Two numbers landed three weeks apart, and they belong on the same page. Irwin Mitchell's analysis of HM Courts and Tribunals Service data, summarised in Legal Futures on 22 April 2026, shows that UK tribunal claims involving neurodiversity discrimination rose by 95% over five years: 517 cases in 2024–25, up from 265 in 2020. The increase between 2024 and 2025 alone was 19%. The most commonly cited conditions are autism and ADHD. Average legal-defence cost per case sits at £8,500, before any award.
The second number is the explanation. VinciWorks, surveying 495 UK HR and compliance professionals over the same period, found that 35% say managers in their organisation lack confidence when discussing reasonable adjustments for neurodivergent employees. Nick Henderson-Mayo, VinciWorks' Head of Compliance, put it plainly: "Organisations who say they are neurodiversity-friendly should have the evidence to back this up."
The two numbers describe the same gap from opposite ends. One side is the manager who, asked to discuss an adjustment, freezes and falls back on policy language. The other side is the employee who, eventually, files.
Why Compliance Training Has Not Closed The Gap
UK organisations have been training on neurodiversity for years. Most large employers run an annual module. Engagement-style awareness sessions are common. The 95% rise has happened anyway. The reason is not new and not specific to neurodiversity: awareness does not transfer to behaviour.
Roediger and Karpicke (2006), in Psychological Science, established the testing effect: active retrieval of material improves long-term retention by around 50% compared with re-reading or passive review. Compliance e-learning sits firmly on the wrong side of that curve. A manager who has watched a module on reasonable adjustments has, in retention terms, done the cognitive equivalent of skimming a chapter. Three months later, faced with an actual conversation in an actual corridor, they have very little to draw on.
The second problem is the Dunning-Kruger pattern. In our own academic behaviour-change work, building on research from UCL, Cambridge and Bocconi, participants' self-rated confidence on communication skills did not predict their observed performance. People thought they could handle the conversation. They could not. We solved this in study design by replacing self-reports with behavioural measurement, role-played scenarios scored by trained observers. The same problem shows up in compliance training: managers who tick "yes, I am confident discussing adjustments" on the post-module survey are not, in practice, the managers who handle the live conversation well.
What The Conversation Actually Looks Like
The corridor version of this conversation is short. An employee says, in some form, I think I might be autistic / I have ADHD / I am being assessed. What happens in the next ninety seconds determines almost everything else.
Three observable manager behaviours predict the good outcome:
- Drop the script. A confident manager does not lead with HR policy language. They ask, as a person: thank you for telling me, what would help? The phrase is small and the effect is large. Edmondson (1999) in Administrative Science Quarterly identified this kind of opening as the single most reliable behavioural marker of a psychologically safe team.
- Hold the silence. Most managers, anxious, fill the pause with reassurance. The good move is to wait long enough for the employee to actually answer. The employee usually knows what they need; the conversation goes wrong when they are interrupted.
- Name the follow-up. "I am going to speak to HR by Friday and come back to you by Tuesday." A specific commitment, with a specific date, anchored by the manager personally rather than passed sideways. This is the behaviour most commonly missing in cases that go on to tribunal.
None of this is in the Equality Act. All of it is observable. None of it transfers from a slide deck.
What We Do About It
Our manager workshops and immersive simulations are built for exactly these corridor conversations. A small group of managers walks into a designed scenario. Professional actors play the employee, the HR partner, the senior stakeholder. The manager has to handle the disclosure live, with feedback, more than once. By the third repetition, the behaviours above are no longer awkward, they are the default. In Sidestream's own UCL-anchored research, participants who learned a communication skill through immersive role-play scored roughly 20% higher on observed behaviour than those who learned the same content through video or slide-show training. That is the gap between a compliance certificate and a manager you would actually want in the conversation.
The 95% spike is not going to reverse on its own. It is the visible end of a conversation that most organisations have not yet taught their managers how to have. Read more on how to handle difficult conversations at work, or book a call to look at what manager rehearsal for adjustments would look like in your context.
Book a free 30-minute diagnostic call → or read about our research-backed approach.