Evaluation of the SEND and AP change programme - interim evaluation summary

Evaluation of the SEND and AP change programme Interim evaluation summary January 2026

Authors: Authors: Priya Cooper, Ali Zaidi, Beth Smiles, Icf Consulting Services Ltd

Introduction

This summary report presents the interim findings from an evaluation of the Special Educational Needs and Disabilities (SEND) and Alternative Provision (AP) Change Programme (hereafter ‘Change Programme’). The main aims of the evaluation are to assess whether the Change Programme is being implemented as intended, the effectiveness and quality of testing activities, and whether the initiatives and changes being tested are on track to achieve their intended outcomes. The evaluation is being undertaken by Icf Consulting Services Ltd, in partnership with Rsm UK Consulting LLP1 and Bmg Research.

The Change Programme was established by the Department for Education (DfE) to test and refine potential reforms outlined in the SEND and AP Improvement Plan (March 2023), which outlined a long-term vision for building a more inclusive, consistent, and financially sustainable system for children and young people with SEND and those requiring AP. The Change Programme launched in September 2023 and will run until August 2026.

The programme is being implemented through 9 Change Programme Partnerships (Cpp), each composed of 2 to 4 local authorities.

Table 1: List of all participating Cpp local areas

Change Programme Partnership Region Local authorities

East of England Bedford, Central Bedfordshire, Luton

East Midlands (referred to as ‘Llr’) Leicestershire, Leicester City, Rutland

London Barnet, Camden, Enfield, Islington

North East Hartlepool, Durham, Gateshead, Stockton-on- Tees

North West Manchester, Oldham, Rochdale, Trafford

South East Portsmouth, East Sussex, West Sussex, Brighton and Hove

South West Gloucestershire, Swindon

West Midlands Telford and Wrekin, Shropshire, Herefordshire, Worcestershire

Yorkshire and Humber Wakefield, Bradford, Leeds, Calderdale

1 The Rsm project team moved over to Economic Research Services Limited under the new brand name Fortia Insight in November 2025.

DfE appointed the Reaching Excellence and Ambition for all Children (REACh) Consortium as its national delivery partner to oversee testing and capture insights from all Cpp local areas2.

Evaluation approach

Over the first 2 years of evaluation, a range of qualitative and quantitative data has been collected to provide a rounded picture of implementation progress, local delivery experiences, and emerging learning. This includes:

• Two waves of a Cpp local area survey undertaken in 2024 (154 responses) and 2025 (74 responses). The survey was distributed to local partnership boards.

• A ‘push to web’ survey of families with children with special education needs, conducted in 2024 (2,237 responses). The survey sample was not limited to families involved in, or directly affected by, the programme.

• A survey of education settings undertaken in 2025 (453 responses). Mainstream schools accounted for the majority of responses (90%), followed by AP providers (4%), special schools (4%), hospital schools (1%), and colleges (1%). Respondents were not limited to schools that had been involved in the programme directly.

• A total of 267 semi-structured interviews over 2 years, including 221 interviews with local area stakeholders (with some stakeholders interviewed more than once), 35 with families, and 11 with national stakeholders.

• A document and data review, covering progress and highlight reports from REACh, local planning and implementation documents, and national policy documentation.

Implementation of the Change Programme

The programme launched in September 2023 and will conclude in August 2026. After a set-up phase (September to December 2023), Phase 1 testing began in January 2024, focusing on 11 initiatives across 4 ‘blocks’:

• Partnerships and Plans (‘Enablers’): Establishing the foundational elements needed for the SEND and AP system to work, including partnership arrangements, local plans (through testing a Local Area Inclusion Plan (Laip)) and use of data (through testing a prototype Inclusion Dashboard).

• Education, Health and Care (EHC) Plan change testing: Cpp local areas were asked to test a standardised EHC Plan and advice templates, explore process

2 The REACh Consortium is made up of Pa Consulting, Impower, Council for Disabled Children and Olive Academies.

changes to multi-agency panels and mediation, and trial Advisory Tailored Lists to help families make informed choices about placements3.

• Inclusive Mainstream Practice (formerly Ordinarily Available Provision): This block was designed to improve support within mainstream settings through early identification and intervention. Cpp local areas were asked to pilot early intervention approaches, work to a 3-tier AP model4, and test workforce models such as the Alternative Provision Specialist Taskforce (Apst)5 and the Early Language Support for Every Child (Elsec) pathfinder pilot6.

