EHCP evidence pack: what every section needs

An EHCP evidence pack is the structured submission a school sends to the local authority when applying for, or contributing to, an Education Health and Care Plan. It's the document that turns "this child needs more support" into "here's the evidence the support is needed." Get it right and the LA has what they need to write a strong EHCP. Get it wrong and the application bounces, the timeline slips, and the child waits longer.

I've written more of these by hand than I can count. The pattern is always the same: a SENDCo at a desk on a Friday afternoon, six tabs open, a Word document slowly accumulating, paragraphs cut and pasted from five different systems. It takes three to four hours per pupil. With a typical secondary school caseload, that's a working week per term, just on the EHCPs that need refreshing.

This article walks through the ten sections that make up a good evidence pack, what each one needs, and how to find the data without losing your evening.

The ten sections of a strong evidence pack

Most LAs will accept different formats, but every reasonable evidence pack includes the same ten elements. Some of these come from your MIS automatically. Some you have to build up over time. The good news: if you're keeping good ISPs and APDR records, most of the pack is already written.

1. Attendance summary

The LA wants to see whether attendance is a barrier to learning. They'll look for:

• Overall attendance percentage (current term, current year, historical)
• Authorised vs unauthorised absence
• Lateness pattern
• Lesson-level attendance if your MIS records it (some pupils attend registration but skip specific lessons)

The trend matters more than the headline number. A pupil at 92% attendance with a downward trend tells a different story from a pupil at 92% with a flat or rising trend.

This section comes straight from your MIS. The challenge isn't the data — it's pulling it cleanly into a format the LA can read. SIMS, Arbor and Bromcom all produce attendance reports, but none of them produces a SEND-friendly summary. You end up exporting to Excel and reformatting.

2. Behaviour and exclusions

The LA wants behaviour data that contextualises need. Not just incident counts, but trends, severity, and patterns.

Include:

• Behaviour incidents over the relevant period (last 12 months, by category)
• Severity breakdown (low, medium, high)
• Patterns (time of day, subject, peer group, time of year)
• Any fixed-term exclusions with dates and reasons
• Any internal exclusions or seclusions
• Pastoral support put in place

Two important framing points. First, behaviour data without context is dangerous. A high incident count for a pupil with a communication need can look like a "bad pupil" instead of a "pupil whose needs aren't being met." The pack needs to frame the behaviour data within the picture of need.

Second, the LA will spot if behaviour data is suppressed or selectively presented. Be honest. If there have been exclusions, list them. Then explain what the school did differently afterwards.

3. Graduated approach (APDR cycles)

This is the single most important section in the pack. The graduated approach — Assess, Plan, Do, Review — is what the LA needs to see to fund an EHCP. The expectation is at least two completed cycles before the EHCP application. Some LAs ask for three.

Each cycle in the pack should show:

• What was assessed and how (observations, parent input, professional input)
• What was planned (provision, outcomes, target dates)
• What was done (actual delivery, including any changes)
• What was reviewed (outcomes RAG-rated, what worked, what didn't)
• What changed for the next cycle

This is where SENDCos who don't have a structured ISP system suffer. The graduated approach evidence is typically scattered across multiple Word documents, intervention spreadsheets, and emails. Pulling it together for the pack is where most of the three-to-four-hours-per-pupil goes.

4. Intervention history with costs

The LA wants to see what the school has already tried, what it cost, and what the outcome was. After the £1.6bn Inclusive Mainstream Fund, this section is being scrutinised more carefully than ever.

Each intervention should record:

• What it was (specific, not "small group support")
• Who delivered it
• How often and for how long
• What it cost
• What outcome it was targeting
• What the outcome RAG status was at review

A common mistake is to list interventions without costs. The LA reads this as "the school hasn't really thought about whether this is the best use of resources." Cost data — even rough estimates — adds weight to the application.

The £6,000 threshold from the SEND Code of Practice is still in play: schools are expected to fund the first £6,000 of additional support per pupil from their own budget, with the LA's funding kicking in above that. The intervention log makes this visible.

5. Professional involvement

The LA wants to see who's already involved. This typically includes:

• Educational Psychologist (EP) — assessments, observations, recommendations
• Speech and Language Therapist (SALT) — assessments, therapy sessions, advice
• Occupational Therapist (OT) — sensory profiles, therapy plans
• CAMHS — referral status, sessions, diagnoses
• School nurse — medical input, care plans
• Specialist teachers (autism, deaf, visually impaired, hearing impaired)
• Local authority teams (advisory teachers, behaviour support)

For each professional, record the dates of involvement, the recommendations made, and how the school has acted on them. The LA will check whether recommendations were implemented.

This section is hard to keep up to date because the data lives in email threads, attached PDFs, and meeting minutes. Schools that maintain a structured professional involvement log find this section writes itself; schools that don't, find it takes an hour per pupil.

