

Note: Control participants were not asked this question

4.2.3. Practical and personal advice and support

In addition to support with gaining employment and self-employment, evidence on other types of practical and personal advice and support received through the programme was collected via the Wave Two Participants Survey.

The types of support that Wave Two survey participants were most likely to have received were financial support to cover the costs of looking for work, with this higher for mandatory (47%) than voluntary participants (41%), and advice or support on their personal wellbeing or mental health (mandatory 44%, voluntary 45% - see Figure 4.5). Almost as many received advice or support on a health condition or disability, with this understandably higher for voluntary (41%) than mandatory participants (35%). Mandatory and control group participants were more likely to get general financial advice than those in the control group (25% mandatory vs.23% voluntary vs. 17% control). Getting help with housing issues was similar among all groups (15% mandatory vs. 14% voluntary vs. 14% control). On average fewer than ten percent of any group got help with their caring responsibilities, drug or alcohol problems, and their criminal record.

Participants in the control group were less likely than WHP participants to report getting financial support to look for work (28%) and advice on personal wellbeing (29%), highlighting the broader range of support offered by the WHP.

Some support types were clearly targeted to need, with participants with health conditions more likely than average to get related support, and with women and younger participants more likely to get help with caring for children or adults.

In addition, among voluntary participants, those aged 30-39 were more likely to have received financial advice (28%) than those aged 50-59 or over (20%), while those with level one qualifications were relatively more likely to get financial advice (30% with Level 1 qualifications vs. 22% with Level 2 qualifications or 20% with less than Level 1 qualifications), help with drug or alcohol problems (11%) and help with a criminal record (12%). Voluntary participants out of work for more than ten years or who had never been in work were less likely to report help with the costs of looking for work (36% and 33% respectively) compared with those out of work for under ten years (43%).

Figure 4.5: Whether received any of listed types of support (Wave Two participants who started receiving support)

Types of support	Mandatory (LTU)	Voluntary (HD/EAG)	Control
Financial support to cover costs of looking for work	47%	41%	28%
Advice/support on personal wellbeing and/or mental health	44%	45%	29%
Advice or support on health condition or disability	35%	41%	32%
Financial advice of some sort	25%	23%	17%
Help with housing issues	15%	14%	14%
Help or advice in relation to looking after children or adults	11%	8%	6%
Help with any drug or alcohol problems that you may have	7%	6%	5%
Help or advice related to having a criminal record	7%	6%	5%

WORKPROG_SUPP1: Have you received any of the following support through the programme? This could be support you would receive from […] or from another organisation that they would refer you to. Base: Wave Two participants who had started receiving support: Mandatory (1,383), Voluntary (2,057), Control (2,042)

4.2.4. Action plans

Key workers developed action plans which outlined activities to support the participant back into work. As shown in Figure 4.6, 77% of mandatory and 74% of voluntary participants reported having an agreed action plan. This was slightly lower for the control group (71%). Among those who reported having an action plan, mandatory participants were slightly more likely to report having completed all or most of their goals (65%) compared with voluntary participants (61%), but not compared with the control group (66%).

While there was no difference in reporting having an action plan by health and disability status, in both groups, those reporting having a plan with any health condition were less likely to report having completed some or all of the goals compared with those without such a condition (mandatory 61% vs. 70%, voluntary 59% vs. 68%).

Age and education played a role for voluntary participants, with older participants and those with lower qualifications less likely to have an action plan. Among those aged 30-49, 79% said they had a plan compared with 70% of those aged 50 and over, as did 68% of those with a qualification below level one compared with 77% with a qualification at level one or above. Among those with a plan, there were no differences by age or education, however, in likelihood of having completed their goals.

Mandatory participants who had never been in work were less likely to report having a plan (70%) than those out of work for up to 10 years (79%). Of those with a plan in this group, those out of paid work for less than three years were more likely to report having completed all or most of their goals (71%) than those out of work for longer (61%).

Figure 4.6: Whether agreed action plan (Wave Two participants who started receiving support) and whether completed goals (Wave Two participants with action plan)

Whether have action plan

Response to having action plan	Mandatory (LTU)	Voluntary (HD/EAG)	Control
Yes	77%	74%	71%
No	17%	18%	22%
Don’t know	6%	8%	7%

Whether completed goals (those with plan)

Goal status	Mandatory (LTU)	Voluntary (HD/EAG)	Control
All/most	65%	61%	66%
Completed all goals	31%	28%	33%
Completed most goals	34%	33%	34%
Completed some goals	23%	23%	22%
Did not complete any goals	7%	11%	6%
Don’t know	6%	5%	6%

NEW_ACTIONPLAN1: And had you agreed an Action Plan with […]?

Base: Wave Two participants who had started receiving support: Mandatory (1,383), Voluntary (2,057), Control (2,042) NEW_ACTIONPLAN3: Have you completed the goals you set in your action plan?

Base: Wave Two participants with action plan: Mandatory (1,062), Voluntary (1,533), Control (1,439)

4.2.5. In-work support

Of those in work at Wave Two, 34% of mandatory and 39% of voluntary participants said they had received in-work support from the programme (Figure 4.7). Despite this, most participants felt the amount of support was about right, albeit with this higher among voluntary (71%) than mandatory participants (61%), Control participants were both less likely to report getting support (20%) and, presumably as a result, less likely to feel they had the right amount of support (54%).

Among participants who had received in-work support, just 6% of mandatory and 7% of voluntary felt this was too little. However, 31% of both mandatory and voluntary participants who had not received any in-work support said they did not receive enough support.

Figure 4.7: Whether participant received in-work support and opinion of amount of contact had from key worker/Work Coach since starting work (Wave Two participants currently in work)

Whether had in-work support

Had in-work support	Mandatory (LTU)	Voluntary (HD/EAG)	Control
Yes	34%	39%	20%
No	62%	60%	77%
Opted out	2%	1%	–
Don’t know	–	–	–

Opinion of amount of contact

Opinion of amount of contact	Mandatory (LTU)	Voluntary (HD/EAG)	Control
To much	4%	4%	2%
About right	61%	71%	54%
Too little	23%	18%	30%
Don’t know	12%	7%	14%

WP_H2: Have you received any in-work support from a […] since you began your current job?

WP_H4: Thinking overall about the amount of contact you had from your […] after you had started work, would you say it was…

Base: Wave Two participants currently in work: Mandatory (183), Voluntary (374), Control (254)

Of the relatively few participants who reported having received in-work contact, 41% of mandatory and 45% of voluntary participants said that it had a positive impact on staying in work, but slightly more saying it made no difference (see Figure 4.8). Too few mandatory participants reported a positive impact to explore this issue further, but among the 62 voluntary participants who said their in-work contact had impacted positively, this was most often related to keeping them motivated (81%) and managing health conditions (68%).

Figure 4.8: Impact of contact on staying in work (Wave Two participants with in-work support) and how contact helped where there was a positive impact (Wave Two voluntary participants with positive impact)

Impact of contact on staying in work

Impact of working	Mandatory (LTU)	Voluntary (HD/EAG)
Positive impact	41%	45%
Made no difference	58%	51%
Negative impact	2%	0%
Don’t know	0%	4%

How contact helped them stay in work (voluntary participants with positive impact)

Staying in work helped with	Percentage
Keeping you motivated	81%
Managing health condition in work	68%
Employer understands difficulties	50%
Negotiating flexible working	48%
Opportunities for progression	41%
Advocating on your behalf	39%
Increasing number of hours worked	38%
Increasing your income	37%
Securing training opportunities	30%

WP_H5: Did the contact you had after you started work have any impact on your staying in work? Would you say it had a…?

Base: Wave Two participants currently in work who had support: Mandatory (60*), Voluntary (136) WP_H6: In what way(s) did the contact you received after you had started work help you stay in work?

Base: Wave Two participants who received positive in work support: Voluntary (62*) *CAUTION: number of respondents below 100

In the deep dive research, provider staff all said they had processes in place to continue to engage with participants after they had started a job and entered employment. This was based on maintaining weekly or fortnightly contact with participants, timed so it did not interfere with their working hours.

‘We flex to them, if they need a call at six o’clock because they work till 5.30pm, we will do that.’ Key worker

This contact was reported to continue until the participant reached a particular earnings threshold which triggers an outcome-related payment to the provider (see Chapter One for a full description). This was generally four to six months after they started work.

In-work support was primarily being delivered by key workers, although some providers said they had in-house teams responsible for this. For example, in one area it had been decided that key workers would continue weekly contact with participants for the first four weeks they were in work and then this would be handed over to their in-work support team.

The nature of in-work support reportedly ranged from general reassurance and encouragement through to direct mediation between a participant and their employer to ensure agreed workplace adjustments were being implemented and to resolve any potential disputes or misunderstandings.

4.2.6. Work Placements

The original WHP design intended to provide work experience placements to prepare participants for the workplace, help them secure jobs and provide in-work support to help participants sustain their employment. When asked about work experience placements, 91% of key workers in the Wave One provider survey said they had made efforts to set these up with local employers. Among those that had attempted to set up work experience placements, 77% said their efforts to set up these placements had been successful.

In the Wave Two provider survey, 19% of key workers said that at least one of their participants was currently in a work placement, but for the most part this was no more than a quarter of their participants.

These results may seem modest, but it was evident from the deep dive research that key workers did not view work placements as an essential step for everyone on their caseload. If participants were motivated to return directly into work, and if the key worker judged them ready for this, then a placement may not be deemed necessary.

4.2.7. Impact of COVID-19 on nature of support delivered – wellbeing and mental health

Provider staff interviewed in the Wave Two deep dive research perceived an increase in the prevalence of anxiety and depression amongst participants due to the pandemic, and all said they had introduced new and/or adapted support in response. This primarily took the form of:

• an increased key worker focus on wellbeing. Key workers already had a regular cycle of meetings with participants on the programme prior to the pandemic, but the emphasis of these reportedly shifted much more towards how the participant was getting on and whether there was anything the key worker could provide or direct them to, to help them address any personal challenges posed by the pandemic. Especially early on in the pandemic, as many employers were freezing recruitment and furloughing or laying off existing staff, key workers said this took precedence over conversations with participants about entering work.
• additional specialist support and resources around wellbeing, anxiety and confidence. This included: increased staff resource for the in-house health teams as some providers had to deliver well-being and mental health-related support; new apps and online resources that participants could access for free; adaptations to the existing support their subcontracted providers delivered to have a greater focus on confidence building and managing anxiety; and commissioning additional new specialist provision with a similar focus. Provider staff highlighted particular support needs around psychologically preparing participants, who may have been isolated at home for an extended period, for attending face-to-face job interviews and starting in a new workplace.

However, providers perceived there was a limit to the role they could play in dealing with mental health issues. This was still perceived to apply in the pandemic. The wellbeing check-ups and additional support were consequently aimed at the mild-moderate end of the mental health scale. Key workers said they were active in trying to help participants with more serious mental health needs access local public provision and mental health charities but highlighted long waiting lists.

4.3. Overall satisfaction with programme

Levels of overall satisfaction were similarly high for both WHP groups at Wave Two, with 84% of mandatory participants and 85% of voluntary participants saying that they were satisfied with the support they had received (see Figure 4.9). Participants in the control group were slightly less likely to be satisfied (81%).

While voluntary participants who were employed at the time of the Wave Two interview were more likely to say they were satisfied (93%) than those not in work (83%), the reverse was true for mandatory participants (in work 78%, not in work 85%).

In both groups, those educated to level three or above were less likely to say they were satisfied than those with a lower level of education (mandatory 79% vs. 88%, voluntary 80% vs. 87%).

Figure 4.9: Overall satisfaction with the support via WHP (Wave Two participants who started receiving support)

Level of satisfaction	Mandatory (LTU)	Voluntary (HD/EAG)	Control
Satisfied	84%	85%	81%
Very satisfied	45%	47%	41%
Fairly satisfied	39%	37%	40%
Fairly dissatisfied	7%	6%	9%
Very dissatisfied	7%	8%	9%
Dissatisfied	14%	14%	18%
Don’t know	–	–	–

CUSTSAT_OVERALLSAT1: Overall how satisfied or dissatisfied are you with the support you have had through […]? Are you…?

Base: Wave Two participants who had started receiving support: Mandatory (1,383), Voluntary (2,057), Control (2,042)

In both waves of the deep dive research, the majority of participants interviewed said they were satisfied with the support they received through the programme. Positive aspects of the provision cited by the interviewees included delivery in a positive and friendly environment, “being treated as an adult” and “not feeling judged”. Several participants had also been on other return to work provision previously and compared the support favourably to that received through the WHP. Specific differentiators included the WHP’s voluntary nature, the flexibility inherent in the programme design, the longer programme duration, the focus on finding work but with additional support as needed, and the extent to which the services received were seen to be tailored to individual need.

The ‘style’ in which the programme was delivered was also important – with key workers being more approachable, “relaxed but still professional”, and the feeling of being cared for and that key workers took a real interest in supporting their clients. As one participant described:

‘They seem to want to get people back into work, while the [other programme] workers could not have cared less.’ Participant

In the minority of cases where dissatisfaction was expressed, these related to the support provided not meeting initial expectations, being insufficiently tailored, or where individuals felt they “had to do all the work”.

4.3.1. Satisfaction with key worker contact and tailoring

Levels of satisfaction with specific aspects of contact with the key worker or Work Coach assigned to them were very similar to those seen overall with the programme, with the vast majority of participants who had started to receive support in each group satisfied with each aspect (see Figure 4.10).

• range of different ways to get in contact (89% mandatory and 85% voluntary)
• frequency of contact (87% mandatory and 88% voluntary)
• availability of Work Coach/key worker (86% each of mandatory and voluntary)
• tailoring of support (84% each of mandatory and voluntary)

Satisfaction (particularly in terms of being very satisfied) was slightly lower among the control group participants, with the exception of satisfaction with tailoring support which was more similar to that reported by mandatory and voluntary participants.

Figure 4.10: Satisfaction with key worker/Work Coach contact (Wave Two participants who started receiving support)

4.10a. The range of different ways you have been able to contact a key worker/Work coach

Level of satisfaction	Mandatory (LTU )	Voluntary (HD/EAG)	Control
Satisfied	89%	89%	84%
Very satisfied	56%	57%	48%
Fairly satisfied	33%	32%	36%
Fairly dissatisfied	5%	4%	7%
Very dissatisfied	4%	4%	6%
Don’t know	3%	3%	3%

4.10b. The frequency of contact you have had with a key worker/Work coach

Level of satisfaction	Mandatory (LTU )	Voluntary (HD/EAG)	Control
Satisfied	87%	88%	84%
Very satisfied	54%	56%	47%
Fairly satisfied	33%	32%	36%
Fairly dissatisfied	6%	5%	7%
Very dissatisfied	5%	4%	6%
Don’t know	3%	3%	4%

4.10c. How available a key worker/Work coach is when you need them

Level of satisfaction	Mandatory (LTU )	Voluntary (HD/EAG)	Control
Satisfied	86%	86%	83%
Very satisfied	54%	54%	49%
Fairly satisfied	31%	33%	35%
Fairly dissatisfied	6%	4%	6%
Very dissatisfied	4%	5%	5%
Don’t know	5%	5%	5%

4.10d. The extent to which key worker/Work coach has tailored support they have offered you to your individual needs

Level of satisfaction	Mandatory (LTU )	Voluntary (HD/EAG)	Control
Satisfied	84%	84%	81%
Very satisfied	51%	51%	49%
Fairly satisfied	33%	32%	32%
Fairly dissatisfied	7%	7%	9%
Very dissatisfied	6%	7%	7%
Don’t know	3%	3%	3%

NEW_COACHSAT1: Are you satisfied or dissatisfied with…

Base: Wave Two participants who had started receiving support: Mandatory (1,383), Voluntary (2,057), Control (2,042)

4.3.2. The impact of COVID-19 on satisfaction

Ad-hoc survey participants were asked how their satisfaction with the support they received through the WHP had changed since the start of the COVID-19 pandemic. Responses were similar for mandatory and voluntary participants: most were at least as satisfied with the support received after the start of the pandemic, with only 15% of mandatory participants and 19% of voluntary participants saying they were less satisfied (Figure 4.11).