• Standards and Commissioning: This block was intended to support the development of National Standards and Bands and Tariffs to promote greater consistency in SEND and AP provision. Activities under this block were later paused pending national policy decisions.

In December 2024, DfE decided to prioritise and expand work on Inclusive Mainstream Practice, reflecting wider policy goals to identify needs earlier, and to ensure mainstream settings are more inclusive and better able to support children with additional needs. This marked the start of Phase 2 of the Change Programme.

Phase 2 placed greater emphasis on helping Cpp local areas identify what works best in their own contexts, using local learning to shape approaches to inclusion and early intervention. Each area was asked to develop a Local Delivery Plan, outlining how they would continue or build on elements of Phase 1 work (such as strengthening their 3-tier AP models and improving co-production mechanisms) alongside broader local priorities.

The Change Programme was formally extended in April 2025 and will continue until August 2026. During this Extension Phase, a new and expanded set of testing activities has been introduced, including:

• Delivering prototype Local Inclusion Support Offers (Liso);

• Piloting Assistive Technology (At) Lending Libraries;

• Collecting feedback from areas on SEN Units and Resourced Provisions;

• Extending the 3-tier AP model, Apst and Elsec programmes; and

4 The AP three-tier model is based on the following tiers; i) an outreach service in mainstream schools to tackle problems early and enable children and young people to remain and thrive in their mainstream school; ii) short-term interventions in AP settings focused on supporting the children and young people to re- integrate back into their school; and (iii) longer-term interventions for a much smaller cohort of children and young people that need more specialist support to get them back on track and reintegrate successfully back into mainstream – at a different school or, for older young people, in post-16 education/training. 5 The Apst is a workforce model where teams of co-located specialists (such as therapists and family support workers) are based onsite within schools to provide integrated, child-centred work with pupils. It was initially tested in a separate DfE-led pilot programme. An initial evaluation report from the Apst Programme was completed in July 2025: Alternative Provision Specialist Taskforce Impact, process and cost evaluation of Years 1 & 2 6 A separate interim evaluation of Elsec, to be published separately by DfE, provides a full assessment of progress to date.

• Integrating Partnerships for Inclusion of Neurodiversity in Schools (Pins) within local plans.

The final evaluation report will focus on the implementation and outcomes of these Extension Phase activities.

Progress with testing changes

The Change Programme has developed over time into an effective mechanism for testing proposed changes and initiatives across participating local areas. After a gradual start, the programme’s test-and-learn approach has become more established, supporting more structured testing, clearer feedback loops, and stronger learning across local areas. Several core initiatives have been tested at scale, with considerable feedback into the conditions under which different initiatives and changes can succeed, the challenges that remain, and the types of adaptations that are likely to be needed for any potential national rollout. The rest of this section provides an overview of progress in Phase 1 and Phase 2 of the programme.

Phase 1

Extent of testing

In Phase 1, the scale and depth of testing varied across local areas. Some initiatives and changes were tested widely and in detail, while others were trialled more selectively or remained at an early, exploratory stage.

Cpp local areas generally focused first on ‘enabler’ activities (such as establishing SEND and AP Partnerships and developing LAIPs) as these provided the essential building blocks for testing other changes. Most areas made steady progress in implementing partnerships, although some boards required adjustments to membership or governance arrangements to work effectively. However, developing the LAIPs took longer than expected due to the complexity of the documents, the need to gather and analyse data from multiple agencies, coordinate input from a wide range of stakeholders, co-produce plans with families, and secure local approvals. Although this extended timelines, these steps were essential for creating a shared understanding of local needs and priorities.

The EHC Plan standardised template was another major focus of Phase 1 and was tested widely across most Cpp local areas. By December 2024, the template had been trialled in 30 out of 32 Cpp areas, reaching over 5,700 families. Testing covered a range of contexts, including new EHC Plan applications, some annual reviews, and transitions between primary and secondary schools.

