Non-medical help works – but cuts are still coming

Just over two years ago, the Department for Education (DfE) opened a call for evidence on non-medical help (NMH).

NMH is the human support component of the Disabled Students’ Allowance (DSA) that funds specialist mentors for mental health and autism, one-to-one study skills support for autism and specific learning difficulties, and specialist support professionals for deaf, blind and multi-sensory impaired students.

The consultation said there was a “fundamental question” as to whether an individual student should have a funding entitlement (ie a state responsibility) for specialist nonmedical help – or whether it should be the responsibility of a higher education provider (ie other students’ tuition fees) to provide that support to students, “assisted by funding from DfE.”

You don’t need your spidey senses to be tingling to know where that was going.

The call closed in June 2024, a response (of sorts) came in January 2025, and somewhere along the way IFF Research was commissioned to do the qualitative work that the call for evidence had not.

That work has now appeared. Non-medical help through DSA: students’ experiences and perceived quality is a 109-page mixed-methods study running 18 scoping interviews, a 2,879-response screening survey across the 2022/23, 2023/24 and 2024/25 application cohorts, 200 in-depth qualitative interviews, and a five-day online ethnography with 12 students in November 2025.

It covers seven NMH role types and includes a boost of 50 specialist mentor for mental health (SM MH) interviews to explore the relationship between NMH and provider-level mental health provision.

The headline reading is positive. Most students rate the support they get as high quality, most get the mode they prefer, and most say it has helped them engage with their studies. The report’s own conclusions describe NMH as broadly successful with operational improvements needed in application, expectation-setting, ADHD provision, and mode flexibility.

That framing, however, is doing some work – and the report is actually more interesting than it lets on. It contains a regulatory mechanism that students do not know exists, a longitudinal decline in mental health support quality that goes uninterrogated, an ADHD gap that the design has not solved, and a significant admission that NMH is doing retention work that providers don’t currently pay for. It also is published into a wider DSA trajectory that is heading the other way.

What counts as quality

Across the seven role types in scope, between 51 and 64 per cent of students rated their support as high quality. SM MH topped the table at 64 per cent, followed by specialist mentor for autism spectrum conditions (SM ASC) at 61 per cent, specialist study skills for specific learning difficulties (SS SpLD) at 58 per cent, and specialist study skills for autism (SS ASC) at 57 per cent.

The two specialist support professional roles for sensory impairments – vision (SSP VI) at 53 per cent and deafness (SSP D) at 51 per cent – sit at the bottom, with around one in six rating their support low quality. These are not catastrophic numbers, but for the categories serving students with the most concrete access requirements, they are not exactly reassuring either.

What students described as “high quality” was consistent across role types – support tailored to the specific disability, flexibility in content and scheduling, continuity of support worker, proactive communication, and clear session aims.

Low quality was the photonegative – generic checklist sessions, frequent changes of support worker, inflexible scheduling, and support workers who appeared poorly trained in the student’s specific needs. None of this will surprise anyone who has read practitioner literature on disability support.

What is more interesting is what students valued that is not officially within scope. Subject-specific knowledge (explicitly outside the NMH remit), institutional knowledge of higher education provider (HEP) processes (effectively unavailable to external providers, who deliver most NMH), proactive contact with academic staff on the student’s behalf, and a wellbeing focus that the report describes as “outside the scope” of NMH but central to why students rated support highly.

Mode preference

Sixty-eight per cent of students access NMH online, 14 per cent face-to-face, 13 per cent a mix. Eighty-two per cent of those with a preference get the mode they want. The report frames this positively. It should not.

For the 18 per cent who do not get their preferred mode, the quality drop is severe. Students with SM MH support who got their preferred mode rated it high quality at 73 per cent, those who did not, 33 per cent. The same pattern holds for SM ASC (71 vs 33), SS ASC (67 vs 27), and SS SpLD (69 vs 30). Mode mismatch is a quality predictor in its own right.

The barrier, by far the most commonly reported, is geographic – NMH providers told students they could not offer face-to-face support because they had no support workers in the relevant region, or all face-to-face workers were at capacity. One student was told their provider would not pay for support workers’ travel expenses. Another was told face-to-face would mean sessions every two months rather than weekly. Several were not offered a choice of mode at all.

And there is a regulatory hole. Footnote 2, on page 9:

DfE policy is that an NMH provider who cannot supply the support required by the student within 14 days should refer the student back to Student Finance England (SFE) to be reallocated to another NMH provider.

The students in this research routinely accepted online support after being told face-to-face was unavailable in their region, with no apparent awareness that they had a regulatory right to be reallocated to a different provider who could supply it.

In other words, the mechanism designed to protect mode preference is functionally invisible to the people it is designed to protect. That is a regulatory failure, not an operational issue, and the report flags it as the latter.