Figure 4.11: Change in satisfaction with support received since COVID-19 pandemic began (ad-hoc participant survey)

Level of satisfaction	Mandatory (LTU )	Voluntary (HD/EAG)
More satisfied	25%	26%
A lot more satisfid now	16%	13%
A little more satisfied now	9%	12%
No difference	46%	46%
A little less satisfied now	8%	10%
A lot less satisfied now	7%	9%
Don’t know	13%	9%

CUSTSAT_OVERALLSAT1b: And would you say you are more or less satisfied with the support you have been receiving since the COVID-19 pandemic began compared with the support you received before?

Base: Ad hoc participants: Mandatory (55*), Voluntary (203) *CAUTION: Number of respondents below 100

5. Outcomes

As throughout the report, this chapter primarily presents descriptive data from the Wave Two survey for voluntary and mandatory WHP participants. This was collected from WHP participants and the control group 18-24 months after the point of allocation (i.e., when WHP participants were either towards the end of or having completed the programme). As elsewhere in the report, this descriptive data was compared (where possible) to the control group as a whole, including participants allocated through both the voluntary and mandatory groupings.

To more accurately assess outcomes, additional statistical analysis was used to estimate the impact of the WHP within the voluntary and mandatory allocations. This was done by comparing the outcomes of those offered support through the WHP (‘WHP participants’) against the outcomes of those randomly allocated to the standard offer of support provided by JCP (‘control group’). In other words, the analysis measured the extent to which the WHP led to better or worse outcomes than JCP’s ‘business as usual’. Further details of the statistical impact analysis are provided in Appendix Two.

The impact estimates were based on the Wave Two survey data, excluding any respondents with missing data on key variables^{[footnote 22]}, The data in the descriptive charts did NOT exclude these cases, so the percentages shown in the descriptive and impact charts may differ by one to two percentage points in some cases.

As elsewhere in the report, the findings should be interpreted in the context of the caveats highlighted in Chapter 1.4.

Summary

• evidence from the participant surveys suggests that the WHP has had some positive impacts on participants’ work readiness compared with the control group. This is most evident for voluntary WHP participants relative to their control group.
• voluntary WHP participants were more likely than their control group to have done some work over the period (27% of voluntary WHP participants compared to 22% of voluntary control participants) and to be in work at the Wave Two survey (19% of voluntary WHP participants compared to 16% of voluntary control participants). While there were also positive differences between the mandatory WHP participants and their control group, these did not reach statistical significance.
• the WHP appeared to have a positive impact on participants’ levels of job search activity, with both voluntary and mandatory WHP participants who were looking for work more likely to report applying for more jobs than before than their control groups.
• among both mandatory and voluntary participants, six in ten thought the support received had increased their chances of moving into work.
• although voluntary and mandatory participants were equally positive that the support had increased their chances of moving into work, voluntary WHP participants were more positive compared with those in the control group. This was not true for the mandatory participants for whom the control group were equally positive about JCP support as the mandatory WHP participants were about the WHP.
• participants in the Wave Two survey were asked about their level of confidence in four interpersonal skills. Levels of confidence were high across all groups. There was no statistically significant evidence among either voluntary or mandatory participants to suggest that the WHP had an impact on participants’ interpersonal skills.

5.1. Employment outcomes

5.1.1. Moving into paid work

Questions about applying for and gaining paid work since being referred to the programme were used at Wave Two of the Participants Survey to assess key employment outcomes of the programme.

Voluntary participants were more likely to report having been employed at any point since referral (26%) than mandatory participants (17%) (Figure 5.1), and also to report working longer hours, with more employed full time (11% vs. 7%) or 16 to 30 hours a week (9% vs. 5%). A further one percent of each group said they had been self-employed.

Among mandatory participants, those with a physical health condition were less likely to report having been employed (14%) than those with no health conditions (20%), and to report having entered full time employment (4% vs. 9%).

While there was no difference in employment levels overall by gender, in both groups men were relatively more likely to report full-time employment, and women to report part-time employment. Participants aged 50 or over were less likely to report being employed at all than those aged under 50 (mandatory 10% vs. 24%, voluntary 19% vs. 32%).

Education also appeared to have played a role, with those with the lowest level qualifications relatively less likely to report having been employed in both groups.

Figure 5.1: Whether started paid work since being referred to WHP (Wave Two participants survey)

Work after being referred to WHP	Mandatory (LTU )	Voluntary (HD/EAG)
Employed	17%	26%
Employed FT – 30 hours + a week	7%	11%
Employed PT – 16-30 hours a week	5%	9%
Employed PT – <16 hours a week	6%	6%
Self-employed	1%	1%
Not been employed since referral	81%	73%

TRAINEE_ECONOUT: Since being referred to the programme have you started paid work at any time?

Base: All Wave Two participants asked: Mandatory (1,348), Voluntary (2,037)

Not all of those who had been in employment were still employed at the time of the Wave Two interview. As shown in Figure 5.2, at the time of interview, 11% of mandatory participants and 17% of voluntary participants were employed at the time of the Wave Two interview (with each slightly more part time than full time on balance), with a further one percent of each group self-employed. Mandatory participants were more likely to say they were focused primarily on looking for work (52%) compared with voluntary participants (37%), with voluntary participants (as expected) more likely to be coping with a long-term condition or disability (22% vs. 15%). Around one in ten of each group said they were not looking for work – among mandatory participants this was relatively higher for, although not restricted to, those with a health condition (13%). Among voluntary participants this was relatively higher for those unemployed for 10 years or more (16%).

Differences in employment at the time of interview by age and education were similar to those seen for any employment since referral. Most notably, those aged 50 and over and those with lower-level qualifications were less likely to report being in employment at the time of interview.

By gender, while women were more likely than men to be in part-time work in both groups, among voluntary participants women were more likely to be currently employed (20%) than men (14%) with men more likely to be focused on looking for work (42% vs. 28%).

Among mandatory participants, those with no health conditions were more likely to be in employment (14%) compared with those with any health condition (9%) and they were also more likely to be primarily focused on finding work (64% vs. 43%); those with a health condition were more likely to report coping with a long-term condition instead.

Figure 5.2: Current employment status (Wave Two participants survey)

Employment status	Mandatory (LTU )	Voluntary (HD/EAG)
Employed	11%	17%
Employed full-time	5%	7%
Employed part-time	7%	9%
Self-employed	1%	1%
Focused primarily on looking for work	52%	37%
Unemployed and not looking for work	8%	10%
Coping with a long-term condition or disability	15%	22%
Caring for children or other people	4%	4%
Other/DK	8%	10%

TRAINEE_ECONACT: Can I check, which of the following best describes your current employment status? Base: All Wave Two participants: Mandatory (1,414), Voluntary (2,111)

In order to assess likely impact, Figure 5.3 compares the percentages of voluntary and mandatory WHP participants and their control groups who had done any work since referral to the WHP and the percentages in work at the time of the Wave Two survey.

The WHP had a statistically significant positive impact on voluntary participants’ entry into work. Voluntary WHP participants were statistically significantly more likely than their control group to have done some work over the period (27% compared to 22%)^{[footnote 23]} and to be in work at the Wave Two survey (19% compared to 16%)^{[footnote 24]}.

While there were also positive differences between the mandatory WHP participants and their control group, these did not reach statistical significance.

Figure 5.3 Impact of WHP on entry to employment (Wave Two participants survey)

Voluntary participants

Employment status	WHP participants	Control group	P-value
Worked at all	27%	22%	0.002*
Currently in work	19%	16%	0.049*

Mandatory participants

Employment status	WHP participants	Control group	P-value
Worked at all	19%	17%	0.621*
Currently in work	13%	9%	0.188*

*Statistically significant

TRAINEE_ECONOUT: Since being referred to the programme have you started paid work at any time?

TRAINEE_ECONACT: Can I check, which of the following best describes your current employment status?

Bases: Voluntary WHP participants (2037); Voluntary control group (1600): Mandatory WHP participants (1348); Mandatory control group (354)

5.1.2. Suitability of employment

At Wave Two, perceptions among those in employment about the suitability of their job were similar between the two groups of participants in terms of meeting their criteria and fitting in with commitments, with mandatory participants more positive about opportunities for progression (see Figure 5.4).

• 71% of mandatory and 67% of voluntary participants felt their job mostly met the criteria they were looking for
• 61% of mandatory and 56% of voluntary participants felt their job mostly fitted in with any commitments or health conditions they may have
• 60% of mandatory and 47% of voluntary participants felt their job mostly offered opportunities for progression if they wanted it.

Figure 5.4: Perceived suitability of current job (Wave Two participants currently in work)

5.4a. Meets the criteria you were looking for

Suitability of job	Mandatory	Voluntary
Mostly	71%	67%
Partly	17%	21%
Not at all	8%	11%
Don’t know	3%	1%

5.4b. Fits in with any commitments or health conditions you may have

Suitability of job	Mandatory	Voluntary
Mostly	61%	56%
Partly	13%	21%
Not at all	22%	17%
Don’t know	5%	6%

5.4c. Offers you opportunities for progression if you want it

Suitability of job	Mandatory	Voluntary
Mostly	60%	47%
Partly	19%	25%
Not at all	17%	22%
Don’t know	4%	5%

JCPOFFER_FJOBMERITS1: Do you believe your job…

Base: Wave Two participants currently in work: Mandatory (183), Voluntary (374)

However, there was no evidence that WHP support improved the suitability of the jobs they entered, based on the percentages saying their job were mostly or partly suitable on these measures. The majority of those in work, among both voluntary WHP participants and their control group, were positive about their jobs, with no statistically significant differences between the two groups. (Figure 5.5)^{[footnote 25]}.

Figure 5.5: Impact of WHP on perceived suitability of work (Wave Two participants survey – % saying mostly or partly)

Voluntary participants

Suitability of work	WHP participants	Control group	p-value
Meets the criteria you were looking for	89%	88%	0.859*
Fits in with any commitments or health conditions you may have	78%	76%	0.736*
Offers you opportunities for progression if you want it	72%	72%	0.968*

*Statistically significant

JCPOFFER_FJOBMERITS1: Do you believe your job…

Bases: Voluntary WHP participants in work (361); Voluntary control group in work (214)

5.1.3. Perceived impact of support on getting work

In answer to a separate question, 59% of mandatory participants and 67% of voluntary participants who were in work at the time of interview, said that the support they had received through the WHP had helped them to get their current job. There was no significant difference between the two groups on this measure.

The statistical impact analysis showed that voluntary WHP participants who were in work at the Wave Two survey were far more likely than their control group to feel that the support they received had helped them get that work (68% compared to 48% - Figure 5.6)^{[footnote 26]}.

Likewise, they were significantly more likely than their control group to report receiving in-work support. Twice as many voluntary WHP participants reported being offered in-work support compared to their control group (40% compared to 21%).

Figure 5.6: Impact of WHP support on finding work (Wave Two participants survey)

Voluntary participants

Support on finding work	WHP participants	Control group	p-value
Support helped get job	68%	48%	<0.001*
Received in-work support	40%	21%	<0.001*

*Statistically significant

YOUTHCONT_QIMPJOB1: Do you feel the support through [the programme/Jobcentre Plus] helped you to get this?

WP_H2: Have you received any in-work support from a [key worker/Work Coach] since you began your current job?

Bases: Voluntary WHP participants in work (361); Voluntary control group in work (214).

5.1.4. Impact of COVID-19 on employment outcomes

Labour market dynamics fluctuated significantly during the pandemic, with obvious implications for the work placement and job opportunities potentially available to programme participants. In the deep dive research, provider staff highlighted concerns early on in the pandemic that not only would there be fewer job opportunities for participants but that they would also face increased competition for these from newly unemployed people, without the same barriers.

‘Due to COVID-19 and the number of people looking for work, WHP participants will be further disadvantaged as they will not be as competitive in the current job market. If you have a choice of employing a highly skilled, healthy reliable person over someone who may be more vulnerable due to the virus due to health and who has not worked for a while who would you choose?’ Key worker

Physical work experience placements were no longer typically an option, although some positive examples were reported of participants being able to undertake virtual/remote placements with an employer.

By the time of the Wave Two deep dive research in late 2021, providers reported almost unprecedented volumes of job vacancies in their local area. Growth in sectors such as health and social care, transportation, COVID-19-related opportunities such as NHS Test and Trace were reported earlier in the pandemic and latterly sectors such as hospitality, manufacturing and retail were reported to have bounced back strongly.

This was cited by providers as a significant facilitator for achieving more employment outcomes with participants. Despite the earlier concerns there had been about increased competition for job opportunities from other newly unemployed people, providers said they were achieving more job starts than at any point since the start of the programme. Equally, some did caveat this by emphasising that not all of these new job opportunities were necessarily desirable or accessible to every WHP participant. Again, those without IT skills were perceived to be potentially disadvantaged, both in the job application stage (which employers increasingly conduct online) and in terms of the nature of the job roles themselves (which are increasingly IT based).

5.2. Job search outcomes

5.2.1. Job search activities

As shown in Figure 5.7, 21% of mandatory and 23% of voluntary participants who were still looking for work at the time of the Wave Two interview said they were sending out more job applications than before the programme, with 41% of mandatory and 44% of voluntary participants reporting sending out fewer than before the scheme.

The under-50s and those with no health condition were relatively more likely to report increasing the number of applications. Similarly, those out of work for less than a year or who had never been in work were also relatively more likely to report an increase.

Mandatory participants with any health condition were more likely to say they were sending out fewer applications compared with those without such a condition (45% vs. 36%).

Figure 5.7: How many job applications sending compared with before the programme (Wave Two participants looking for work)

Number of job applications	Mandatory (LTU )	Voluntary (HD/EAG)
More	21%	23%
The same	35%	30%
Less	41%	44%
Don’t know	4%	3%

YOUTHCONT_QAPP3: On average, how many job applications per week are you sending out now compared to before the programme? Would you say you are sending out…

Base: Wave Two participants looking for work: Mandatory (913), Voluntary (1,082)

Comparing WHP participants with the control groups, the WHP appeared to have a positive impact on participants’ levels of job search activity, with both voluntary and mandatory WHP participants who were looking for work more likely to report applying for more jobs than before than their control groups. Among the voluntary group, the difference between WHP participants and their control group is statistically significant (23% compared to 17%) (see Figure 5.8). Table 10 in Appendix Five shows that voluntary WHP participants were more likely than their control group to report applying for more or the same number of jobs as before, while the control group were more likely than WHP participants to report applying for fewer jobs than before (53 per cent of the control group compared to 44 per cent)^{[footnote 27]}. While the pattern of results was very similar for mandatory participants, it did not reach statistical significance, almost certainly because of the smaller sample size. Table 10 in Appendix Five shows that mandatory WHP participants were more likely than their control group to report applying for more or the same number of jobs as before, while the control group were more likely than WHP participants to report applying for fewer jobs than before (52 per cent of the control group compared to 40 per cent) (p-value: 0.064).

Figure 5.8: Impact of WHP on job search activity (Wave Two participants survey)

Level of job search activity	WHP participants	Control group	p-value
Voluntary participants: applying for more jobs than before	23%	17%	0.012*
Mandatory participants: applying for more jobs than before	21%	16%	0.158

*Statistically significant

YOUTHCONT_QAPP3: On average, how many job applications per week are you sending out now compared to before the programme? Would you say you are sending out…

Base: WHP participants looking for work: Voluntary (1043), Mandatory (870); Control group looking for work: Voluntary (866) Mandatory (1,168)

5.2.2. Participant perceptions and attitudes to finding work

Most participants who were not in work at the time of the Wave Two survey indicated that they would like paid work in the future. On scale of 0 to 10 where 0 was not at all and 10 was a lot, more than half of either group indicated a strong desire to move into work by selecting scores between 8-10 (a lot). Half of participants selected the top point of the scale.