In contrast, the extent to which advice templates and proposed changes to multi-agency panels and mediation processes were tested was more limited. Guidance for these areas was released later in the programme, and some local areas reported finding it difficult to

prioritise them alongside other requests. The Advisory Tailored Lists initiative was also trialled on a smaller scale than planned. While the concept of supporting families to make more informed choices about placements was welcomed, local areas raised significant concerns about how Advisory Tailored Lists would be kept up to date, how parental preference would be respected in practice, and how to avoid creating unrealistic expectations that inclusion on a list meant a school place was available or likely, given ongoing eligibility and capacity constraints.

Work under the Inclusive Mainstream Practice block in Phase 1 focused primarily on laying the foundations for system-level changes. By the end of Phase 1, most Cpp local areas had begun developing inclusion frameworks, self-assessment tools, and improvement plans, with some launching early initiatives to test approaches and strengthen understanding of inclusive practice. Additionally, by December 2024:

• 17 Cpp local areas had implemented or embedded the 3-tier AP model, with others planning or starting implementation. Progress varied by starting point, with most areas reviewing provision, clarifying the commissioning role of schools, improving monitoring, and developing quality assurance frameworks. Some areas introduced new provision focused on Tier 1 and 2, such as outreach, step-out placements, reintegration support, and alternative curricula.

• The Apst model was supported in most areas, with 24 of 32 either operating or planning to implement an Apst. Testing remained early-stage, focused on establishing the model, recruiting staff, and securing cross-sector support. In a few cases, local areas drew on established practice, partnerships and learning from existing APSTs in neighbouring areas.

• The Elsec programme was live in all pathfinder sites, reaching nearly 6,000 children with universal support, over 1,200 with targeted interventions, and training around 800 staff to identify and deliver Slcn support.

The variation in the extent and depth of testing described above also reflected a common set of implementation constraints, including:

• Recruitment delays for project leads and specialist staff, which reduced the time available for testing and co-design.

• The late sequencing of some testing asks and guidance, which created uncertainty about what Cpp local areas were expected to test and led to multiple initiatives being introduced simultaneously, resulting in competing priorities.

• Differing levels of local SEND and AP capacity and readiness to test and implement changes locally, as well as competing pressures.

• External factors, including the pre-election period in mid-2024 and the subsequent change of government, further disrupted momentum and required some areas to pause or re-prioritise activity.

Feedback collected

Multiple channels were used to gather feedback on the changes being tested, including local area reports, stakeholder interviews, surveys, and parent and carer input. Feedback served a range of purposes from guiding local implementation, providing early insights on usability and local adaptation, and informed policy-level learning. The depth and consistency of feedback differed across initiatives.

For the foundational ‘enabler’ changes, most Cpp areas provided insights on Local SEND and AP Partnerships and LAIPs.

Changes to local partnership working were broadly welcomed, with stakeholders noting that they strengthened governance, co-production mechanisms, and strategic oversight, and helped boards build on existing practice and improve relationships with forums and partners. At the same time, testing highlighted areas for improvement, particularly around the practical operation of SEND and AP Partnership Boards and the use of the standardised Terms of Reference template (developed for testing through the Change Programme). This feedback was largely formative and provided valuable insights to inform DfE thinking on how boards can function most effectively.

Testing of LAIPs generated substantial feedback on their content, usability, and co- production processes. Stakeholders were generally supportive of the idea of having a plan to clarify local priorities and provide an evidence-informed framework, but the drafting process highlighted several challenges. These included difficulties in engaging partners, schools, and families, accessing and presenting data, and managing the overall complexity of the process. This feedback was provided back to DfE to inform future thinking.

The Inclusion Dashboard generated more limited feedback, as its restricted rollout meant only a subset of intended users were able to test its usability and utility.

Feedback on changes linked to the EHC Plan process varied. Stakeholders across all Cpp local areas provided considerable feedback and practical insights on the usability, implementation and potential improvements to the EHC Plan template. This feedback helped DfE understand how these resources function in practice and where improvements may be needed, though some local stakeholder reported uncertainty about whether their feedback had been fully considered at a national level. Feedback on the accompanying advice templates was more limited, reflecting later rollout and inconsistent use, but where collected, it highlighted usability challenges, adaptation to local processes, and professional concerns around word limits.