6. Medical needs

For pupils with medical conditions, the LA wants:

• Diagnosed conditions
• Medication and dosage if school holds it
• Care plan summary (the full care plan can be referenced rather than reproduced)
• Allergies and emergency procedures
• EpiPen / inhaler / insulin status
• Any healthcare professionals involved
• Any safeguarding considerations linked to medical need

Some of this comes from your MIS (most have a medical conditions module). Some comes from the school nurse or a separate care plan document. Pull it together but reference the full care plan rather than reproducing it.

7. Safeguarding (stub only)

The pack should indicate whether a safeguarding record exists, but not include the contents. The DSL holds the full record; the LA will request access through the proper channel if relevant.

A typical entry: Safeguarding records are held by the school's Designated Safeguarding Lead. The DSL can be contacted directly via the school office for any specific information required for the EHCP assessment.

This is a contextual redaction — the LA knows it exists and how to ask for it. The pack itself doesn't carry confidential details that don't need to be there.

8. Attainment and progress

The LA wants academic context. Include:

• Current attainment (key stage tests, GCSE projections, reading age)
• Working at vs working towards age-related expectations
• Progress over the period (improving, plateaued, declining)
• Targets and progress against them

The trend matters as much as the absolute level. A pupil working two years below age-related expectations who's making expected progress tells a different story from a pupil working at age-related expectations whose progress has stalled.

This data lives in your MIS or assessment system. Most produce some form of pupil progress report; the work is reformatting it into something the LA can read alongside the rest of the pack.

9. EHCP outcomes

If the pupil already has draft outcomes (or proposed outcomes), they go here. This is where the school proposes what an EHCP should be aiming for, with rationale.

Strong proposed outcomes:

• Are specific and measurable
• Map to the four areas of need
• Have a realistic time horizon (typically the EHCP is reviewed annually)
• Are framed around the pupil's life — not just academic achievement

For older pupils, proposed outcomes should consider preparation for adulthood: employment, independent living, community participation, health.

10. Pupil and parent voice

The pack closes with the voices the LA most wants to hear: the pupil's and the parent's.

For pupil voice, structured prompts work best:

• What do you enjoy at school?
• What do you find hard?
• What helps?
• What would make school better for you?
• What do you want to do when you finish school?

For parent voice, similar prompts but framed for adult perspective:

• What's working at home?
• What concerns you?
• What does your child say about school?
• What support has been most helpful?
• What outcomes do you most want for your child?

Both should be quoted directly where possible. The LA reads voice sections more carefully than people realise — they cut through the formal language of the rest of the pack.

How to format the pack itself

Some LAs accept Word; most prefer PDF. Either way, format for readability:

• Clear section headings
• Page numbers
• A contents page if the pack is over 15 pages
• Source attribution under each table or extract
• Date of generation on the front page
• The school's name, the pupil's name, the SENDCo's name and contact details

Most LAs have a preferred template. Use it if there is one. If not, build your own and use it consistently across pupils.

How long should the pack be?

For a primary-age pupil with relatively focused need, 12–15 pages is typical. For a secondary pupil with complex need and a long history, 25–30 pages is normal. Beyond 40 pages and the LA officer is skimming.

Be substantive but not padded. Every page should earn its place. If a section is brief because there's not much to say, that's fine — list what there is and move on.

What this looks like with the right tools

If your school keeps structured ISPs with APDR cycles, intervention logs with costs, professional involvement records, and a parent portal for voice contributions, the evidence pack writes itself. The data is already there; the pack is a structured export of it.

If your school doesn't have those things, the pack writes you. Three to four hours per pupil, every term, for every pupil who's at EHCP stage or close to it. Multiply by 30 pupils and you've got the workload picture.

SENDCo View generates the structured pack in one click, pulling live MIS data alongside the SEND-specific data the SENDCo has logged through normal daily work. We've published a sample pack that shows what a complete one looks like.

Frequently asked questions

Does every LA require all ten sections? No, but most expect all ten or close to it. A pack with fewer sections doesn't necessarily fail, but it leaves the LA officer doing more work. Schools that submit a complete pack get faster decisions.

What about pupils on SEN Support who aren't progressing to EHCP? The same data structure applies. The "evidence pack" becomes the SEN Support evidence pack — useful for showing the graduated approach, for parents who want to see the picture, and as a starting point if the pupil's needs escalate.

How often should the pack be regenerated? At every formal review of the EHCP (annually, statutorily). For pupils approaching application, the pack should be regenerated when the application is submitted. For pupils mid-cycle, the underlying data should be kept current; regenerating the pack is then trivial.

What if the LA asks for additional information? They will. EHCP processes always have follow-up questions. The pack is a starting point, not the final word.

Do we have to share the pack with parents? Yes. Parents are co-producers in the EHCP process and have a right to see what's being submitted on their child's behalf. The pack should be shared before submission, with time for parents to read and comment.

Download the checklist

We've turned this article into a printable checklist you can use to audit your own EHCP packs against. Free, no email gate.

If you'd rather skip the manual work entirely, SENDCo View generates the pack in one click.