There was a relatively weaker desire to move into paid work in both groups for participants with a health condition, those aged 60 and over, those with a qualification below level 1 and those who have been out of work for ten years or more.

Figure 5.9: Extent to which would like paid work in future (Wave Two participants not currently in work)

Extent to which would like work	Mandatory (LTU)	Voluntary (DG/EAG)
Already in work	13%	18%
8 - 10 (high)	63%	56%
4 - 7 (medium)	17%	19%
0 - 3 (low)	17%	19%
Don’t know	2%	3%

ESASUPP_WORKCONF2: On a scale of 0-10, where 10 is a lot and 0 is not at all, to what extent would you like to undertake paid work in the future?

Base: All Wave Two participants: Mandatory (1,414), Voluntary (2,111)

In order to measure the impact of WHP on all participants’ desire to work, non-working participants responses were grouped into ‘zero to three’ (low), ‘four to seven’ (medium), ‘eight to ten’ (high), and also included those already in work. There was no statistically significant evidence that the WHP has an impact on how much participants would like to be in work (see Figure 5.10).

Figure 5.10: Impact of WHP on wanting to undertake paid work in the future (Wave Two participants survey)

Voluntary

Impact wanting paid work	WHP participants	Control group
Already in work	19%	16%
8 - 10 (high)	55%	58%
4 - 7 (medium)	19%	19%
0 - 3 (low)	5%	5%

p-value: 0.191

Mandatory

Impact wanting paid work	WHP participants	Control group
Already in work	13%	9%
8 - 10 (high)	63%	67%
4 - 7 (medium)	17%	18%
0 - 3 (low)	6%	2%

p-value: 0.064

ESASUPP_WORKCONF2: On a scale of 0-10, where 10 is a lot and 0 is not at all, to what extent would you like to undertake paid work in the future?

Base: Voluntary WHP participants (2037); Voluntary control group (1600); Mandatory WHP participants (1348); Mandatory control group (354)

5.3. Self-efficacy outcomes

Among both voluntary and mandatory participants, there was no statistically significant evidence to suggest that the WHP had an impact on participants’ interpersonal or job application skills. However, there was evidence that the WHP had a positive impact on self-reported confidence finding a job and increased motivation to find work among voluntary participants. There was no significant evidence of the same impact among mandatory participants.

5.3.1. Confidence in interpersonal and job application skills

Key self-efficacy outcomes at Wave Two of the participants’ survey included levels of confidence in a range of interpersonal and job application skills.

Participants in the Wave Two survey were asked about their level of confidence in four interpersonal skills. While the majority of participants were confident on all four measures, levels of confidence were higher in each of the skills among mandatory participants compared to voluntary participants (see Figure 5.11):

• 85% of mandatory and 80% of voluntary participants were confident at working in a team with other people
• 80% of mandatory and 74% of voluntary participants were confident at having a go at things that are new to them
• 78% of mandatory and 70% of voluntary participants were confident at meeting new people
• 78% of mandatory and 70% of voluntary participants were confident at putting forward their ideas

Across both groups, confidence in these skills remained broadly consistent between the two waves of fieldwork, suggesting that confidence in interpersonal skills was not a strong proxy for moving closer to work (Figure 5.11).

Among both mandatory and voluntary participants, confidence levels were generally higher (usually significantly more saying ‘very confident’) across all statements among men, ethnic minorities (excluding white minorities), those without any health condition or disability, those with a higher level of education and those currently in employment. While all differences were significant, the difference tended to be largest between those with and without health conditions, with this reflected in the voluntary group in higher confidence for EAG than Disability group participants (with a difference of around 10 percentage points on each of the four issues).

Gender differences were more pronounced among mandatory participants than voluntary participants, while differences by ethnicity were similar across the two groups.

Figure 5.11: Confidence in interpersonal skills (Wave Two participants survey)

Participant group	Confident	Very confident	Fairly confident	Not very confident	Not at all confident
Mandatory: Working with other people in a team	85%	49%	36%	10%	5%
Voluntary: Working with other people in a team	80%	40%	40%	14%	6%
Mandatory: Having a go at new things	80%	34%	46%	14%	6%
Voluntary: Having a go at new things	74%	26%	48%	20%	6%
Mandatory: Meeting new people	78%	39%	39%	20%	6%
Voluntary: Meeting new people	70%	31%	39%	17%	9%
Mandatory: Putting forward their ideas	78%	33%	44%	15%	8%
Voluntary: Putting forward their ideas	70%	26%	44%	22%	8%

WORK_READY1: How do you feel about the following things even if you have never done them before…

Base: All Wave Two participants: Mandatory (1,414), Voluntary (2,111)

Figure 5.12 shows that, among both voluntary and mandatory participants, there was no statistically significant evidence to suggest that the WHP had an impact on participants’ interpersonal skills^{[footnote 28]}.

Figure 5.12: Impact of WHP on interpersonal skills (Wave Two participants survey)

Voluntary participants

Skills	WHP participants	Control group	p-value*
Working with others in a team	80%	78%	0.304
Having a go at new things	74%	74%	0.964
Meeting new people	70%	71%	0.617
Putting forward ideas	70%	70%	0.981

Mandatory participants

Skills	WHP participants	Control group	p-value*
Working with others in a team	85%	83%	0.562
Having a go at new things	79%	82%	0.390
Meeting new people	77%	78%	0.884
Putting forward ideas	78%	77%	0.681

*Statistically significant

WORK_READY1: How do you feel about the following things even if you have never done them before… (% confident)

Base: Voluntary WHP participants (2037); Voluntary control group (1600: Mandatory WHP participants (1348); Mandatory control group (354)

Participants in the Wave Two survey were also asked about their level of confidence in three different skills relating to job applications. As seen in relation to interpersonal skills, while the majority of participants were confident in all three skills, levels of confidence tended to be higher among mandatory participants than voluntary participants (Figure 5.13):

• 79% of mandatory and 72% of voluntary participants were confident in completing a job application or CV
• 77% of mandatory and 73% of voluntary participants were confident in making a good impression in a job interview, and
• 74% of mandatory and 70% of voluntary participants were confident int applying for jobs online Figure 5.13 shows that across both groups, confidence in these skills remained consistent between the two waves of fieldwork, suggesting that confidence in job application skills was not a strong proxy for moving closer to work.

As for interpersonal skills, levels of confidence were generally higher for all three job application skills among men, those with no health condition and those who had been out of paid work for less time (particularly those who had been out of work for less than three years). Among mandatory participants, the differences tended to be biggest in relation to health (15 to 20 percentage points) but with slightly smaller differences by health condition among voluntary respondents, meaning there was much less difference in confidence on these issues between Disability group and EAG participants, compared with the differences seen for interpersonal skills.

In addition, in both groups, those with level one qualifications or above were more likely to be very confident than those with lower-level qualifications in completing applications and applying for jobs online. Confidence in applying online was also higher for the under-50s than for those aged 50 and over (mandatory 50% vs. 33%, voluntary 41% vs. 28%).

Figure 5.13: Confidence in job application skills (All Wave Two participants)

Participant group	Confident	Very confident	Fairly confident	Not very confident	Not at all confident
Mandatory: completing a job application and CV	79%	39%	40%	14%	7%
Voluntary: completing a job application and CV	72%	32%	40%	19%	9%
Mandatory: making a good impression in an interview	77%	31%	46%	16%	7%
Voluntary: making a good impression in an interview	73%	28%	45%	20%	7%
Mandatory: applying for jobs online	74%	42%	33%	15%	11%
Voluntary: applying for jobs online	70%	35%	34%	17%	13%

WORK_READY2: How confident do you feel about doing the following things?

Base: All Wave Two participants: Mandatory (1,414), Voluntary (2,111)

Figure 5.14 shows that, among both voluntary and mandatory participants, there was no statistically significant evidence to suggest that the WHP had an impact on participants’ confidence in their job application skills^{[footnote 29]}.

Figure 5.14: Impact of WHP on job application skills (Wave Two participants survey)

Voluntary participants

Impact on job application	WHP participants	Control group	p-value*
Making good impression at interview	73%	72%	0.493
Completing job applications/CV	72%	72%	0.783
Applying for jobs online	69%	70%	0.486

Mandatory participants

Skills	WHP participants	Control group	p-value*
Making good impression at interview	77%	75%	0.480
Completing job applications/CV	79%	80%	0.675
Applying for jobs online	74%	78%	0.243

*Statistically significant

WORK_READY2: How confident do you feel about doing the following things? (% confident)

Base: Voluntary WHP participants (2037); Voluntary control group (1600: Mandatory WHP participants (1348); Mandatory control group (354)

5.3.2. Participant perceptions of impact on confidence and motivation to find work

Participant perceptions of the impact of advice and support were explored in the Wave Two survey by asking participants whether it had helped them with:

• building confidence about finding a job they could do
• increasing motivation to find work, and
• increasing chances of finding suitable work (suitable work was further explained as work which accommodates any health or caring requirements they have and reflects the type of work they wish to do long term).

At Wave Two, over two thirds of both mandatory and voluntary participants who had received support said that it had helped to do each of these things (see Figure 5.15).

In both groups, those without a health condition were consistently more likely than those with a health condition to say the support helped them with all three issues.

Figure 5.15: Whether advice and support has helped … (Wave Two participants who started receiving support)

Advice/support helped	Yes	No	Hard to say/Don’t know	Not applicable
Mandatory: to build confidence about finding a job you could do	69%	27%	3%	–
Voluntary: to build confidence about finding a job you could do	68%	28%	3%	–
Mandatory: to increase motivation to find work	71%	25%	2%	–
Voluntary: to increase motivation to find work	69%	27%	2%	–
Mandatory: to increase chances of finding suitable work	67%	28%	4%	–
Voluntary: to increase chances of finding suitable work	67%	28%	3%	–

JCPOFFER_SOFTOUTCOME: Do you believe that the advice and support from […] has helped …

Base: Wave Two participants who had started receiving support: Mandatory (1,383), Voluntary (2,057)

Participants aged under 30 were also consistently more likely to agree support helped with each of the issues, particularly compared with those aged 50 and over. This difference was more marked among mandatory than voluntary participants, with a difference of around 20 percentage points between the youngest and oldest participants on each issue. Among mandatory participants, those with level one qualifications were more likely to agree support helped with each of the issues compared with those with level three qualifications or above (see Table 3). While there were few gender differences, female voluntary participants were more likely to say support had built their confidence (71%) compared with males (65%).

Table 3: Whether advice and support has helped … (Wave Two mandatory participants who started receiving support, by age)

Advice and support helped to…	Up to 29 years old	30-39 years old	40-49 years old	50-59 years old	60+ years old
…build confidence finding a job that you could do	78%	70%	65%	73%	61%
…increase motivation to find work	81%	70%	69%	73%	63%
…increase your chances of finding suitable work	78%	68%	66%	68%	57%
Base	186	220	267	379	239

The self-reported impacts among voluntary participants on all three aspects were statistically significantly better than the control group (see Figure 5.16).

• 68%of voluntary WHP participants felt that the support had built their confidence about finding a job that they could do, compared to 61% of their control group.
• 69%of voluntary WHP participants said that it had increased their motivation to find work, compared to 63% of their control group.
• 67% of voluntary WHP participants said that it had increased their chances of finding suitable work, compared to 61% of their control group. Conversely, the mandatory WHP participants’ ratings of WHP support were not statistically significantly different to those of their control group across any of the outcomes.

Figure 5.16: Impact of WHP on perceptions of support increasing confidence and motivation (Wave Two participants survey)

Voluntary participants

Impact on confidence/motivation	WHP participants	Control group	p-value*
Built confidence that could find job	68%	61%	<0.001*
Increase motivation to find work	69%	63%	0.002*
Increased chances of finding suitable job	67%	61%	0.001*

Mandatory participants

Impact on confidence/motivation	WHP participants	Control group	p-value*
Making good impression at interview	69%	69%	0.935
Increase motivation to find work	71%	74%	0.345
Increased chances of finding suitable job	67%	70%	0.337

*Statistically significant

JCPOFFER_SOFTOUTCOME: Do you believe that the advice and support from […] has helped …

Base: Voluntary WHP participants (2037); Voluntary control group (1600); Mandatory WHP participants (1348), Mandatory control group (354)

Among both mandatory and voluntary participants, 61% of mandatory and 60% of voluntary participants thought the support received had increased their chances of moving into work (Figure 5.17).

As might be expected, participants in both groups who were currently in work were more likely to say it had increased their chance by a large extent than those not currently in work (mandatory 44% vs. 18%, voluntary 45% vs. 17%). In addition, mandatory participants aged under 50 were more likely to say it had helped to a large extent than those aged 50 or over (24% vs. 17%), while there was a similar difference among voluntary respondents in terms of saying it had helped at all (66% vs. 55%).

Among both mandatory and voluntary participants, those with a health condition were less likely to say it had helped even to some extent compared with those with no health condition (mandatory 55% vs. 67%, voluntary 58% vs. 69%).

Figure 5.17: Wave Two Perceptions of support increasing chances of moving into work

Increase chances of getting work	Mandatory (LTU)	Voluntary (HD/EAG)
Yes	61%	60%
Yes – to large extent	21%	22%
Yes – to some extent	40%	39%
No	35%	34%
No – not really	19%	18%
No – not at all	16%	16%
Don’t know	3%	3%
Not applicable	–	–

JCPOFFER_USEFUL1: And do you feel the support you received through […] has increased your chances of moving into work?

Base: Wave Two participants who had started receiving support: Mandatory (1,383), Voluntary (2,057)

Although voluntary and mandatory participants were equally positive that the support had increased their chances of moving into work, voluntary WHP participants were statistically significantly more positive compared with the control group (61% of WHP participants and 53% of the control group, see Figure 5.18). This was not true for the mandatory participants for whom the control group were equally positive about JCP support as the mandatory WHP participants were about the WHP.

Figure 5.18: Impact of WHP on perceptions of support increasing chances of moving into work (Wave Two participants survey)

Support increasing chances of work	WHP participants	Control group	p-value*
Voluntary participants: increase chances of moving into work	61%	53%	<0.001*
Mandatory participants: increase chances of moving into work	60%	61%	0.841

*Statistically significant

JCPOFFER_USEFUL1: And do you feel the support you received through […] has increased your chances of moving into work? (% yes)

Base: Voluntary WHP participants (2037); Voluntary control group (1600) Mandatory WHP participants (1348); Mandatory control group (354)

Among the 27% of mandatory participants and 21% of voluntary participants who had received a work experience or volunteering placement (see Figure 4.3), opinions on whether this had increased their chances of moving into work were similar for both groups. Around three quarters said that their placement had increased their chances of moving into work (mandatory 75%, voluntary 73%) (see Figure 5.19).

Those with any health condition were less likely to say it had helped to a large extent in both groups compared with those without such a condition (mandatory 26% vs. 38%, voluntary 30% vs. 44%).

Figure 5.19: Whether work experience or volunteering increased chances of moving into work (Wave Two participants who have had work experience or volunteering placement)

Experience/volunteering increase chances of moving into work	Mandatory (LTU)	Voluntary (HD/EAG)
Yes	75%	73%
Yes – to large extent	31%	34%
Yes – to some extent	45%	40%
No	18%	17%
No – not really	10%	12%
No – not at all	8%	5%
Don’t know	2%	4%
Not applicable	5%	5%

JCPOFFER_USEFUL2: And thinking specifically about your work experience or volunteering placement, do you feel it has increased your chances of moving into work?

Base: Wave Two participants who had work experience or volunteering placement: Mandatory (371), Voluntary (445)

Figure 5.20 shows no evidence of WHP placements having greater impact on participants’ perceptions of their chances of moving into work than placements arranged via JCP.