Where changes were tested, formative feedback on multi-agency panels and mediation processes highlighted both challenges and practical lessons. For multi-agency panels, feedback included challenges in increasing parental engagement, managing workloads for panel members, and adapting processes to local contexts. Stakeholders also noted that

increased representation on panels was beneficial, but workload pressures remained a key challenge, and one-size-fits-all approaches were not always effective. Feedback on strengthening mediation practices highlighted a lack of trust in existing processes and practical challenges for local authorities, including capacity constraints and prioritisation of dispute prevention. Insights from local areas informed adjustments such as offering parents a choice of mediation providers, collaborating with external services, introducing pre-mediation support, and revising commissioning practices.

Whilst most local areas chose not to implement Advisory Tailored Lists, they provided feedback to DfE on the reasons for this to inform their future planning. Where lists were implemented, stakeholders shared learning from their experience, suggesting ways to make them more user-friendly, including incorporating parent feedback, adding functionality for families to explore options, and building on existing local resources.

Feedback across planned changes and initiatives under the Inclusive Mainstream Practice block was largely formative by the end of Phase 1, reflecting the early-stage and preparatory nature of testing. For example:

• Cpp local areas reported that resources such as inclusion frameworks, self- assessment tools, and improvement plans helped clarify local priorities and identify areas for development. Early feedback highlighted practical learning on how frameworks and tools could be used to support schools in embedding inclusive practice, and provided insights on areas where additional guidance, capacity, or resources might be needed.

• Feedback on the 3-tier AP model highlighted practical learning around service design, monitoring, and quality assurance including for non-school alternative provision (Nsap) providers, as well as challenges in adapting provision to local context. Where new provision was introduced, early insights were being captured on the benefits of structured step-out placements, outreach services, reintegration and transition support.

• Feedback from areas adopting the Apst and Elsec models provided insights into early operational arrangements, workforce planning, and cross-sector collaboration. Stakeholders highlighted the benefits of integrated, multi-agency approaches, including the ability to provide intensive, co-ordinated support for pupils. There were also reported implementation challenges in recruiting and retaining specialists.

Phase 2

Extent of testing

Phase 2 marked a clear shift from planning to implementation, with most Cpp local areas beginning to embed inclusive education approaches across mainstream settings. Activity increasingly focused on testing, adapting, and scaling the models developed in Phase 1.

However, uncertainty over continuation funding and delays in the extension decision constrained planning, recruitment, and delivery in early 2025. Capacity pressures across local authorities, schools, and AP providers also persisted, with staff balancing programme activity alongside core statutory responsibilities.

Despite these challenges, many areas made tangible progress across the 3-tier AP model. Tier 1 work centred on strengthening support within mainstream schools through inclusion coaching, behaviour units, and outreach services. Tier 2 testing focused on creating more consistent, structured “step-out” provision and reintegration pathways, replacing ad hoc arrangements. A smaller number of areas advanced work under Tier 3 by improving commissioning oversight and developing quality assurance frameworks for AP providers. Testing of the Apst model also expanded, though recruitment and retention of specialist staff (particularly for education psychology, speech and language therapy, and mental health roles) remained a significant barrier in several Cpp areas.

Feedback collected

Feedback to date has indicated growing confidence in the 3-tier AP approach and the ability to improve inclusive mainstream practices in schools. Local leads reported early signs of more consistent practice across schools, with inclusion toolkits, staff coaching, and “step-out” provision helping mainstream settings retain pupils who might otherwise have been referred to AP. In several Cpp areas, schools reported improved collaboration and shared ownership of inclusion through clearer expectations and the use of common templates or assessment tools.

Stakeholders also noted positive impacts from early Apst activity, particularly where APSTs were embedded within multi-agency teams or short-term intervention centres. These approaches improved coordination between education, health, and care partners and enabled quicker access to specialist support such as speech and language or mental health input. However, recruitment and retention difficulties, especially for specialist roles, limited reach and consistency in some areas.

Despite these achievements, many local stakeholders stressed that sustaining momentum would require ongoing investment, workforce stability, and clear national guidance to embed the gains made and scale effective practice across all areas.

Emerging outcomes

Partnerships and Plans (‘Enablers’)

Early evidence suggests the programme has strengthened local collaboration, co- production and evidence-informed decision making. Stakeholders reported that partnership arrangements as “maturing” across Cpp local areas, with improved dialogue clearer shared priorities, and more structured meetings across education, health and local

authority teams. Engagement from social care and AP providers remains more variable, highlighting an area for further focus.