The application as gatekeeper

The application process for DSA is, on the report’s own evidence, a structural drag on access. Forty-six per cent of students found it easy, 32 per cent middling, 21 per cent difficult.

A minority of students did not begin support until the second semester, with the report noting that this hits first-years hardest – they have no other institutional support yet in place, and often do not even know what NMH is when they start the academic year (47 per cent had no prior expectations of what the support would involve).

The report’s acknowledged limitation, buried near the end of the methodology, matters more than it lets on – the research focuses on students who did receive support in 2024/25, which “may limit insights into the experiences of those who were not able to access support due to barriers such as issues with applications”.

In other words, the people for whom the application gate did not open are absent from the data. The conclusion that NMH is broadly working is one tainted by survivor-bias. We have no idea how large the locked-out population is or what their characteristics are, because no one has asked.

ADHD and the role that does not exist

There is no NMH role tailored to ADHD. Students with ADHD are allocated SM MH, SM ASC, or SS ASC depending on what the needs assessor recommends. Across all three of those role types, the report finds students with ADHD describing support that does not fit – SM MH workers focused on anxiety and depression management when the student wanted help with deadlines and assignment planning, SM ASC workers offering autism-framed strategies when the student needed something different, SS ASC workers delivering generic study skills.

The report identifies the gap repeatedly, in chapter after chapter. But the nearest it gets to a recommendation is to note “the challenge of tailoring support without a specific NMH role designed for ADHD”. It’s a big big concrete design failure that the research surfaces, and the response is to merely describe it.

Meanwhile SM MH is the largest NMH category and the most politically sensitive. It is also, on the report’s own data, the only role type where perceived quality varies significantly by application year. Seventy-two per cent of students who applied in 2022/23 rated their SM MH support as high quality. For 2023/24 applicants, that fell to 63 per cent. For 2024/25 applicants, 58 per cent.

A 14-percentage-point fall over three application years is flagged once, in chapter 5, as a fact about quality variation. It does not interrogate it, propose explanations, or recommend further work. Something is happening in the SM MH market – capacity stretching as demand rises, supplier consolidation, training pipeline issues, the post-pandemic mental health caseload, the shift to online by default – and the report’s response is to note the data and move on. For the largest NMH category, that absence of analysis is hard to defend.

Retention

An especially striking paragraph in the report is in chapter 10, the conclusions, on page 78.

Overall, NMH support often fills gaps where HEP support services are inconsistent. This suggests NMH is functioning as a de-facto cornerstone of study skills support for disabled students.

A few lines earlier:

The findings indicate that to an extent NMH plays a retention support function for HEPs, even though not designed as such.

This is an admission about cost-shifting that the report makes in passing. The argument is that NMH – funded by DfE through DSA, delivered overwhelmingly by external providers – is doing work that DfE argues providers ought to be planning around and paying for.

Where HEP disability and wellbeing services are inconsistent, NMH is the consistency. Where HEP mental health provision caps out at six to eight counselling sessions, SM MH runs weekly for the year. The mental health support package the typical disabled student receives is composed of an HEP component that is patchy and a DSA component that is the actual ongoing infrastructure.

It’s an argument. But we are where we are. It means provider-level disability and wellbeing service quality has been allowed to drift downward (or at least to vary wildly) because DSA is plugging the gap. It means broader student support cannot be assessed coherently without DSA in the frame. And it means any reduction to DSA – which is precisely what is happening elsewhere – removes load-bearing infrastructure from a system that is no longer designed to function without it.

The procurement question

It is most common for NMH to be delivered by external support workers employed by third-party providers – HEP-based provision is the minority. The report notes throughout that HEP-based support workers can do things external workers cannot – work institutional systems, advocate to academic staff, follow up on reasonable adjustment processes, signpost specifically rather than generically. The case studies of “positive experiences” frequently feature this kind of institutional embedding.

External provision is unable to deliver this joining-up function. Support workers can’t reach into HEP processes they do not know, and the procurement architecture has not paid for them to learn. The cheaper market has been allowed to dominate, and the report’s own evidence is that what cheaper purchases is less effective at the integration that students consistently identify as central to good support.

The same students who are not getting face-to-face support because their provider has no coverage in their region are also disproportionately getting the support model that cannot do the institutional advocacy work. Geographic capacity gaps and the structural limits of external delivery are compounding for the same students.

Brace brace

DfE now has all the inputs it needs to make decisions about NMH, and it certainly feels like what I would regard as cuts are coming.

But the big question is – if NMH is the de-facto cornerstone of disabled student support that the report describes, what is DfE’s plan for protecting it as the rest of the DSA package is reduced around it? On the current evidence, there isn’t one.

There is a research report describing how essential the support is, and a separate trajectory of decisions reducing the support, along with an assumption that the responsibility for funding can be dumped on universities without considering whether they can afford to. Miserable stuff.