I once spent four months turning a 90-page research report into a plain-language summary, then watched a participant struggle through it anyway because I'd used the word "methodology" six times in the first paragraph. That was the day I stopped thinking of accessible dissemination as a checkbox. It's a craft, and most of us are bad at it.
Here's what I've learned over roughly five years of working on accessible research dissemination formats for disabled audiences: the field has decent guidance for sensory and mobility needs, and almost nothing useful for cognitive access, deafblind readers, or the messy logistics of actually producing alternative formats on a research timeline.
Key Takeaways
- Standard accessible formats (tagged PDFs, captions, transcripts) cover sensory needs but leave cognitive and deafblind audiences underserved
- Easy-read and plain-language versions aren't "dumbed down" research—they're a translation task with its own skill set
- DAISY, braille, and tactile formats exist but require real budget and lead time, usually 6–12 weeks
- Evidence on which formats actually improve comprehension is thinner than the guidance suggests
- Co-design with disabled readers catches problems that accessibility checklists miss entirely
Accessible research dissemination formats: the real gap isn't the checklist
Ask ten researchers about accessible dissemination and you'll get ten variations on the same list: alt text, captions, transcripts, tagged PDFs, high-contrast slides. All good. All necessary. And all aimed at roughly the same slice of disability—blindness, low vision, deafness, motor access.
Which slice is missing? Cognitive and intellectual disability, learning disabilities, acquired brain injury, dementia, and deafblindness. Scroll through any university's accessibility guidance and count how many pages mention easy-read formats versus how many mention alt text. The ratio is lopsided, and it isn't because alt text matters more. It's because alt text is easy to automate.
Why the standard formats plateau
A tagged PDF does nothing for a reader with a cognitive disability who needs shorter sentences, one idea per line, and images that carry meaning rather than decoration. A captioned video doesn't help someone who is deafblind and needs a tactile or braille transcript with precise timestamps. These aren't edge cases. According to the UK's Family Resources Survey, around 1 in 5 people report a disability, and cognitive and sensory disabilities overlap constantly.
So the honest framing is this: your accessibility checklist probably passes. Your dissemination probably still excludes people.
Formats that actually reach disabled audiences (beyond the obvious)
When I first started building dissemination plans about four years ago, I treated "accessible" as a binary. Was it tagged? Were there captions? Done. Then a colleague who is deafblind asked me for a research summary and I realized I had no idea what to send her.
That conversation reshaped how I plan formats. Here's the working set I use now.
Easy-read and plain-language versions
Easy-read is not a summary. It's a separate document that uses short sentences, present tense where possible, one concept per line, and images that anchor meaning. Picture a two-column layout: text left, supporting icon right. That's the standard format used by organizations like Mencap and the UK's easy-read guidance.
Plain-language versions sit between the full report and easy-read. Same length as a long blog post, jargon removed, acronyms spelled out, and a clear "what this means for you" section. If you only have budget for one alternative format, I'd argue this is it—broadest reach per pound spent.
Audio-first and spoken-word formats
Podcasts with transcripts are table stakes. What's less common: audio summaries under five minutes, recorded by the researcher in their own voice, with a plain-language script. I've found these outperform written summaries for participants with low literacy, fatigue conditions, or vision loss who listen on the move.
DAISY, braille, and tactile formats
DAISY (Digital Accessible Information System) is the standard for producing navigable audio-plus-text books and reports. Braille production is slower and more expensive than most researchers expect. When I priced it for a 40-page report in 2023, a contracted braille service quoted roughly £8–£12 per braille page, with a four-week turnaround. That's the reality you're budgeting for.
Sign language and deafblind-specific output
A sign-language summary of a research finding—usually 3–5 minutes, filmed with a BSL or ASL interpreter—reaches Deaf audiences who use sign as a first language. Deafblind readers often need a mix: braille, large print, or haptic communication depending on remaining vision and hearing. There is no single format.
| Format | Primary audience | Typical lead time | Cost band |
|---|---|---|---|
| Tagged PDF + alt text | Screen reader users | Days | Low |
| Plain-language version | Cognitive / general | 1–2 weeks | Medium |
| Easy-read version | Intellectual disability | 2–3 weeks | Medium |
| DAISY audio + text | Blind, dyslexic | 2–4 weeks | Medium-high |
| Braille hard copy | Braille readers | 4–8 weeks | High |
| Sign language video | Deaf signers | 2–3 weeks | Medium-high |
None of this is cheap if you do it properly. The mistake I made early on was planning formats after the research finished, then scrambling. Now I bake format decisions into the funding proposal, because retrofitting is where budgets go to die.
Do research dissemination formats for disabled audiences actually improve comprehension?
Honestly? The evidence is thinner than the guidance implies. There's solid work on plain-language comprehension in health contexts—the UK's Plain English Campaign and the Cochrane collaboration have published on it for years—but comparative studies specifically measuring whether easy-read outperforms standard research summaries for cognitively disabled readers are scarce.
I ran a small internal test in 2024 with 24 participants. Easy-read scored noticeably higher on recall of key findings than a standard summary, but people preferred the plain-language version when asked which felt more respectful. That tracks with what disability advocates have told me for years: easy-read can feel infantilizing if it isn't done with care.
What this means for your format choices
- Offer a plain-language version as the default, not an afterthought
- Produce easy-read when the audience genuinely benefits—don't assume
- Co-design with disabled readers before you draft, not after
- Track comprehension if you can, even informally
- Ask people what they want. It sounds obvious. Almost nobody does it.
Practical logistics, and the legal floor you can't skip
The legal side is clearer than the evidence side. The UN Convention on the Rights of Persons with Disabilities, Articles 30 and 31, establishes the right to access information and the obligation to make research accessible. In the UK, the Equality Act 2010 replaced the Disability Discrimination Act 1995 and carries the same expectation: reasonable adjustments on request.
Reasonable adjustments means if a participant asks for braille, you produce braille. Not next quarter. On their timeline.
Tools I actually use
For transcripts, I run recordings through automated transcription, then edit by hand—raw output is never good enough for publication. For tagged PDFs, I use whatever the source software offers, then verify with a screen reader. For plain-language drafts, I read them aloud. If a sentence sounds like a grant application, it goes.
The tooling isn't the bottleneck. Time is. I now block roughly 15% of any dissemination budget for alternative formats and assume each format needs a named owner with a real deadline. Otherwise it slips.
What I got wrong, repeatedly
Two years ago I delivered a "fully accessible" report to a disability advisory panel. Every PDF was tagged. Every video captioned. Three of five panel members told me the executive summary was unreadable—too dense, too abstract, no clear takeaways. They weren't wrong. Accessibility is not the same as usability, and it took that feedback session to make the distinction stick.
Since then, I've added a step: before publishing anything, one disabled reader from the target audience reviews the content, not just the format. That single change has caught more problems than any automated checker.
What good looks like when you get it right
Accessible research dissemination for disabled audiences isn't a format problem. It's a translation problem—several translations, running in parallel, each aimed at a different reader. Tagged PDFs, plain language, easy-read, audio, braille, sign language. Different tools for different access needs, planned from the start, reviewed by the people you're trying to reach.
The researchers I admire in this space don't have bigger budgets. They have earlier conversations. They ask a disabled reader to sit with the draft before it's final. They accept that "accessible" is a moving target, not a certificate.
Which leaves an uncomfortable question: if you can't name a single disabled reader who saw your last dissemination output before it went live, what exactly were you making it accessible for?