Cpp local areas have also implemented improved co-production practices with families and Parent Carer Forums. Local partnerships have developed new mechanisms (such as parent-led panels and co-chaired workstreams) to embed lived experience in planning and service design. However, the strength of co-production still varies, and some family representatives highlighted the need for clearer roles and sustained support beyond the Change Programme.

There are also early signs of more evidence-based planning and local decision-making. Stakeholders noted that LAIPs and other data tools have improved information-sharing, oversight, and transparency, helping partners identify gaps, monitor progress, and make more informed decisions.

EHC Plan change testing

Families generally welcomed the new national EHC Plan template, describing it as clearer, more manageable and less overwhelming. Caseworkers similarly valued the logical structure and online accessibility, which, in their view, supported better information capture and joint working. Many Cpp areas have continued to use the template, often with small local adaptations, reporting greater plan quality and more consistent decision-making.

Where process changes have been implemented, multi-agency panels have also become more collaborative and transparent, with clearer criteria and stronger professional input across education, health and care. Stakeholders noted that these changes have improved both the timeliness and quality of decisions, strengthened governance, and increased confidence in the system. However, in some areas this did not differ substantially from what they had in place prior to the Change Programme.

Some areas have continued to focus on strengthening mediation practices even after testing paused: several Cpp areas have embedded new liaison or mediation coordinator roles, introduced pre-mediation meetings, or strengthened family engagement. One area reported that of 40 mediation requests, 12 were resolved without escalation and only 2 progressed to appeal, demonstrating tangible impact in reducing disputes and building trust.

While it remains too early to evidence impact on longer-term outcomes, Cpp local area stakeholders were cautiously optimistic that stronger mediation processes, more consistent panels, and better-quality plans would lead to lasting improvement in families’ experiences and children’s outcomes.

Inclusive Mainstream Practice

Evidence from across Cpp local areas indicates steady progress in embedding more inclusive practice within mainstream settings. Schools are increasingly taking collective responsibility for meeting SEND needs, supported by new tools, guidance, and local initiatives such as inclusion frameworks, resource bases, and the 3-tier AP model. Stakeholders reported early signs of impact following local implementation, including more pupils being retained in mainstream education, fewer exclusions, and stronger collaboration between schools, local authorities, and AP providers.

Staff capability also appears to be improving. In some areas, training programmes such as Autism Education Trust modules, Elklan communication training, and Pins inclusion training have been rolled out, helping teachers manage diverse needs and reduce reliance on specialist services.

APSTs are also showing positive early impact. Across several Cpp local areas, APSTs have played a key role in supporting reintegration from AP back into mainstream settings and preventing escalation into exclusion. Stakeholders reported that the co-location of Apst specialists within schools (a departure from the original pilot model) has enabled earlier identification of emerging Semh needs and faster access to targeted interventions, helping pupils remain engaged in learning.

Confidence in the 3-tier AP model remains high across Cpp local areas, with stakeholders and AP providers describing it as a clear, flexible framework that supports collaboration between schools, Pupil Referral Units, and local authorities. The model is valued for promoting early intervention through Tier 1 support, helping to prevent escalation into higher tiers, and providing structured pathways for reintegration.

There is also growing optimism that with continued investment and clearer guidance, the inclusive approaches developed through the Change Programme can be embedded and scaled nationally, strengthening early intervention and reducing demand for specialist support over time.

Management of the programme

National support offer

Over the course of the programme, trust between DfE, REACh, and local areas strengthened, creating an environment that supported testing, learning, and adaptation.

At the national level, DfE has provided guidance, set expectations, and offered workshops to help local areas implement changes. In Phase 1, national guidance was often perceived as high-level and unclear. Local areas reported difficulty understanding testing requirements, timescales, and key activities. The volume of communications across multiple initiatives/areas of testing, combined with delays in updates, sometimes slowed

implementation and caused frustration. However, support from DfE was broadly perceived to have become more coherent and responsive over the course of the programme. Workshops and information sessions provided opportunities for local areas to ask questions and test approaches suited to their context. Regional teams supporting CPPs were also widely praised for being accessible, visible, and responsive, helping local areas adapt initiatives and proposed changes to their local priorities.