Figure 5.20: Impact of WHP on perceptions of doing voluntary or work placements increasing chance of moving into work (Wave Two participants survey)

Voluntary/work placements increase chances of moving into work	WHP participants	Control group	p-value*
Voluntary participants: increase chances of moving into work	74%	74%	<0.854*
Mandatory participants: increase chances of moving into work	75%	75%	0.995

*Statistically significant

JCPOFFER_USEFUL2: And thinking specifically about your work experience or volunteering placement, do you feel it has increased your chances of moving into work? (% yes)

Base: Voluntary WHP participants who did work/voluntary placement (429); Voluntary control group who did work/voluntary placement (396) Mandatory WHP participants who did work/voluntary placement (351); Mandatory control group who did work/voluntary placement (94)

5.4. Key wellbeing outcomes

There was some, limited, evidence that the WHP had a positive effect on the wellbeing of voluntary participants. Voluntary WHP participants were statistically significantly more likely than the control group to have a high level of life satisfaction, however the differences for the other wellbeing measures did not meet the threshold for statistical significance. There was no statistically significant evidence to suggest that the WHP had a positive impact on the wellbeing of mandatory participants.

5.4.1. General wellbeing

Key wellbeing outcomes at Wave Two of the participants survey included the four Office for National Statistics personal wellbeing questions (standardised questions that have been used by the ONS to track wellbeing since 2011), self-assessed health and self-reported use of the health service in the past year.

To provide context for these findings, Figure 5.21 and Figure 5.22 show both the survey findings and the ONS mean scores^{[footnote 30]} for the UK population covering the same period as the Wave Two survey fieldwork (Oct 2020 to Apr 2021). For analysis purposes (and in line with the ONS measures these questions replicate) scores are divided into high (seven-ten – those positive about wellbeing), medium (four-six – those neutral about wellbeing) or low (zero-three – those negative about wellbeing). For these three items, ‘high’ indicates a positive score.

The UK-wide personal wellbeing scores (from ONS) in October to December 2020 were among the most negative recorded in the previous decade, driven by COVID-19, but they had recovered to pre-pandemic levels by April to June 2021 at the end of the second wave of COVID-19. The mean scores among WHP participants were less positive on all four measures than those seen for the UK population across this period.

Figure 5.21: ONS personal wellbeing question scores (Wave Two participants survey)

Participation group	Mean score	High 7-10	Medium 4-6	Low 0-3	Prefer not to say	UK Q4 2020*	UK Q4 2021*
Mandatory: satisfaction with life	6.2	49%	34%	16%	–	7.3	7.6
Voluntary: satisfaction with life	5.7	42%	36%	20%	–	7.3	7.6
Mandatory: how worthwhile things they do are	6.8	58%	29%	10%	–	7.7	7.8
Voluntary: how worthwhile things they do are	6.3	51%	30%	16%	–	7.7	7.8
Mandatory: how happy felt yesterday	6.5	55%	27%	17%	–	7.2	7.5
Voluntary: how happy felt yesterday	6.0	47%	31%	20%	–	7.2	7.5

ONS_4_1/2/3: Overall, how satisfied are you with your life nowadays? Overall, to what extent do you feel the things you do in your life are worthwhile? Overall, how happy did you feel yesterday? (ALL asked on a scale of 0 to 10, where 0 is not at all and 10 is completely)

Base: All Wave Two participants: Mandatory (1,414), Voluntary (2,111)

*UK mean scores taken from ONS estimates from Annual Population Survey

For these three areas (life satisfaction, feeling worthwhile and happiness), scores were consistently lower for voluntary participants than for mandatory participants. For the first three areas (life satisfaction, feeling worthwhile and happiness), voluntary participants were less likely to give a high score and more likely to give a low score, with this reflected in the differences in mean scores summarised in Figure 5.21. Additionally, participants who were out of work were more likely to give lower scores than those in employment, as were those in the Early Access Group (EAG) compared with those in other groups.

Figure 5.22 shows WHP participants’ perceptions of their level of anxiety, again compared with figures from the ONS Annual Population Survey. In this instance ‘high’ indicates a negative reaction – a high level of anxiety indicated by giving a score of six - ten, more moderate/neutral levels are considered four - five and low levels of anxiety zero - three. Levels of anxiety among the WHP participants were higher than those seen among the population in general, however, compared to the wellbeing measures, scores for anxiety were more similar in each group^{[footnote 31]}.

Participants reporting a long-term health condition, particularly those with a mental health condition, answered more negatively on all four ONS measures than those with no long-term health condition.

Figure 5.22: ONS personal anxiety question scores (Wave Two participants survey)

	Mean score	High 6-10	Medium 4-5	Low 0-3	Prefer not to say	UK Q4 2020*	UK Q4 2021*
Mandatory: how anxious felt yesterday	4.3	37%	20%	41%	–	3.4	3.0
Voluntary: how anxious felt yesterday	4.5	37%	24%	38%	–	3.4	3.0

ONS_4_4: Overall, how anxious did you feel yesterday? (ALL asked on a scale of 0 to 10, where 0 is not at all and 10 is completely)

Base: All Wave Two participants: Mandatory (1,414), Voluntary (2,111)

*UK mean scores taken from ONS estimates from Annual Population Survey

Figure 5.23 shows the percentage of WHP participants and the control group with a ‘high’ score (a score of seven or more on satisfaction, feeling worthwhile and happiness, and six or more for anxiety) and examines impact. For the first three items, ‘high’ indicates a positive score, while for anxiety ‘high’ indicates a negative score.

There was some, limited, evidence that the WHP had a positive effect on the wellbeing of voluntary participants. Voluntary WHP participants were statistically significantly more likely than the control group to have a high level of life satisfaction (42% compared to 37%)^{[footnote 32]}. However, the mean score difference between the participants and control group (see Table 14, Appendix Five) was not statistically significant. Moreover, while the percentage point and mean score differences for the other three measures between voluntary WHP participants and their control group were all positive, they did not reach the threshold for statistical significance.

There was no statistically significant evidence to suggest that the WHP had a positive impact on the wellbeing of mandatory participants.

Figure 5.23: Impact of WHP on life satisfaction and wellbeing (Wave Two participants survey)

Voluntary participants

Satisfaction and wellbeing	WHP participants	Control group	p-value*
Satisfied with life 7+	42%	37%	0.010*
Life is worthwhile 7+	51%	48%	0.090
Happy yesterday 7+	46%	47%	0.531
Anxious yesterday 6+	37%	39%	0.437

Mandatory participants

Satisfaction and wellbeing	WHP participants	Control group	p-value*
Satisfied with life 7+	48%	51%	0.434
Life is worthwhile 7+	58%	60%	0.535
Happy yesterday 7+	55%	56%	0.806
Anxious yesterday 6+	37%	37%	0.853

*Statistically significant

ONS_4_1/2/3/4: Overall, how satisfied are you with your life nowadays? Overall, to what extent do you feel the things you do in your life are worthwhile? Overall, how happy did you feel yesterday? Overall, how anxious did you feel yesterday? (ALL asked on a scale of 0 to 10, where 0 is not at all and 10 is completely)

Base: Voluntary WHP participants (2037); Voluntary control group (1600) Mandatory WHP participants (1348); Mandatory control group (354)

5.4.2. Perceived health benefits of being in work

Participants in both groups who were working at the time of the interview were more likely to say that being in work benefited their mental well-being than their physical health (Figure 5.24); Among mandatory participants 68% felt that work was mostly good for their mental health and 58% felt that work was mostly good for their physical health. Among voluntary participants 70% felt that felt that work was mostly good for their mental health and 50% felt that work was mostly good for their physical health.

Small numbers of respondents limit the scope for further analysis.

Figure 5.24: Perceived benefit of work on mental and physical health (Wave Two participants currently in work)

Effect of work on mental/physical health	Mostly	Partly	Not at all	Don’t know
Mandatory: has benefited your mental wellbeing	69%	27%	3%	–
Voluntary: has benefited your mental wellbeing	68%	28%	3%	–
Mandatory: has benefited your physical health	71%	25%	2%	–
Voluntary: has benefited your physical health	69%	27%	2%	–

JCPOFFER_FJOBMERITS1: Do you believe your job…

Base: Wave Two participants currently in work: Mandatory (183), Voluntary (374)

Despite the preceding discussion, there was no evidence that WHP support improved the perceived impact of working on mental or physical wellbeing. Voluntary group WHP participants who were in work at the Wave Two survey were no more likely to report positively on these measures than those in the control group (Figure 5.25)^{[footnote 33]}.

Figure 5.25: Whether current job has benefitted mental and physical health at all (wave participants survey)

Voluntary participants	WHP participants	Control group	p-value*
Benefitting physical health	76%	75%	0.934
Benefitting mental health	88%	89%	0.792

*Statistically significant

JCPOFFER_FJOBMERITS1: Do you believe your job…

Base: Voluntary WHP participants currently in work (361); Voluntary control group in work (214)

5.4.3. Impact on perceptions of health and employability

There was no significant evidence of the WHP having a direct impact on participants’ health using measures of self-reported health, number of GP appointments or number of hospital appointments in the previous 12 months, with no significant differences in any of these between WHP participants and the control group for either of mandatory or voluntary participants.

However, among voluntary participants with a health condition or disability, the WHP had a statistically significant positive effect on how they perceived that their condition might affect their ability to work. Those with a health condition or disability were asked the extent to which they agreed or disagreed with the following statements, using an 11-point scale from 0 ‘strongly disagree’ to 10 ‘strongly agree’:

• I am worried people won’t employ me because of my health condition/disability
• the idea of working makes me feel anxious
• I am worried that my health condition or disability will prevent me from staying in a job

The responses to these questions were discussed in detail in Chapter Three, along with other health-related measures. Figure 5.26 shows the percentages among WHP participants and the control group who agreed with each statement, scoring six or more out of 10 (agree). Table 15 in Appendix Five also shows the mean scores.

Voluntary WHP participants were statistically significantly less likely to agree that they were worried that people would not employ them because of their disability. Among voluntary participants, 60% scored six or more on the scale compared to 65% of their control group. The differences in the mean scores between the two groups was also statistically significant. Voluntary WHP participants were also less likely than the control group to agree to the other two statements, although differences were not statistically significant.

While the pattern of findings was similar for the mandatory group, the differences between the two groups did not reach statistical significance.

Figure 5.26: Impact of WHP on perceptions of employability (Wave Two participants survey)

Voluntary participants

Employability	WHP participants	Control group	p-value*
Worried people won’t employ me	60%	65%	0.012*
Idea of work makes me anxious	46%	49%	0.166
Health condition will prevent me from staying in a job	61%	64%	0.108

Mandatory participants

Employability	WHP participants	Control group	p-value*
Worried people won’t employ me	54%	57%	0.541*
Idea of work makes me anxious	45%	44%	0.892
Health condition will prevent me from staying in a job	55%	62%	0.225

*Statistically significant

ESO1: To what extent do you agree or disagree with the following statements about your health condition or disability and finding work? Please answer using a scale of 0-10 where 0 is strongly disagree and 10 is strongly agree. (% scoring 6-10/10)

Base: Voluntary WHP participants with a health condition/disability (1,436); Voluntary control group with a health condition/disability (1,167) Mandatory WHP participants with a health condition/disability (749); Mandatory control group with a health condition/disability (191)

6. Conclusions

This report has set out the findings from the participant and provider surveys and deep dive fieldwork with JCP staff, providers and participants in WHP and PSC areas. As previous chapters have shown, experiences of WHP support were reported positively by the majority of participants, with many also reporting positive personal and work outcomes.

The conclusion is structured around three key questions:

1. What is innovative about WHP?

2. Which participants benefit from the programme and how?

3. Which aspects of the programme work well?

6.1. What is innovative about WHP?

There were a few core aspects of the WHP that differentiated it from previous JCP support programmes. Voluntary participation was highlighted as a critical aspect of the WHP and created a basis for genuine conversation with key workers about readiness for work, the barriers participants faced and their individual needs, for those in the voluntary groups.

The extended time available for key workers to build rapport with participants was also mentioned as a significant development. This was an important factor for participants who appreciated having a dedicated person to focus on them and listen to their experience and needs. In contrast, the JCP Work Coaches interviewed, acknowledged that time constraints meant they were not able to focus on individuals as much as they might have wanted to.

The advantages of having additional time to focus on individuals were extended by the longevity of support offered by WHP. Providers felt that this was a necessary element of the programme if it were to help those furthest from the labour market.

The key worker approach to delivering WHP support was also an important differentiating factor. Participants reported that WHP key workers were more approachable compared with both Work Coaches and key workers they had interacted with on previous DWP programmes. Participants also felt that WHP key workers took a real interest in supporting them - highlighted in contrast with previous provision participants had experienced (where key workers were generally seen to have had less interest in helping participants return to work). Participants reported that the WHP support was positive and friendly and that they appreciated ‘being treated as an adult’ and ‘not being judged’.

The flexibility of the programme also offered advantages. Key workers recognised that participants who had been out of work for a long time faced multiple barriers before reaching a point where they could consider returning to the workplace. The WHP gave them the time and tools to help participants work through those barriers.

The WHP aimed to promote localism and ensure local needs were met. The involvement of local authorities in the DDA commissioning was perceived to have encouraged bidders to tailor their offer to the characteristics of the local area.

6.2. Which participants benefit from the programme and how?

The overall findings indicated a range of positive outcomes for all WHP participants. However, the impact analysis showed that outcomes were generally more positive for voluntary participants than mandatory participants.

6.2.1. Participant skills

Most participants reported positive views of their interpersonal skills and job application skills both at the start and end of the programme. Accordingly, there was no statistically significant evidence to suggest that the WHP had an impact on these skills among either voluntary or mandatory participants.

There were statistically significant differences showing that the programme had improved voluntary participants’ self-efficacy (their confidence, motivation and views of their chances of finding work) in comparison with the control group. But, consistent with the findings that providers were not always able to source specialist health support (including for mental health conditions, physical disabilities, mental disabilities), self-reported self-efficacy outcomes were more positive among those without a health condition. There was no statistically significant difference between mandatory WHP participants self-efficacy ratings and those of their control group.

6.2.2. Perceptions of the support

Voluntary and mandatory participants were equally positive that the support had increased their chances of moving into work, but voluntary WHP participants were statistically significantly more positive about their increased chances of moving into work compared with those in their control group. This was not true for the mandatory participants for whom control participants were equally positive about the role business-as-usual JCP support had played in increasing their chances of moving into work.

6.2.3. Impact of work experience

Opinions on whether work experience or volunteering placements had increased their chances of moving into work were similar for mandatory and voluntary participants who had been on a work experience of volunteering placement. Around three-quarters of both mandatory and voluntary participants said that their placement had increased their chances of moving into work. However, as noted in Chapter Four, work placements were not particularly prevalent and key workers did not view work placements as an essential step for everyone on their caseload.

6.2.4. Employment outcomes

Overall, slightly more WHP participants reported being in work at the time of the Wave Two participant survey compared with the control group. The survey data suggested that this difference may be driven by a difference between those with and without health conditions. WHP participants without a health condition were no more likely than control participants without a health condition to report being in work. However, WHP participants with a health condition were more likely than control participants with a health condition to say they were working.

The statistical impact analysis of survey responses showed that voluntary participants were more likely than their control group counterparts to have worked at any point since being referred to the programme. Once they were offered a job, twice as many voluntary WHP participants reported being offered in-work support compared to their control group. They were also more likely than their control group to attribute their employment outcomes to the support they had received.

There was also some limited evidence that the WHP improved employment outcomes for mandatory participants, with more mandatory participants reporting that they had been employed since referral to the programme than their control group. However, this difference did not meet the threshold for statistical significance.

Employment outcomes varied by participant age although this is not attributable to the WHP. Among both voluntary and mandatory participants, older people (particularly those aged 50 or older) were less likely to report having been employed at all compared with younger people (those aged under 50).

6.3. Which aspects of the programme work well?

6.3.1. Clear communication and collaboration

Clear communication and collaboration both within and between JCPs and providers facilitated successful delivery of the WHP. Effective communication and training were important for ensuring that Work Coaches understood the programme and had the knowledge needed to refer appropriate participants.

