About
Who made this, and how.
Where this came from, what it's built on, how accessible it is, and where every figure on the site comes from. If you're considering using or recommending this resource, this page is for you.
About
Built in-house. By a charity that's been doing this work for 49 years.
Let's Talk About Charlie is a service of the RCA Trust, Paisley. We have given counselling and support across alcohol, drug and gambling addiction for 49 years.
We built this because the same gap kept showing up. Families, young people and frontline staff were all asking us the same questions about cocaine. There's good clinical guidance out there for workers. The NHS Greater Glasgow & Clyde cocaine toolkit is part of it. There's far less written for the person on the other side of the desk. This is meant to sit alongside that work, not replace it.
49
years delivering frontline addictions support
3
addictions covered — alcohol, drugs, gambling
Open
referral — anyone can refer themselves or someone else. Waiting list may apply.
PA1 1HP
8 Incle Street, Paisley — phone & in-person support
Accessibility
Built so as many people as possible can use it.
We've built this so as many people as possible can use it. That includes anyone using a screen reader, a keyboard on its own, or bigger text. That was a requirement from the start, not something added at the end. For anyone who needs the technical detail: the site is built against the WCAG 2.2 AA standard. It has not yet been independently audited, and we say so plainly below rather than implying otherwise.
Reading age
Written to be understood, not admired.
across the site
Health information is no use if people can't read it. The NHS recommends patient-facing material is written for a reading age of 9 to 11. When the NHS checked its own website, the average came out at almost 16. Around eight in ten UK adults don't read at that level.
This site averages a reading age of 10. That sits inside the NHS band. The pages that matter most in an emergency are plainer still, and that was deliberate.
Measured with the Flesch–Kincaid grade formula across all body text, converted to UK reading age. Headings, phone numbers and navigation are left out: they are not sentences, and counting them flatters the score. Automated scores only count sentence and word length — they can't tell you whether writing actually sounds like a person. So every section has also been read by workers, and by people who have used cocaine themselves. Their changes are in the words you're reading.
Short sentences do the heavy lifting.
Reading age is only half of it. A page can score well and still be a wall of text. Someone reading this is often worried, exhausted, coming down, or all three — so sentence length matters as much as word choice.
The rules we hold ourselves to
- Plain word first, medical word second. Not "agranulocytosis — your body stops making white cells", but "it stops your body making the cells that fight infection. Doctors call it agranulocytosis." The meaning arrives before the terminology.
- Full stops instead of commas. Where a sentence ran long we broke it up rather than cutting it. Nothing was removed to hit a number.
- Long sentences allowed when they're lists. “A particular pub, one or two names in the phone, Friday from about 6pm” is long on purpose. Being that exact is what makes someone recognise themselves.
- No euphemisms. If it kills people we say so. Softening the language doesn't soften the risk.
- Nothing implied that isn't stated. Where we don't know, or the data isn't out yet, the page says so.
Nothing is hidden, but not everything is shouted
The home page holds 11,100 words. About 8,600 are visible when you land on it. The rest sits behind panels you open if it applies to you — the injecting guidance, the sex section, the individual stories.
That's deliberate. Someone who snorts doesn't need to read the crack pipe guidance to reach what they came for. Collapsing content breaks your browser's find-on-page. So our own search covers everything, closed panels included, and opens the right one when you click a result.
What's in place today
- You can use the whole site with a keyboard, and always see where you are
- Built with the structure screen readers rely on — proper headings, landmarks and lists
- Text stands out clearly against its background, everywhere on the site
- Medical words are explained in plain English
- Every button and link is labelled for screen readers
- Quick Exit on every page — clears your history and leaves in one tap
- Laid out to reflow on small screens and when text is enlarged
- Every form field has a clear, permanent label
Where we've got to
The site has been checked against the WCAG 2.2 AA standard using automated testing tools and a line-by-line review of the code. We found and fixed problems with keyboard focus, colour contrast, form labels, heading order, and how icons are named for screen readers. Automated testing now returns no errors. Every text and colour pairing on the site meets the AA contrast standard. That includes the ones you only see once a panel is open.
Two things we did that go beyond the standard. Long sections fold into panels. You open the ones that apply to you. Nobody has to scroll past injecting guidance to reach what they came for. Collapsing content breaks your browser's find-on-page. So we built a search that covers the hidden panels too, and opens the right one when you pick a result. The second exists because of the first. It would have been easy to ship the convenience and leave the cost with the reader.
What that doesn't cover. Automated tools catch a good deal but not everything. We have not yet tested the site with real screen reader software, or with people who use one day to day. Until that's done we're not claiming full WCAG 2.2 AA conformance. Only that we built to the standard, and fixed everything we could find. An independent audit is something we intend to arrange.
Found a problem?
If something on this site is hard to use, tell us. Email info@rcatrust.org.uk with "Accessibility" in the subject line. We'll reply within five working days and fix it as soon as we can.
Method
How the check-ins are scored.
There are two check-ins on the home page. One is for families and partners. One is for people using cocaine. They are the part of this site people are most likely to ask us about, so here is exactly how they work.
The maths
Eight questions. Each has four answers. They score 0, 1, 2 or 3, lowest first. So the total runs from 0 to 24. We cut that at 6, 12 and 18, which gives four bands. Every question counts the same.
| Score | Band | Family check-in | Use check-in |
|---|---|---|---|
| 0–6 | 1 | Low concern | Low risk |
| 7–12 | 2 | Some concern | Emerging pattern |
| 13–18 | 3 | Significant concern | Established pattern |
| 19–24 | 4 | Act now | Get support now |
We show people their band, not their score. Eight questions can't be as exact as a number like “19 out of 24” makes them look. A score like that gets screenshotted and passed round as if it meant something.