Further support has been provided through DfE’s commissioned delivery partner, REACh. In Phase 1, a dedicated team offered strategic advice, acted as a ‘critical friend’, and helped areas test changes. Local stakeholders reported broadly positive experiences of working with REACh, highlighting that support was responsive, practical, and helped them problem-solve. Over time, communication from REACh to all Cpp stakeholders was generally reported to have improved, with monthly newsletters and other regular updates proving helpful. In the final stages of the programme, REACh introduced Change Coaches, who act as the main point of contact for local areas, participate in leadership meetings, and help teams reflect on progress. The effectiveness of this support, and how it contributes to local implementation, will be explored in the next stage of the evaluation.

Feedback loop

An integral feature of the Change Programme has been the feedback loop, managed by REACh in partnership with DfE, which captures learning from local implementation and informs national policy development. Stakeholders reported that it enabled them to share observations and challenges in near real-time, supporting iterative improvements and more responsive programme delivery.

In Phase 1, feedback was collected through structured cycles, including surveys and focus groups. While effective at gathering information, many stakeholders expressed a desire for greater clarity on how their input was used at the national level, noting that links between local feedback and resulting decisions were not always visible.

In Phase 2, the feedback loop evolved to capture learning in a more flexible and responsive way. Regional workshops, focus groups, and Practice Sharing Forums enabled local areas to provide richer, more context-specific insights. These were summarised in concise practice notes, highlighting key findings and actionable recommendations, which could be shared efficiently with DfE.

The effectiveness of the feedback loop will be examined fully in the final report.

Working as a Cpp

The Cpp model remains a central strength of the Change Programme. It has provided a structured platform for collaboration, enabling local areas to convene, share learning, and collectively address complex challenges. Dedicated project leads within each Cpp, often

embedded in lead local authorities, provide strategic oversight, coordinate activities, and provide a link back to DfE and REACh.

However, some challenges have been noted. Areas entered the programme with varying levels of partnership maturity and differing approaches to implementing changes, which meant that not all local areas progressed at the same pace. These differences sometimes limited the immediate impact of working together as a partnership, and also required flexibility in oversight and support to ensure all areas could embed changes successfully.

Despite these challenges, stakeholders consistently highlighted the value of having protected time and structured mechanisms to share practice and learn from neighbouring areas, with many describing this way of working as a valuable legacy of the programme.

Next steps for the evaluation

A final evaluation report will be produced following the completion of the Change Programme. This report will build on learning to date and incorporate additional evidence from multiple sources, including:

• Interviews with 45 education settings and 65 families (November 2025 to February 2026)

• Visits to 9 educational settings across different Cpp local areas (January to February 2026)

• In-depth fieldwork in Cpp local areas (February to April 2026) to provide a detailed understanding of local implementation, emerging outcomes, and contextual factors shaping the programme.

• Follow-up surveys of education settings (February to March 2026) and families (April-May 2026)

• Final interviews with national and Cpp local area stakeholders at the end of the programme to reflect on outcomes, challenges, and lessons learned.

In parallel, a feasibility impact assessment is being developed. This will explore the potential for using a quasi-experimental design to measure programme impacts. A separate feasibility report will outline the options and provide recommendations for implementing this approach. If deemed feasible, an impact assessment will be included in the final evaluation report.

This publication is licensed under the terms of the Open Government Licence v3.0, except where otherwise stated. To view this licence, visit nationalarchives.gov.uk/doc/open- government-licence/version/3. Where we have identified any third-party copyright information you will need to obtain permission from the copyright holders concerned. Reference: RR1600 Isbn: 978-1-83870-744-6

For any enquiries regarding this publication, contact www.gov.uk/contact-dfe.

This document is available for download at www.gov.uk/government/publications.

Why this page, not just a PDF? The original is a PDF - fine to read, but a poor fit for assistive technology: heading structure is often inconsistent or absent, text doesn't reflow for zoom or a narrow screen, and screen readers can struggle with multi-column layouts and scanned pages. This page is the same content rebuilt as plain, semantic HTML - real headings, text that reflows, nothing a screen reader or zoom tool trips over.

Republished from Department for Education under the Open Government Licence v3.0. View the original PDF.