Strong partnerships between the organisations, which sometimes involved key workers spending time in JCP offices, also allowed for more effective face-to-face handovers between JCP and the provider. Where this happened, all parties viewed it as beneficial. Positive relationships between JCP and provider staff also helped the two teams to work together to re-engage participants who had disengaged with the support.

6.3.2. Key worker role

The role of the key worker was also a key element in successful delivery of the WHP. Participants reported that having a dedicated person to focus on them and listen was beneficial, particularly when that person was able to provide flexible support which was tailored to meet their individual needs. The benefits provided by key workers extended beyond the additional time that key workers were able to spend with participants (discussed above), which allowed them to build rapport. WHP key workers also had broad skillsets which allowed them to focus on participants’ individual needs.

Previous support programmes required key workers to have knowledge of welfare to work and case management skills, but the WHP also trained key workers to work with people with complex needs including physical and mental health conditions. This allowed them to effectively tailor the support provided to each participant.

6.3.3. Remote support

Notwithstanding the huge disruption to the delivery of the WHP, the shift to largely remote key worker service provision implemented due to the COVID-19 pandemic, brought benefits for some participants. In some cases, participants felt more comfortable taking part in the programme remotely and engaged more fully than they might have in face-to-face provision. However, participants who lacked basic IT skills struggled to engage in this way.

In addition to the support offered by key workers, WHP participants benefitted from the specialist support provided by external partners and suppliers including specialist health support and dedicated support for Early Access Group participants. As with key worker support, there were benefits to the shift to online provision as it was no longer limited to particular geographic areas and was therefore available to a wider cross-section of participants.

7. Appendices

7.1. Appendix One: Survey methodology

7.1.1. Questionnaire design

Both the Participant and Provider and surveys were tested before the start of mainstage fieldwork. The objective of the testing was to:

• confirm that respondents understood terminology used.
• test survey length against the costed interview length.
• understand what decision processes the respondent uses in coming to an answer.
• test interview flow for respondents.
• understand the period of recall and how clearly respondents could remember the referral process and support received in particular.
• test overall feelings about the questionnaire:
  • order of questions.
  • respondents’ feelings of complex areas - areas that were difficult to answer.
  • areas that were causing discomfort or respondent fatigue.
• make recommendations on how the questions can be improved or refined.

Key findings were shared with DWP, and the findings were used to improve the questionnaire draft.

Cognitive testing – Participant Survey Wave One

Kantar Public conducted 19 cognitive interviews between Wednesday 28th November and Tuesday 4th December 2018. Interviews were conducted by telephone and lasted between 30 and 45 minutes. Cognitive testing focused on programme participants and the PSC group who had been allocated to their respective allocation group between August and September 2018. No interviews were conducted with participants in the control group.

Pilot fieldwork – Participant survey Wave One

Kantar Public conducted 40 pilot interviews between Wednesday 16th and Saturday 19th January 2019. Interviews were conducted by telephone and lasted between 18 and 57 minutes. Interviews were conducted with all participant groups, including WHP, LGP, PSC and control participants.

Cognitive testing – Participant survey Wave Two

Kantar Public conducted 20 cognitive interviews between Tuesday 25th February and Friday 6th March 2020. Interviews were conducted by telephone and lasted around 45 minutes. Cognitive testing included Work and Health Programme participants from all branches (CPA, DDA, LGPs), Public Sector Comparator and control respondents who had been referred to the programme in September 2018.

Pilot fieldwork – Participant survey Wave Two

Kantar Public conducted 40 pilot interviews between Monday 2nd and Friday 6th March 2020 with respondents who were referred to the programme in October 2018. Interviews were conducted by telephone and lasted between 17 and 70 minutes. Interviews were conducted with all participant groups, including WHP, LGP, PSC and control participants.

7.1.2. Analysis weights

Three different analysis weights have been produced, each covering a different subset of respondents to the Wave One and Two surveys:

1. All Wave One survey respondents

2. All Wave Two survey respondents who were sampled for – and responded to – the Wave One survey

3. All Wave Two survey respondents

In each case, the target population remains the same: those allocated to one of (i) the WHP, (ii) the BAU control group, or (iii) the PSC, between November 2018 and December 2019 inclusive, with the exception of August and September 2019.

The sample design for Wave One was intentionally disproportionate, over-sampling the less numerous subpopulations and under-sampling the more numerous subpopulations. However, a ‘top-up’ sample was drawn for Wave Two which was largely sourced from the more numerous subpopulations. Consequently, the combined Wave Two sample (comprising Wave One respondents plus this top-up sample) was more proportionate to the overall population than was the case for the Wave One sample.

Stage One: the design weight

All the weights have a design weight at their core. The design weight for each sampled individual is equal to one divided by its sampling probability. For weights one and two, this sampling probability is for Wave One only. For weight 3, this sampling probability is a joint sampling probability: the probability of being sampled either for Wave One or for the Wave Two top-up.

However, these sampling probabilities were not known exactly for any sampled individual. DWP drew monthly samples from each subpopulation following instructions from Kantar Public with respect to the target sample size. Before drawing these samples, DWP excluded some members of the target population who could not be surveyed for one or more reasons. The number of exclusions from each subpopulation could not be reported to Kantar Public, which meant the true sampling probability could not be calculated for any individual.

The solution to this problem was to estimate the sampling probability by mapping the sample profile to the population profile, and effectively ignoring the (potentially variable) impact of exclusions. Kantar Public calculated a ‘quasi’ sampling probability by taking the mean of three estimates of the sampling probability, derived from mapping the sample against the population for three different three-way combinations of the four available profile variables:

• allocation month * programme (WHP / BAU / PSC) * referral type (disabled, EAG or LT unemployed)
• allocation month * programme (WHP / BAU / PSC) * CPA/DDA
• allocation month * CPA/DDA * referral type (disabled, EAG or LT unemployed)

This approach was taken because the sample size for each four-way combination was too small for reliable estimation.

A final step was to take the design weight based on the quasi-sampling probability and use the raking algorithm to ensure a perfect fit for the design-weighted counts against three broader combinations of profile variables:

• programme (WHP / BAU / PSC) * referral type (disabled, EAG or LT unemployed)
• CPA/DDA
• allocation month * referral type (disabled, EAG or LT unemployed)

Stage Two: the non-response weight

The samples supplied by DWP included not only the four population profile variables described above but also sex and postcode (with some missing data). After a small amount of data imputation, every sampled individual had a value for sex and some geodemographic data that had been attached via home postcode (the index of multiple deprivation and the Output Area Classification based on 2011 Census data).

A logistic regression model was fitted to each of two datasets: (i) the Wave One sample with a response indicator added; and (ii) the Wave Two top-up sample with a response indicator added. The response indicator was the dependent variable in both models. The same set of predictor variables was also used for both models, but the model coefficients differed, reflecting both the different response rates between the two surveys and some small differences in response patterns. The non-response weight for each respondent was equal to one divided by the model fitted value: an individual level estimate of response probability (conditional on being sampled for the relevant survey).

A more complex model was fitted to estimate the conditional probability of response at Wave Two, given response at Wave One. As well as the sample data, Wave One questionnaire variables were included as candidate predictor variables. Only a few of these variables improved the fit of this model: (i) programme status at the time of Wave One (started and still on programme, started but no longer on the programme, did not start the programme, allocated to BAU control group); (ii) whether in paid work at the time of Wave One; (iii) the number of self-reported motor function problems at the time of Wave One; (iv) the number of self-reported sensory problems at the time of Wave One; (v) the number of self-reported chronic health problems at the time of Wave One; and (vi) age at the time of Wave One.

Stage 3: the calibration weight

Although the combination of design weight and non-response weight produces a sample profile that is very similar to the population profile, the raking algorithm was used to ensure a perfect fit for the weighted counts against three broad combinations of profile variables:

• programme (WHP / BAU / PSC) * referral type (disabled, EAG or LT unemployed), but with PSC as a single category
• CPA/DDA
• allocation month * referral type (disabled, EAG or LT unemployed), except for Weight Two where this is just Allocation month

Summary of weights

Each weight may be broken down into its component factors and is equal to the product of these factors. The process of calculating each component is as described above. Here, we show how all of the different components – labelled A to H – fit together to form each of the three weights.

Weight One

[A] 1/p(sampled for Wave One) *
[B] 1/p(Wave One response | sampled for Wave One, sample variables) *
[C] Calibration factor One

Weight Two

[A] 1/p(sampled for Wave One) *
[B] 1/p(Wave One response | sampled for Wave One, sample variables) *
[D] 1/p(Wave Two response | Wave One response, sample variables + Wave One questionnaire variables) *
[E] Calibration factor Two

Weight Three

Wave One sampled cases:

[F] 1/p(sampled for Wave One or Wave Two) *
[B] 1/p(Wave One response | sampled for Wave One, sample variables) *
[D] 1/p(Wave Two response | Wave One response, sample variables + Wave One questionnaire variables) *
[G] Calibration factor Three

Wave Two sampled cases:

[F] 1/p(sampled for Wave One or Wave Two) *
[H] 1/p(Wave Two response | sampled for Wave Two, sample variables) *
[G] Calibration factor Three

Note that, for Weight 3, Wave One and Wave Two sampled cases were combined before calibration, with input weight factors FBD for Wave One sampled cases, and F*H for Wave Two sampled cases.

Scaling of weights

All three weights have been scaled to two different totals: (i) the relevant respondent sample size, and (ii) the target population size. The latter weight should be used for estimates of the number of people in category x or for any analysis of sums. Either weight can be used for analysis of means, proportions or variances. To obtain accurate standard errors, specialised statistical software should be used that accounts for the complex sample and weight design. Suitable tools are available in SPSS, Stata, SAS and R.

Effective sample sizes

Table x.1 below shows the actual respondent sample size (N), effective sample size accounting for weights (Neff), and the implied weighting efficiency (Neff/N) for the Wave Two weights two and three, disaggregated by programme and referral type.

In general, weighting efficiency is quite high with the finest disaggregation (programme * referral type) yielding efficiencies ranging from 80% to 88% (weight two) and from 78% to 90% (weight three). Efficiencies are somewhat lower at the broader aggregations but still quite high (e.g., for weight three, 78% for those allocated to the WHP and 84% for those allocated to business-as-usual (BAU), the control for this study).

Table 4: Actual and effective sample sizes for different subgroups (weights two and three)

Programme	Weight two: Referral type	Weight two: N	Weight two: Neff	Weight two: Weight efficiency	Weight three: N	Weight three: Neff	Weight three: Weight efficiency
PSC	–	88	78	88%	88	69	78%
BAU (Control)	Disabled	305	258	85%	1,623	1,453	90%
BAU (Control)	EAG	40	34	85%	63	51	81%
BAU (Control)	LT Unemployed	118	99	84%	356	309	87%
WHP	Disabled	401	332	83%	1,750	1,520	87%
WHP	EAG	74	59	80%	361	311	86%
WHP	LT Unemployed	118	99	84%	1,414	1,279	90%
All allocations	Disabled	781	461	59%	3,448	2,118	61%
All allocations	LT Unemployed	247	130	53%	1,781	1,561	88%
All allocations	EAG	116	79	68%	426	357	84%
PSC	All referral types	88	78	88%	88	69	78%
BAU (Control)	All referral types	463	384	83%	2,042	1,717	84%
WHP	All referral types	593	460	78%	3,525	2,736	78%
All allocations	All referral types	1,144	630	55%	5,655	3,683	65%

7.1.3. Statistical significance

Results from all surveys are attempts to estimate “true values” in a wider population; all figures come with an associated margin of error. As such, all differences quoted in the main survey report have been tested for statistical significance; that is, the difference is significant once the margins of error have been accounted for.

Unless otherwise specified, all commentary in the report focuses on differences that are statistically significant at a 95 per cent confidence level. In basic terms, you would only expect to see the result caused by chance one in 20 times.

7.2. Appendix Two: Impact analysis - Figure format, statistical tests and p-values for the impact analysis

Most of the figures in this report use the same format, presenting results for each outcome measure for the WHP participants and the control group, with separate figures for the voluntary and the mandatory groups.

Most outcomes are presented as binaries (for example having found work vs. not having found work, or a high or low impact on health) with figures including the percentage with a positive outcome^{[footnote 34]}. Where the outcome did not lend itself to a binary split, the full outcome measure was presented. Impact estimates using the full outcome measures, and mean scores where relevant, were included in the tables in Appendix Five and commented on in the report text where notably different from the binary.

The figures (and appendix tables) show for each outcome the p-value significance level of the difference between the WHP participants and control group. The p-value is the probability of an observed difference being due to chance alone, rather than being a real underlying difference for the population. A p-value of less than five per cent is conventionally taken to indicate a statistically significant difference (p<0.05). The p-values have been calculated in the complex samples module of SPSS and take into account the (non-response and propensity score) weighting of the data. Where the differences between the two groups are statistically significant (that is the p-value is less than 0.05), these are highlighted in red and with an asterisk. The term ‘statistically significant’ is often abbreviated in the text to ‘significant’.

P-values are dependent on sample size. For any given observed difference, the smaller the sample size the larger the p-value. Because the survey sample size is larger for the voluntary participants than the mandatory participants, the impacts have to be slightly larger for the mandatory participants to reach significance.

The unweighted sample sizes are cited at the end of each Figure.

7.3. Appendix Three: Survey participant attitudes and characteristics

Table 5: Participant gender, age, highest level of qualification, and ethnicity by allocation group (Wave Two participants survey)

Gender/Age	Mandatory	Voluntary	Control
Male	61%	59%	63%
Female	37%	40%	35%
In another way	1%	1%	1%
Up to 29	14%	17%	18%
30 – 39	16%	17%	17%
40 – 49	21%	20%	17%
50 – 59	28%	30%	30%
60+	18%	15%	16%

Highest qualification	Mandatory	Voluntary	Control
Level 3 or higher	29%	31%	30%
Level 2	27%	28%	27%
Level 1	12%	12%	13%
Below level 1	22%	22%	21%

Ethnicity	Mandatory	Voluntary	Control
White	77%	78%	75%
Mixed/Multiple	4%	4%	3%
Asian	6%	7%	8%
Black/African/Caribbean	10%	9%	11%
Other	1%	1%	2%

Base: All respondents asked – Mandatory: 1,348, Voluntary: 2,037 and Control: 1,954

(Gender) Self-reported gender; (Age) Self-reported age; (ESASUPP_K9) Highest level of qualification attained by participants; (ETHNICITY) Ethnicity of respondent.

Table 6: Deep dive participant characteristics

Eligibility Group/Gender/Age	Wave One	Wave Two
Disability	66%	66%
Long Term Unemployed	19%	15%
Early Access Group	15%	19%
Male	56%	62%
Female	44%	38%
Up to 29	18%	20%
30 – 39	20%	23%
40 – 49	26%	23%
50 – 59	23%	20%
60+	14%	14%

Base: All participants – Wave one: 113 and Wave two: 81

7.4. Appendix Four: Public Sector Comparator

This appendix draws together evidence from the deep dives and participant surveys relating to the design and delivery of the Public Sector Comparator group (PSC). Given the limited sample size for the PSC group in the Wave Two survey, the findings should be regarded as indicative.

Summary

• PSC participant characteristics: 64% of PSC participants were male. Participants covered the full age range but were relatively more likely to be aged 40-49 (30%). 13% had a Level 1 qualification or below while 41% had a Level 3 qualification or higher.
• PSC participants reported a positive view of the support they received. Most were satisfied with the support they received and agreed that it had increased their chances of moving into work.
• at the time of the Wave Two interview, 23% of PSC participants had started work at any time since being referred to the programme, and 14% were currently employed.
• drivers of success among PSC participants included the flexibility to schedule longer appointments which allowed Work Coaches to perform a similar role to WHP key workers.
• in comparison with the WHP, PSC participants were not able to access equivalent specialist support offered by provider supply chains.

7.4.1. PSC overview

In four local authorities JCPs were provided with additional funding and resources to run an equivalent programme of additional support for eligible participants themselves. The support provided by the PSC was similar to the support provided by WHP. The aim of this programme was to establish whether given similar resources, JCPs could deliver a similar or better performance in terms of outcomes and costs.