Two answers that count on their own
Two answers bring up their own warning, whatever the total. They are about dying this weekend, not about how set the pattern is.
- Using alone, answered “sometimes” or “often”. There are opioids in the cocaine supply now. Alone means nobody to call 999.
- Mixing with alcohol, answered “usually” or “always”. The two together make cocaethylene, which is harder on the heart and builds up.
In the family check-in, saying “yes” to having found something sends the person to our guide on what the different forms look like.
What it is not
This is not a test and it is not a diagnosis. It has not been checked against a clinical tool. We make no claim about how often it is right, and you should not read one in.
It cannot tell you it is cocaine. The family questions are about changes in behaviour, and those changes fit depression, another drug, or a hard year just as well.
We wrote the questions here, from frontline work. The ones for people using cover the same ground clinical tools cover: how often, tolerance, how much of your week it takes, what it costs you, how hard it is to stop. But it is not built from any of those tools and should not be quoted as if it were. A worker will use a proper assessment.
What it is for is narrower. It gives someone a clear look at what they have already noticed, and it shortens the walk to the phone.
What it gets wrong
- There is no “I don't know” answer. Not knowing scores zero, which drags the total down. So a family who hasn't been able to watch closely will land lower than they should. The result page tells people to treat their band as a floor.
- In the family check-in, how long you have worried counts the same as everything else. Those are two different things.
- Because every question counts the same, two very different sets of answers can land in the same band.
- The band lines sit at a quarter, a half and three quarters of the range. They are arithmetic. They have not been set against any real population.
The first thing we would change is adding “I don't know”. We would want to hear from workers before we did it.
Data
We collect none. The scoring runs in your browser. Nothing is sent, logged or saved. The site sets no cookies. We count page visits, but that counter cannot see your answers — it only knows a page was opened. Your answers sit in the open tab and go when you close it. There is no record for us to hold, hand over, or be made to produce.
If you work in this field and you can see something wrong here, we would rather know. Email info@rcatrust.org.uk and put “Self-assessment method” in the subject line. Nobody outside RCA Trust has reviewed these two check-ins yet.
Sources
Where the figures come from.
Every statistic on this site is listed here with its source and date, so you can check it or cite it. Some figures come from an earlier report than the latest one. Where that happens we say so, rather than letting it look newer than it is.
Deaths and treatment
- National Records of Scotland — Drug-related deaths in Scotland, 2024 (published September 2025). Source for the 479 figure, the 47% share, and the 13% overall fall. The 2025 annual figures are expected around September 2026 and this site will be updated when they are published.
- Public Health Scotland — RADAR quarterly report, April 2026. Source for powder cocaine at 33% of specialist treatment assessments, ahead of heroin at 22%. Position unchanged in the July 2026 report, which does not restate the percentage.
- Public Health Scotland — RADAR quarterly report, July 2026 (covering March–May 2026). Source for the 4% quarterly fall in suspected drug deaths, the rise in emergency department attendances, nitazene detection, and the WEDINOS testing figure. The Scottish Government's separate "Suspected drug deaths in Scotland" report was discontinued after its final release in March 2026. Equivalent data now sits in the RADAR quarterly reports.
What cocaine is cut with
- Levamisole prevalence — UK and international seizure analysis; figures in the 60–90% range are widely reported across the UK, US, Spain, Netherlands and Switzerland.
- European Medicines Agency, February 2026 — PRAC recommendation to withdraw levamisole-containing human medicines, on the basis of leukoencephalopathy risk.
- Fentanyl in powder cocaine — US community drug-checking data, adjusted prevalence approximately 1 in 7 samples. This is US data and is labelled as such on the site; UK prevalence is lower but rising.
- Medetomidine and xylazine in the Scottish supply — Public Health Scotland RADAR, July 2026.
- Pink cocaine composition — UK and international drug-checking data, 2024–2025, including WEDINOS submissions.
Reading age
- Flesch–Kincaid grade formula, applied to all body text on the site and converted to UK reading age. Measured 4 August 2026.
- NHS guidance on reading age for patient-facing material (9–11), and the published audit finding an average reading age of almost 16 across NHS website content.
Clinical content
- Cardiovascular, neurological, respiratory and gastrointestinal effects are consistent with standard clinical toxicology and with NHS Greater Glasgow & Clyde's Alcohol and Drug Recovery Services Cocaine Toolkit (v2, reviewed September 2025), which we'd recommend to any clinician wanting the full picture.
- Cocaine withdrawal — DSM-5-TR stimulant withdrawal criteria, and Gawin & Kleber's three-phase model (1986), which remains the basis for how the crash, withdrawal and extinction phases are described.
- Naloxone guidance — Scottish Government national naloxone programme; supply routes via HaRRT, Renfrewshire ADP and Scottish Families Affected by Alcohol & Drugs.
- Resuscitation Council UK — CPR ratios and recovery position guidance.
A note on what this is and isn't. This site is written for the public, not as clinical guidance. Every section has been checked line by line for accuracy. Workers and people who have used cocaine have reviewed it too. It carries a named author, above. What it has not been through is formal NHS clinical governance, and its author is not a clinician. If you need a governed clinical document for your service, use your board's own guidance — NHS GGC's toolkit is a good example.
If you spot something you believe is wrong, please tell us: info@rcatrust.org.uk. We would rather be corrected than be confidently wrong.