The PSC was rolled out in January 2018 in:

• Lincolnshire, Nottinghamshire and Rutland
• Dorset, Wiltshire, Hampshire and the Isle of Wight
• Leicestershire and Northamptonshire
• Devon and Cornwall

PSC allocation was halted at the start of the COVID-19 pandemic (Spring 2020) to allow the Work Coaches involved to focus on processing the influx of new benefit claims.

7.4.2. PSC programme design

Senior JCP staff in the deep dive PSC areas reported they had been given flexibility in determining how to structure their local PSC support within an agreed delivery framework. Two different delivery models were subsequently adopted:

1. In three of the four PSCs, each participating JCP office had one or two designated PSC Work Coaches. After the point of referral, participants were reassigned onto the caseload of one of the designated PSC Work Coaches, who they would then see for more frequent and extended appointments in line with the framework design and standards. The rationale for this approach was that some Work Coaches would be better equipped to deliver PSC support than others, and that once in the role those designated to do it would be able to build up their knowledge and expertise in working with this customer group further.

Work Coaches were invited to put themselves forward to perform this role through an internal expression of interest process. There was reportedly a high level of interest from Work Coaches, reflecting the appeal of being able to spend longer with, and potentially do more to help, participants than was possible through more time-constrained business as usual contact. Senior staff said they had selected Work Coaches for the role who had previous experience of working with claimants with health or other additional barriers to work.

2. In the one other PSC area, they implemented a dispersed model whereby all Work Coaches in participating JCP offices were responsible for delivering PSC support. When a referral was made, the participant’s existing Work Coach would continue to have them on their caseload and see them for more frequent and extended appointments. The rationale for this approach was that it provided continuity for the participant, gave all Work Coaches the opportunity to work for longer and more supportively with participants and enabled a greater spread of capability in supporting this customer group.

7.4.3. PSC participant characteristics

A small number of PSC participants were included in the Wave Two survey. Although the numbers are too small to allow comparison with other participant groups, we have included an overview of the key characteristics of this group.

Voluntary participants made up 80% of the PSC, with mandatory participants forming the remaining 20%. The Disability group constituted 73% of referrals, with 8% in the EAG, and 20%in the LTU (Figure 7.1).

Figure 7.1 Breakdown of Wave Two PSC participant sample by mandatory / voluntary participation and eligibility group

	Percentage
Mandatory	20%
Voluntary	80%
Disabilty Group	73%
Early Access Group	8%
Long-term Unemployed	20%

Base: All respondents Wave Two (5655) PSC Wave Two (88*) *CAUTION: Number of respondents below 100

Please note data on this slide combines responses collected at Wave One from longitudinal participants who participated at Wave Two and retrospective responses collected at Wave Two from fresh respondents.

PSC demographic characteristics by gender, age, highest qualification and ethnicity are shown in Table 7.

Almost two-thirds of PSC participants were male, most were over 40 years old, and 96% were white.

Table 7: Participant gender, age, and highest level of qualification, by allocation (Wave Two participants survey)

Gender/Age	Public Sector Control
Male	64%
Female	35%
Described themselves in another way	1%
Up to 29	14%
30 – 39	17%
40 – 49	30%
50 – 59	25%
60+	13%

Highest qualification	Public Sector Control
Level 3 or higher	41%
Level 2	25%
Level 1	12%
Below level 1	13%

Ethnicity	Public Sector Control
White	96%
Mixed/Multiple	–
Asian	1%
Black/African/Caribbean	1%
Other	1%

Base: All respondents asked 88

(Gender) Self-reported gender; (Age) Self-reported age; (ESASUPP_K9) Highest level of qualification attained by participants; (Ethnicity) Ethnicity of respondent

Physical and mental health conditions were common among PSC participants, with 70% reporting a health condition or disability expected to last six months or more. Conditions varied but mental health (68%), musculo-skeletal (37%), and chronic/progressive conditions (36%) were most frequently mentioned. Other common conditions are outlined in Figure 7.2.

Figure 7.2: PSC participant health conditions (Wave Two participants survey)

Whether respondent has any health conditions

Has health condition	Percentage
Yes	70%
No	30%

Which health conditions respondents have (those with any)

health condition	Percentage
Mental health	68%
Musculo-skeletal/physical injury	37%
Chronic/systemic	36%
Learning difficulties/cognitive	23%
Sensory impairment/communication	6%
Other condition	18%

WPK23: Can I check, do you have any physical or mental health conditions, disabilities or illnesses lasting or expected to last for six months or more?

WPK25: Could you tell me what your illness, health condition or disability is? Base: Wave Two participants who were asked: PSC (68*) *CAUTION: Number of respondents below 100

7.4.4. PSC outcomes

At the time of the Wave Two interviews, 23% of PSC participants had started work at any time since being referred to the programme, and 14% were currently employed (see Figure 7.3).

Figure 7.3: PSC employment outcomes

Employment since the start of the programme

Employment status	Percentage
Employed	23%
Employed FT – 30 hours + a week	11%
Employed PT – 16-30 hours a week	3%
Employed PT – <16 hours a week	9%
Self-employed	2%
Not been employed since referral	73%

Current employment status

Employment status	Percentage
Employed	14%
Employed full-time	6%
Employed part-time	7%
Self-employed	3%
Focus primarily on looking for work	38%
Unemployed and not looking for work	4%
Coping with a long-term condition or disability	24%
Caring for children or other people	4%
Other/DK	13%

TRAINEE_ECONOUT: Since being referred to the programme have you started paid work at any time?

TRAINEE_ECONACT: Can I check, which of the following best describes your current employment status?

Base: All Wave Two participants asked: PSC (88*) *CAUTION: Number of respondents below 100

Most PSC participants (63%) agreed that the support had increased their chances of moving into work (see Figure 7.4).

Figure 7.4: Participant perceptions of whether support had increased chances of moving into work (Wave Two participants who started receiving support)

Increased chances	Percentage
Yes – total	63%
Yes – to large extent	33%
Yes – to some extent	29%
No – total	35%
No – not really	28%
No – not at all	7%
Don’t know	–

JCPOFFER_USEFUL1: And do you feel the support you received through […] has increased your chances of moving into work?

Base: Wave Two participants who had started receiving support: PSC (87*) *CAUTION: Number of respondents below 100

7.4.5. Satisfaction with PSC support

Overall, interviewees in the Wave One deep dive research were positive about the effectiveness of the PSC support being delivered in their area. Stakeholders perceived it to match much of what the equivalent WHP commissioned provision offered and potentially even have some advantages over it. The Wave Two participant survey showed that PSC participants continued to be positive about the support offered to them. Among PSC participants, 81% were satisfied with the support offered (see Figure 7.5).

Figure 7.5: PSC participant satisfaction with support (Wave Two participants who started receiving support)

Level of satisfaction	Percentage
Satisfied	83%
Very satisfied	43%
Fairly satisfied	40%
Fairly dissatisfied	9%
Very dissatisfied	6%
Don’t know	–

CUSTSAT_OVERALLSAT1: Overall how satisfied or dissatisfied are you with the support you have had through […]? Are you…? Base: W2 participants who had started receiving support: PSC (87*) *CAUTION: Number of respondents below 100

7.4.6. Drivers and barriers to PSC success

Deep dive interviewees reported constraints and challenges. The main factors that were perceived to mediate the success of PSC support in achieving positive outcomes with participants were.

Dedicated time for PSC Work Coach and participant contact

Work Coaches designated to deliver PSC support had the flexibility to schedule longer appointments (of reportedly 30 minutes to an hour) with participants than the appointments they would ordinarily have with a claimant in the course of their benefits claim. This was perceived by all parties (including most participants) to be an important success factor for PSC support. It provided scope for the Work Coach to build up more of a personal relationship with the participant and explore their barriers, capabilities and potential support needs in relation to work.

‘There are so many people we see, with 15-minute appointments it’s hard. There are many ways the Work Coach would like to help if only they had the time. With the PSC they can do this.’ JCP manager

‘We have got to know each other through the appointments, and so she knew what I needed to talk about, how I was anxious about work.’ PSC participant

‘At the usual appointments it’s more ‘have you done this, this, and this’ to tick boxes. This was more personal.’ PSC participant

With the benefits of more time, the evidence indicated PSC Work Coaches were able to perform a similar role to a key worker employed by one of the WHP commissioned providers. This included providing general encouragement and confidence-building but also extended to PSC Work Coaches providing some practical forms of employment-related support (such as help with CVs and job searching) that they did not ordinarily always have the time to undertake.

This was supported by the Wave Two survey findings which showed that more than eight in ten PSC participants were satisfied with the range of ways they had been able to contact their Work Coach, the frequency of contact with their Work Coach, how available the Work Coach was when they needed them and the extent to which the Work Coach tailored the support offered (see Figure 7.6).

Figure 7.6: Satisfaction with Work Coach contact

Type of contact with Work Coach	Satisfied	Very satisfied	Fairly satisfied	Fairly dissatisfied	Very dissatisfied	Don’t know
Range of different ways to contact a key worker/Work Coach	81%	53%	29%	11%	4%	5%
Frequency of contact you’ve had with key worker/Work Coach	85%	54%	31%	10%	3%	–
Availability a key worker/Work Coach is when you need them	82%	48%	34%	9%	4%	5%
Extent key worker/Work Coach tailored support offered you to your individual needs	83%	55%	28%	14%	1%	–

NEW_COACHSAT1: Are you satisfied or dissatisfied with…

Base: Wave Two participants who had started receiving support: PSC (87*) *CAUTION: Number of respondents below 100

PSC Work Coach skills and training

Similarities between the PSC Work Coach and WHP key worker role extended to the skills they were perceived to need to make successful progress with participants – with the emphasis on softer inter-personal skills. Many PSC participants in particular emphasised the empathy, listening skills, and motivational skills of their Work Coach in same way that WHP participants did when talking about their key worker. Knowledge and experience of working with individuals with different health conditions was considered useful but – as with WHP key workers – this did not reportedly have to be specialist in-depth knowledge in order for them to still be able to perform their role effectively.

The initial training for Work Coaches to deliver PSC support reportedly varied. In one of the PSCs, it was described as extensive and included training by work psychologists in coaching skills, professional boundaries, developing action plans, and using coaching tools such as life wheels^{[footnote 35]}. This was highly valued by the Work Coaches who received it. They felt it had built on their pre-existing skills and gave them additional approaches to draw on with PSC participants. In the other PSCs the initial training was described as more light touch, comprising, for example, a one-off training session delivered by a DEA about how to work with individuals with a health condition or disability. Once delivery had started, DWP also facilitated regular cross-PSC sharing of good practice, learning and tools.

The evidence from the deep dive research indicated that the more training and support there was for PSC Work Coaches the better. Particularly in the PSC where all Work Coaches were given responsibility for delivering PSC support (rather than individual Work Coaches selected on the basis of relevant previous experience) it was felt that more initial training would have been beneficial.

Access to wider support

Figure 7.7 shows that PSC participants who responded to the Wave Two survey received support across a range of areas. More than half received support with job applications and training. And more than four in ten said they had received help with their health and wellbeing or with their finances.

Figure 7.7: Support received by PSC participants

7.7a: Job applications and training

Types of support	Percentage
Writing CV, applications or interview skills	61%
Chance of doing a training course	59%
Careers advice	58%
Assessment of skills	43%
Work experience or voluntary work	37%
Support/training in maths, reading, writing, or English language, IT skills	27%
Support from local employers	23%
Support/advice for setting up own business or self-employed	13%

7.7b: Health and wellbeing

Types of support	Percentage
Advice/support about your wellbeing and/or mental health	46%
Advice/support about a health condition or disability you have	38%
Session on motivation or confidence	29%

7.7c: Finances

Types of support	Percentage
Financial support to help with costs with looking for work	40%
Financial advice	19%

7.7d: Other

Types of support	Percentage
Help with housing	12%
Help/advice in relation to looking after children or adults	7%
Help with any drug or alcohol problems	5%
Help/advice about a criminal record	4%

WORKPROG_SUPP1: Have you received any of the following support through the programme? This could be support you would receive from […] or from another organisation that they would refer you to.

Base: Wave Two participants who had started receiving support: PSC (87*) *CAUTION: Number of respondents below 100

PSC Work Coaches were able to provide this support despite having no direct equivalent to the supply chains of more specialist support that WHP key workers could refer participants onto. Interviewees cited this as one limitation the PSC support had in comparison to commissioned WHP provision and some suggested participating JCP offices should have been allocated a budget to be able to subcontract specialist provision for PSC participants.

‘It doesn’t seem fair on the Public Sector Comparator Work Coaches, who are trying their best to work well with this group of people, if they haven’t got access to all the same resources.’ JCP district manager

‘All I could give them was my time rather than additional provision or anything like that.’ PSC Work Coach

However, there were also positive examples, to a greater or lesser extent in all four of the PSCs, of proactive approaches taken by JCP staff to give PSC participants access to a wider array of support.

This was partly through capitalising on existing in-house JCP resources and partly through effective signposting to other local support. For example, DEAs were reported to have provided additional health-related input and support for PSC participants. PSC Work Coaches were also encouraged to make full use of the Flexible Support Fund^{[footnote 36]} to, for example, meet the costs if a participant attended a training course. PSC Work Coaches could also signpost or refer participants onto other DWP programmes. For example, one JCP office identified a potential synergy with other nationally funded provision that was already running in their local area and had prioritised PSC participants in allocating places onto this.

In the PSCs where these opportunities had been mostly fully exploited, the Work Coaches involved expressed the most confidence in the extent to which they could meet participants’ needs and help them achieve positive outcomes. More generally, PSC Work Coaches said they had built up their knowledge of local public and third sector services that they could signpost participants to over time. The overall direction of travel at the time of the Wave One deep dive research was towards Work Coaches making increased use of more sources of internal and external support (and thereby possibly narrowing the gap between the breadth of support what they could help participants access in comparison to WHP providers).

The JCP office environment

At the time of the Wave One deep dive research (and pre-pandemic) it was reported that all contact between PSC Work Coaches and participants was taking place on-site in JCP offices. Some PSC Work Coaches perceived it would be beneficial if they could engage with participants in informal environments such a coffee shop or café. This, they suggested, would help to further differentiate the PSC engagement from business-as-usual appointments in the JCP office and further encourage participants to talk freely about their personal challenges, aspirations and support needs.

Work Coaches could complete an outreach work assessment to engage with claimants outside of JCP settings but those interviewed indicated that they had been deterred from doing this because of a perception that the associated administrative process would be onerous. The positive evidence from the participant interviews and surveys also showed that, even in the JCP office environment, Work Coaches were in most cases successful in making their PSC appointments ‘feel’ different and in enabling participants to talk freely.

Taken as a whole, the evidence indicated that dedicated time for PSC Work Coach and participant contact and Work Coach skills were important determinants of the programme effectiveness (much more so than the physical environment in which appointments took place). There were also some potential benefits to conducting PSC appointments in JCP offices. The PSC Work Coaches interviewed generally reported fewer issues than WHP providers with participants failing to attend initial and subsequent appointments. This was likely to be due to the familiarity of the JCP office setting to participants and that PSC Work Coaches reported scheduling their contact to, as far as possible, coincide with when participants would be at the JCP office for other appointments.

7.5. Appendix Five: Outcome tables with subgroups

Table 8: Impact of the Work and Health Programme on entry to employment

Employment status	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Worked at all since allocation	27%	22%	5	0.002*	19%	17%	2	0.621
Employed/self-employed 30+ hours per week	12%	10%	2	0.007*	7%	8%	0	0.792
Employed/self-employed <30 hours per week	15%	12%	3	–	11%	9%	2	–
Not worked	73%	78%	-5	81%	83%	-2	–
% Currently working	19%	16%	3	0.049*	13%	10%	4	0.188
Employed/self-employed	19%	16%	3	0.031*	13%	10%	4	0.142
Training, education or apprenticeship	1%	2%	-1	–	1%	1%	0	–
Looking for work	37%	41%	-4	–	52%	51%	1%	–
Unemployed not looking for work	10%	8%	2	–	8%	10%	-1	–
Sick or disabled	26%	27%	-2	–	17%	24%	-7	–
Caring	4%	3%	1	–	4%	3%	1	–
Other	4%	3%	1	–	4%	2%	2	–
Base: all	2037	1600	–	–	1348	354	–	–

Table 9: Impact of the Work and Health Programme on employment support, job suitability and length of employment

Level of employment support, job suitability and length of employment	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value
% Perceiving that support helped them get their job	68%	48%	20	<0.001*
% Offered in-work support	40%	21%	19	<0.001*
Whether job fits commitments or health conditions: % Mostly or partly	78%	76%	1	0.736
Whether job fits commitments or health conditions: Mostly	56%	56%	0	0.932
Whether job fits commitments or health conditions: Partly	22%	21%	1	–
Whether job fits commitments or health conditions: Not at all	22%	24%	-1	–
Whether job meets criteria looking for: % Mostly or partly	89%	88%	1	0.859
Whether job meets criteria: Mostly	67%	62%	6	0.451
Whether job meets criteria: Partly	22%	27%	-5	–
Whether job meets criteria: Not at all	11%	12%	-1	–
Whether job offers progression: % Mostly or partly	72%	72%	0	0.968
Whether job offers progression: Mostly	47%	52%	-5	0.471
Whether job offers progression: Partly	25%	20%	5	–
Whether job offers progression: Not at all	28%	28%	0	–
Whether job has benefitted physical health: % Mostly or partly	76%	75%	0	0.934
Whether job has benefitted physical health: Mostly	50%	54%	-3	0.660
Whether job has benefitted physical health: Partly	25%	21%	4	–
Whether job has benefitted physical health: Not at all	24%	25%	0	–
Whether job has benefitted mental health: % Mostly or partly	88%	89%	-1	0.792
Whether job has benefitted mental health: Mostly	70%	68%	3	0.635
Whether job has benefitted mental health: Partly	18%	22%	-4	–
Whether job has benefitted mental health: Not at all	12%	11%	1	–
Length of current employment	–	–	–	0.001*
Length of current employment: Fewer than two months	11%	19%	-8	–
Length of current employment: Two months, less than six months	22%	9%	13	–
Length of current employment: Six months or more	67%	72%	-5	–
Length of current employment: Don’t know	1%	0%	0	–
Base: all currently working	361	214	–	–

Table 10: Impact of the Work and Health Programme on job search

Number of job applications compared to before	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Applying for more than before	23%	17%	5	0.012*	21%	16%	5	0.158
Applying for more	23%	17%	5	0.004*[footnote 37]	21%	16%	5	0.064[footnote 38]
Applying for the same number	30%	27%	4	35%	27%	8	–
Applying for fewer	44%	53%	-9	40%	52%	-12	–
Don’t know	3%	3%	0	3%	4%	-1	–
Base: all currently looking for work	1043	866	–	–	870	244	–	–

When think might find work	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
In next three months	16%	14%	2	0.209	16%	15%	1	0.895
In next six months	11%	12%	-1	–	12%	10%	2	–
Longer	29%	31%	-3	–	26%	28%	-1	–
Don’t know	44%	43%	2	–	46%	47%	-1	–
Base: all not currently working	1.661	1,379	–	–	1,168	325	–	–

Table 11: Impact of the Work and Health Programme on perceptions of support

Satisfied with support	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Very or fairly satisfied	85%	80%	5	0.002*	84%	85%	-1	0.784
Very satisfied	48%	41%	6	0.001*[footnote 39]	45%	41%	4	0.446[footnote 40]
Fairly satisfied	37%	39%	-2	–	39%	44%	-5	–
Fairly dissatisfied	6%	10%	-4	–	7%	6%	1	–
Very dissatisfied	8%	9%	-1	–	7%	9%	-1	–
Don’t know	1%	1%	0	–	1%	0%	1	–

Increased chances of moving into work	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Yes, to a large or some extent	61%	53%	7	<0.001*	60%	61%	-1	0.841
Yes, to a large extent	22%	19%	3	0.005*[footnote 41]	21%	21%	-1	0.727[footnote 42]
Yes, to some extent	39%	35%	4	–	40%	40%	0	–
No not really	18%	23%	-5	–	19%	20%	-1	–
No not at all	16%	17%	-2	–	16%	13%	3	–
Don’t know	3%	4%	-1	–	3%	5%	-2	–
Not applicable	2%	2%	0	–	2%	2%	0	–

Built confidence that could find a job	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Yes	68%	61%	7	<0.001*	69%	69%	0	0.935
Yes	68%	61%	7	0.001*	69%	69%	0	0.863
No	28%	34%	-5	–	27%	28%	-1	–
Don’t know	3%	4%	-1	–	3%	3%	0	–
Not applicable	2%	2%	0	–	1%	1%	0	–

Increased motivation to find work	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Yes	69%	63%	6	0.002*	71%	74%	-3	0.345
Yes	69%	63%	6	0.004*	71%	74%	-3	0.661
No	27%	31%	-4	–	25%	22%	3	–
Don’t know	2%	4%	-1	–	2%	3%	0	–
Not applicable	2%	2%	0	–	2%	1%	1	–

Increased chances of finding a suitable job	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Yes	67%	61%	6	0.001*	67%	70%	-4	0.337
Yes	67%	61%	6	0.007*	67%	70%	-4	0.426
No	28%	34%	-6	–	28%	25%	3	–
Don’t know	3%	4%	0	–	4%	4%	0	–
Not applicable	2%	2%	0	–	1	0	1	–
Base: all	2037	1600	–	–	1348	354	–	–

Table 12: Impact of the Work and Health Programme work or voluntary placements

Placement increased chances of moving into work	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Yes, to a large or some extent	74%	74%	-1	0.854	75%	75%	0	0.995
Yes, to a large extent	34%	32%	2	0.483[footnote 43]	31%	41%	-10	0.116[footnote 44]
Yes, to some extent	40%	42%	-2	–	44%	34%	10	–
No not really	12%	13%	-1	–	10%	15%	-5	–
No not at all	5%	5%	0	–	8%	6%	2	–
Don’t know	4%	5%	-1	–	2	4	-2	–
Not applicable	6%	3%	3	–	5%	0%	5	–

Placement meant developed new skills	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Agree	77%	79%	-2	0.493	77%	80%	-3	0.631
Strongly agree	29%	30%	-1	0.151[footnote 45]	30%	32%	-2	0.539[footnote 46]
Agree	48%	49%	-1	–	47%	48%	-1	–
Disagree	12%	15%	-3	–	15%	13%	2	–
Strongly disagree	3%	2%	2	–	2%	3%	-1	–
Don’t know	4%	2%	2	–	2%	4%	-2	–
Not applicable	4%	2%	2	–	4%	0%	3	–

Placement meant more attractive to potential employers	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Agree	80%	81%	-1	0.696	78%	82%	-4	0.535
Strongly agree	31%	26%	4	0.336[footnote 47]	25%	39%	-14	0.253[footnote 48]
Agree	49%	54%	-5	–	53%	42%	11	–
Disagree	11%	12%	-1	–	12%	9%	2	–
Strongly disagree	1%	2%	-1	–	3%	1%	2	–
Don’t know	5%	3%	1	–	3%	5%	-1	–
Not applicable	4%	2%	2	–	4%	4%	0	–
Base: all going on placements	429	396	–	–	351	94	–	–

Table 13: Impact of the Work and Health Programme on self-efficacy and job search confidence

Meeting new people	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Very or fairly confident	70%	71%	-1	0.617	77%	78%	0	0.884
Very confident	31%	31%	0	0.452[footnote 49]	39%	37%	2	0.946[footnote 50]
Fairly confident	39%	40%	-1	–	39%	41%	-2	–
Not very confident	21%	19%	2	–	17%	17%	0	–
Not at all confident	9%	11%	-1	–	6%	5%	0	–

Having a go at new things	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Very or fairly confident	74%	74%	0	0.964	79%	82%	-3	0.390
Very confident	26%	29%	-3	0.119[footnote 51]	34%	33%	1	0.781[footnote 52]
Fairly confident	48%	45%	3	–	46%	49%	-4	–
Not very confident	20%	19%	1	–	15%	12%	2	–
Not at all confident	6%	7%	-1	–	6%	6%	0	–

Working with others in a team	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Very or fairly confident	80%	78%	2	0.304	85%	83%	2	0.562
Very confident	40%	41%	-1	0.403[footnote 53]	49%	48%	1	0.914[footnote 54]
Fairly confident	40%	37%	3	–	36%	36%	1	–
Not very confident	14%	15%	-1	–	10%	12%	-2	–
Not at all confident	6%	7%	-1	–	5%	5%	0	–

Putting forward ideas	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Very or fairly confident	70%	70%	0	0.981	78%	77%	1	0.681
Very confident	26%	29%	-3	0.185[footnote 55]	33%	31%	1	0.710[footnote 56]
Fairly confident	44%	41%	3	–	45%	45%	0	–
Not very confident	22%	21%	1	–	14%	13%	1	–
Not at all confident	8%	9%	-1	–	8%	10%	-2	–

Completing job applications/CV	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Very or fairly confident	72%	72%	0	0.783	79%	80%	-1	0.675
Very confident	32%	33%	-1	0.952[footnote 57]	39%	40%	-1	0.626[footnote 58]
Fairly confident	40%	39%	1	–	40%	40%	-1	–
Not very confident	19%	19%	-1	–	14%	11%	3	–
Not at all confident	9%	9%	0	–	7%	9%	-2	–

Making a good impression at interview	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Very or fairly confident	73%	72%	1	0.486	77%	75%	2	0.480
Very confident	28%	30%	-3	0.051[footnote 59]	31%	29%	3	0.369[footnote 60]
Fairly confident	45%	42%	4	–	46%	46%	0	–
Not very confident	20%	19%	1	–	16%	15%	1	–
Not at all confident	7%	10%	-2	–	7%	11%	-4	–

Applying for jobs online	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Very or fairly confident	70%	69%	1	0.493	74%	78%	-4	0.243
Very confident	35%	35%	0	0.886[footnote 61]	42%	44%	-2	0.686[footnote 62]
Fairly confident	35%	33%	1	–	32%	34%	-2	–
Not very confident	17%	18%	-1	–	16%	13%	3	–
Not at all confident	13%	13%	-1	–	10%	9%	1	–

How much like to work in future (scale from 0 (not at all) to 10 (a lot))	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
0 to 3	5%	5%	-1	0.191	6%	2%	3	0.064
4 to 7	19%	19%	0	–	17	18	-2	–
8 to 10	55%	58%	-3	–	63%	67%	-4	–
Already in work	19%	16%	3	–	13%	9%	4	–
Don’t know	3%	3%	1	–	2%	3%	1%	–
Mean score (for those not in work)	8.15 (sd 2.54)	8.12 (sd 2.61)	–	0.829	8.23 (sd 2.58)	8.54 (sd 2.08)	–	0.069
Base: all	2037	1600	–	–	1348	354	–	–

Table 14: Impact of the Work and Health Programme on ONS life satisfaction and wellbeing measures

Satisfied with life (scale from 0 (not at all) to 10 (completely))	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Satisfied (7+)	42%	37%	5	0.010*	48%	51%	-3	0.434
Mean score	5.71 (sd 2.76)	5.58 (sd 2.77)	–	0.241	6.16 (sd 2.72)	6.19 (sd 2.53)	–	0.874

Life is worthwhile (scale from 0 (not at all) to 10 (completely))	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Worthwhile (7+)	51%	48%	3	0.090	58%	60%	-2	0.535
Mean score	6.31 (sd 2.80)	6.21 (sd 2.80)	–	0.354	6.82 (sd 2.66)	6.70 (sd 2.72)	–	0.604

Happy (scale from 0 (not at all) to 10 (completely))	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Happy (7+)	47%	46%	1	0.531	55%	56%	-1	0.806
Mean score	6.02 (sd 2.93)	6.03 (sd 2.94)	–	0.913	6.49 (sd 2.94)	6.48 (sd 2.68)	–	0.968

Anxious (scale from 0 (not at all) to 10 (completely))	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Anxious (6+)	37%	39%	-2	0.387	37%	37%	-1	0.853
Mean score	4.45 (sd 3.21)	4.57 (sd 3.25)	–	0.339	4.36 (sd 3.37)	4.56 (sd 3.21)	–	0.444
Base: all	2037	1600	–	–	1348	354	–	–

Table 15: Impact of the Work and Health Programme on health

Self-reported health (scale from 0 (very poor) to 10 (very good))	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Good (7+)	37%	37%	0	0.851	49%	49%	1	0.834
Mean score	5.48 (sd 2.65)	5.61 (sd 2.60)	–	0.191	6.22 (sd 2.71)	6.15 (sd 2.50)	–	0.723

GP appointments in past 12 months	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
None	24%	24%	-1	0.367	33%	32%	1	0.377
One or two	23%	23%	0	–	24%	25%	-1	–
Three to nine	34%	34%	0	–	30%	32%	-2	–
Ten or more	17%	15%	2	–	11%	11%	0	–
Don’t know	2%	3%	-1	–	3	1%	2%	–

Hospital appointments in past 12 months	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
None	43%	44%	-1	0.951	53%	53%	1	0.865
One or two	24%	25%	0	–	24%	26%	-3	–
Three to nine	24%	23%	1	–	19%	16%	2	–
Ten or more	7%	7%	0	–	3%	3%	0	–
Don’t know	2%	2%	0	–	1	1	0	–
Base: all	2037	1600	–	–	1348	354	–	–

Worry that people will not employ me because of my health condition/disability (scale from 0 (strongly disagree) to 10 (strongly agree))	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Agree (7+)	60%	65%	-5	0.012*	54%	57%	-3	0.541
Mean score	6.22 (sd 3.24)	6.64 (sd 3.28)	–	0.003*	5.91 (sd 3.52)	6.22 (sd 3.17)	–	0.368

Idea of working makes me anxious (scale from 0 (strongly disagree) to 10 (strongly agree))	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Agree (7+)	46%	49%	-3	0.166	45%	44%	1	0.892
Mean score	5.03 (sd 3.41)	5.37 (sd 3.44)	–	0.026*	–	5.03 (sd 3.49)	5.11 (sd 3.26)	–	0.810

Worried my health condition/disability will prevent me staying in a job (scale from 0 (strongly disagree)
to 10 (strongly agree))	Voluntary group (VG): WHP participants	VG: control group	VG: percentage point difference	VG: p-value	Mandatory group (MG): WHP participants	MG: control group	MG: percentage point difference	MG: p-value
% Agree (7+)	61%	64	-3	0.108	55%	62%	-6	0.225
Mean score	6.33 (sd 3.20)	6.63 (sd 3.22)	–	0.036*	5.87 (sd 3.55)	6.19 (sd 3.13)	–	0.336
Base: all with a health condition or disability	1,167	1,436	–	–	749	191	–	–

Table 16: Impacts of the Work and Health Programme by health condition or disability: voluntary participants

Employment/health status	Mental/physical: WHP participants	Mental/physical: control group	Mental/physical: p-value	Mental, not physical: WHP participants	Mental, not physical: control group	Mental, not physical: p-value	Physical not mental: WHP participants	Physical not mental: control group	Physical not mental: p-value	No health condition: WHP participants	No health condition: control group	No health condition: p-value	p-value for diff’al impact
% Worked at all since allocation	23%	14%	0.014*	31%	27%	0.358	24%	22%	0.491	27%	24%	0.387	0.418
% Currently working	16%	11%	0.074	19%	20%	0.827	17%	16%	0.649	20%	17%	0.403	0.518
% Thinking support increased chances of moving into work to a large or some extent	52%	37%	<0.001*	65%	55%	0.039*	55%	57%	0.709	69%	61%	0.036*	0.019*
% Very or fairly confident completing job applications/CV	63%	61%	0.693	74%	63%	0.013*	72%	75%	0.334	77	82	0.073	0.015*
% Satisfied with life (7+ on scale from 0 (not at all) to 10 (completely))	24%	16%	0.026*	33	27	0.155	41%	38%	0.374	63%	58%	0.173	0.577
% Good health (7+ on scale from 0 (very poor) to 10 (very good))	16%	13%	0.323	33%	30%	0.572	25%	27%	0.574	68%	70%	0.643	0.610
Base: all	400	295	–	318	278	–	568	455	–	558	399	–	–
% Applying for more job than before	16%	17%	0.879	30%	13%	0.005*	19%	16%	0.433	26%	22%	0.262	0.189
Base: all currently looking for work	147	118	–	161	113	–	288	247	–	345	286	–	–

Table 17: Impacts of the Work and Health Programme by health condition or disability: mandatory participants

Employment/health status	Has health condition or disability: WHP participants	Has health condition or disability: control group	Has health condition or disability: p-value	Does not: WHP participants	Does not: control group	Does not: p-value	p-value for differential impact
% Worked at all since allocation	16%	10%	0.114	22%	27%	0.368	0.073
% Currently working	10%	6%	0.246	18%	14%	0.475	0.626
% Thinking support increased chances of moving into work to a large or some extent	55%	55%	0.943	67%	69%	0.795	0.807
% Very or fairly confident completing job applications/CV	74%	78%	0.361	86	82	0.378	0.211
% Satisfied with life (7+ on scale from 0 (not at all) to 10 (completely))	35%	39%	0.403	66%	65%	0.836	0.484
% Good health (7+ on scale from 0 (very poor) to 10 (very good))	29%	30%	0.931	76%	73%	0.507	0.578
Base: all	749	191	–	569	158	–	–
% Applying for more job than before	16%	15%	0.800	27%	17%	0.113	0.379
Base: all currently looking for work	440	120	–	108	33	–	–

Table 18: Impacts of the Work and Health Programme by length of unemployment: voluntary participants

Unemployment/health status	Less than a year: WHP participants	Less than a year: control group	Less than a year: p-value	1 year, less than 3: WHP participants	1 year, less than 3: control group	1 year, less than 3: p-value	3 years, less than 10: WHP participants	3 years, less than 10: control group	3 years, less than 10: p-value	10 or more years: WHP participants	10 or more years: control group	10 or more years: p-value	Never in paid work: WHP participants	Never in paid work: control group	Never in paid work: p-value	p-value for differential impact
% Worked at all since allocation	43%	45%	0.743	28%	23%	0.144	20%	12%	0.021*	14%	8%	0.072	21%	12%	0.014*	0.079
% Currently working	29%	30%	0.790	20%	15%	0.102	15%	9%	0.056	10%	6%	0.140	14%	12%	0.618	0.338
% Thinking support increased chances of moving into work to a large or some extent	62%	58%	0.321	65%	52%	0.002*	57%	48%	0.015*	57%	55%	0.627	65%	56%	0.109	0.456
% Very or fairly confident completing job applications/CV	79%	78%	0.797	75%	75%	0.945	75%	71%	0.265	62%	62%	0.950	65%	68%	0.573	0.830
% Satisfied with life (7+ on scale from 0 (not at all) to 10 (completely))	44%	40%	0.323	44%	35%	0.017*	38%	35%	0.387	37%	36%	0.930	46%	44%	0.671	0.633
% Good health (7+ on scale from 0 (very poor) to 10 (very good))	42%	39%	0.401	39%	39%	0.940	35%	34%	0.690	25%	30%	0.343%	35%	46%	0.041*	0.222
Base: all	445	322	–	441	330	–	469	364	–	280	227	–	236	207	–	–
% Applying for more job than before	30%	26%	0.499	23%	19%	0.452	16%	14%	0.580	13%	9%	0.383	30%	18%	0.067	0.830
Base: all currently looking for work	194	150	–	244	192	–	256	200	–	140	129	–	125	116	–	–

Table 19: Impacts of the Work and Health Programme by length of unemployment: mandatory participants

Unemployment/health status	Less than 3 years: WHP participants	Less than 3 years: control group	Less than 3 years: p-value	3 years or more, or never worked: WHP participants	3 years or more, or never worked: control group	3 years or more, or never worked: p-value	p-value for differential impact
#% Worked at all since allocation	29%	29%	0.991	12%	11%	0.760	0.833
% Currently working	20%	20%	0.937	9%	5%	0.142	0.296
% Thinking support increased chances of moving into work to a large or some extent	61%	59%	0.725	60%	65%	0.335	0.391
% Very or fairly confident completing job applications/CV	87%	82%	0.382	74%	79%	0.214	0.166
% Satisfied with life (7+ on scale from 0 (not at all) to 10 (completely))	50	44	0.342	47	57	0.056	0.060
% Good health (7+ on scale from 0 (very poor) to 10 (very good))	55	51	0.596	46	48	0.706	0.515
Base: all	481	119	–	741	213	–
% Applying for more job than before	25%	13%	0.063	19%	17%	0.681	0.210
Base: all currently looking for work	311	81	–	494	147	–	–

Table 20: Impacts of the Work and Health Programme by gender: voluntary participants

Employment/health status	Men: WHP participants	Men: control group	Men: p-value	Women: WHP participants	Women: control group	Women: p-value	p-value for differential impact
% Worked at all since allocation	26%	20%	0.002*	28%	25%	0.247	0.285
% Currently working	16%	14%	0.147	22%	18%	0.182%	0.977
% Thinking support increased chances of moving into work to a large or some extent	61%	55%	0.012*	61%	51%	0.002*	0.377
% Very or fairly confident completing job applications/CV	75%	73%	0.177	67%	70%	0.287	0.093
% Satisfied with life (7+ on scale from 0 (not at all) to 10 (completely))	42%	37%	0.030*	43%	38%	0.166	0.795
% Good health (7+ on scale from 0 (very poor) to 10 (very good))	36%	37%	0.558	38%	36%	0.688	0.498
Base: all	1222	997	–	790	580	–	–
% Applying for more job than before	23%	18%	0.062	23%	16%	0.041*	0.541
Base: all currently looking for work	703	578	–	328	278	–	–

Table 21: Impacts of the Work and Health Programme by gender: mandatory participants

Type of impact	Men: WHP participants	Men: Control group	Men: p-value	Women: WHP participants	Women: Control group	Women: p-value	p-value for differential impact
% Worked at all since allocation	19	14	0.007*	18	21	0.545	0.211
% Currently working	13	5	0.650	13	16	0.577	0.025*
% Thinking support increased chances of moving into work to a large or some extent	62	59	0.731	59	62	0.615	0.499
% Very or fairly confident completing job applications/CV	82	81	0.758	75	79	0.485	0.448
% Satisfied with life (7+ on scale from 0 (not at all) to 10 (completely))	49	50	0.501	46	51	0.470	0.694
% Good health (7+ on scale from 0 (very poor) to 10 (very good))	50	47	0.588	48	49	0.832	0.566
Base: all	816	225	–	503	121	–	–
% Applying for more job than before	20	16	0.390	23	11	0.056	0.276
Base: all currently looking for work	546	170	–	308	69	–	–

Table 22: Impacts of the Work and Health Programme by age: voluntary participants

Employment/health status	Aged up to 34: WHP participants	Aged up to 34 : control group	Aged up to 34: p-value	Aged 35 to 49: WHP participants	Aged 35 to 49: control group	Aged 35 to 49: p-value	Aged 50 or over: WHP participants	Aged 50 or over: control group	Aged 50 or over: p-value	p-value for differential impact
% Worked at all since allocation	36%	29%	0.088	30%	23%	0.035*	20%	16%	0.072	0.912
% Currently working	23%	22%	0.714	22%	16%	0.040*	14%	12%	0.270	0.451
% Thinking support increased chances of moving into work to a large or some extent	67%	60%	0.033*	64%	53%	0.004*	55%	49%	0.064	0.440
% Very or fairly confident completing job applications/CV	75%	75%	0.861	73%	71%	0.565	70%	70%	0.958	0.847
% Satisfied with life (7+ on scale from 0 (not at all) to 10 (completely))	43%	43%	0.906	45%	38%	0.056	40%	34%	0.020*	0.257
% Good health (7+ on scale from 0 (very poor) to 10 (very good))	41%	44%	0.456	41%	38%	0.344	31%	32%	0.677	0.449
Base: all	516	403	–	581	421	–	914	753	–	–
% Applying for more job than before	32%	24%	0.089	26%	19%	0.113	15%	13%	0.393	0.726
Base: all currently looking for work	280	209	–	289	225	–	462	423	–	–

Table 23: Impacts of the Work and Health Programme by age: mandatory participants

Employment/health status	Aged up to 34: WHP participants	Aged up to 34 : control group	Aged up to 34: p-value	Aged 35 to 49: WHP participants	Aged 35 to 49: control group	Aged 35 to 49: p-value	Aged 50 or over: WHP participants	Aged 50 or over: control group	Aged 50 or over: p-value	p-value for differential impact
% Worked at all since allocation	29%	35%	0.438	23%	17%	0.388	11%	9%	0.417	0.409
% Currently working	20%	19%	0.865	15%	11%	0.479	8%	4%	0.069	0.546
% Thinking support increased chances of moving into work to a large or some extent	68%	71%	0.715	62%	67%	0.511	56%	52%	0.466	0.58
% Very or fairly confident completing job applications/CV	85%	87%	0.706	79%	74%	0.457	76%	80%	0.440	0.559
% Satisfied with life (7+ on scale from 0 (not at all) to 10 (completely))	59%	56%	0.713	48%	35%	0.105	43%	57%	0.007*	0.012*
% Good health (7+ on scale from 0 (very poor) to 10 (very good))	60%	66%	0.428	56%	38%	0.024*	40%	45%	0.320	0.035*
Base: all	296	85	–	392	85	–	630	179	–	–
% Applying for more job than before	32%	19%	0.130	26%	16%	0.212	13%	14%	0.899	0.419
Base: all currently looking for work	184	58	–	264	62	–	405	120	–	–

1. See Improving lives: the future of work, health and disability ↩

2. See Work and Health Programme statistics to May 2023 - GOV.UK (www.gov.uk) ↩

3. Note: The Wave One participant survey did not focus on employment outcomes given participants were interviewed shortly after they joined the programme. ↩

4. Table 8 in Appendix Five shows that the impact on having been in any work since allocation was split between participants having worked full-time (30 hours or more per week) or part-time (fewer than 30 hours). ↩

5. This positive impact was due to Disability WHP participants. EAG WHP participants were not significantly more likely to be in work than their control group. ↩

6. See Department for Work and Pensions’ settlement at the Spending Review - GOV.UK (www.gov.uk) ↩

7. See Improving lives: the future of work, health and disability ↩

8. It should be noted that random allocation was paused in March 2020 during the onset of the COVID-19 pandemic. ↩

9. See Work and Health Programme Provider Guidance, chapter 1: introduction and overview ↩

10. Providers receive payments to cover their agreed costs to deliver the provision plus an agreed profit margin. The cost-plus contracts still have performance management requirements with options for DWP to withhold an amount of the payment if delivery falls below acceptable standards. ↩

11. See Work and Health Programme statistics to May 2023 ↩

12. See Work and Health Programme statistics to May 2023 ↩

13. Weighting was applied to account for the purposive over-sampling of WHP participants and control cases during these months. Further details are provided in Appendix One. ↩

14. Due to a coding error, 228 respondents did not have gender or age recorded. These cases have been excluded from the impact analysis. ↩

15. Noting that random allocation was paused in March 2020 during the onset of the COVID-19 pandemic. ↩

16. See Work and Health Programme statistics to November 2021 ↩

17. ‘Welfare to work’ was used in this context to refer to previous government programmes or initiatives (such as the Work Programme) which aimed to help unemployed people find and sustain work. ↩

18. Disability Confident is a government scheme designed to encourage employers to recruit and retain disabled people and those with health conditions. For more details, see: Disability Confident Campaign ↩

19. Other than in Chapter Six, where outcomes are considered, the control group data is shown for the group as a whole (the comparison groups for both voluntary and mandatory groups). As voluntary participants make up a comparatively larger portion of WHP participants (and their respective control group), the control group tends to look more similar to the voluntary group. In Chapter Six, WHP participants are compared with their respective mandatory and voluntary control groups. ↩

20. Figures for the control group are not charted here for either question as these are very similar to the responses given by voluntary WHP participants ↩

21. The response percentages for each answer code have not been reported due to low base sizes. The base sizes were only sufficient to report on the recurring themes from the open-ended question. ↩

22. Due to a coding error, 228 respondents did not have gender or age recorded. These cases have been excluded from the impact analysis. See Chapter One and Appendix Two for fuller details of the impact analysis method. Impact analysis was not carried out elsewhere on the survey data as this would not have been feasible within the evaluation budget. ↩

23. Table 8 in Appendix Five shows that the impact on having been in any work since allocation was split between participants having worked full-time (30 hours or more per week) or part-time (fewer than 30 hours). ↩

24. This positive impact was due to Disability WHP participants. EAG WHP participants were not significantly more likely to be in work than their control group. ↩

25. There were too few mandatory control group participants in work (n=29) to produce impact estimates for the mandatory group. ↩

26. There were too few mandatory control group participants in work (n=29) to produce impact estimates for the mandatory group. ↩

27. This positive impact was due to Disability and Health WHP participants. EAG WHP participants were (non-significantly) less likely to be applying for more jobs than their control group. ↩

28. Table 13 in Appendix Five provides the responses to the full scales which also show no statistically significant impacts of the WHP. ↩

29. Table 13 in Appendix Five provides the responses to the full scales which, also show no statistically significant impacts of the WHP. ↩

30. Source of UK general populationstatistics ↩

31. Note that anxiety is a negative emotion in contrast to the previous three positive emotions, so the scale works in a reverse order ↩

32. This positive impact was due to Disability and Health WHP participants. EAG WHP participants were (non-significantly) less likely to be satisfied than their control group. ↩

33. There were too few mandatory control group participants in work (n=29) to produce impact estimates. ↩

34. With the exception of the ONS measure on anxiety and perceptions of the extent to which their condition affects their capacity to work, which show the percentages with a negative outcome. ↩

35. Life wheels are tools commonly used by professional life coaches in a range of contexts to help individuals consider different aspects of their lives in turn, and identify areas where support is needed or as foci for continued development. They feature a series of scaled axes, the ‘spokes of the wheel’, which allow a visual representation of where individuals’ capabilities are strong and where additional support may be required, as well as showing participants’ the progress they have made over time. ↩

36. See Flexible Support Fund ↩

37. P-value based on ordinal test <0.001*. ↩

38. P-value based on ordinal test 0.014*. ↩

39. P-value based on ordinal test <0.001*. ↩

40. P-value based on ordinal test 0.384. ↩

41. P-value based on ordinal test 0.001*. ↩

42. P-value based on ordinal test 0.488. ↩

43. P-value based on ordinal test 0.602. ↩

44. P-value based on ordinal test 0.491. ↩

45. P-value based on ordinal test 0.831. ↩

46. P-value based on ordinal test 0.714. ↩

47. P-value based on ordinal test 0.169. ↩

48. P-value based on ordinal test 0.039*. ↩

49. P-value based on ordinal test 0.926. ↩

50. P-value based on ordinal test 0.816. ↩

51. P-value based on ordinal test 0.257. ↩

52. P-value based on ordinal test 0.806. ↩

53. P-value based on ordinal test 0.992. ↩

54. P-value based on ordinal test 0.690. ↩

55. P-value based on ordinal test 0.440. ↩

56. P-value based on ordinal test 0.582. ↩

57. P-value based on ordinal test 0.952. ↩

58. P-value based on ordinal test 0.828. ↩

59. P-value based on ordinal test 0.795. ↩

60. P-value based on ordinal test 0.313. ↩

61. P-value based on ordinal test 0.751. ↩

62. P-value based on ordinal test 0.367. ↩