Which medications are anticholinergic? What 200,000 dementia cases say about burden

Short answer: Anticholinergic drugs block acetylcholine, a chemical the brain uses for memory and attention, and dozens of ordinary medicines have that effect as a side effect — older allergy tablets, bladder drugs, some antidepressants, some antipsychotics, some anti-nausea and anti-vertigo drugs. Adding them up gives your anticholinergic burden. The largest study ever done on this — a Swedish analysis of 199,526 people with dementia matched to as many controls — found something the popular lists miss: the risk climbed steadily with cumulative use of strong anticholinergics (up to 66% higher odds at the highest exposure), but showed no dose-response at all for the weak ones. And a Cochrane review of the trials found no proof yet that lowering your burden improves anything.

This is educational content, not medical advice, and it is not a list of drugs to stop. Several of the medicines below treat conditions that matter considerably more than this risk does. Any change belongs with the person who prescribed them.

What "anticholinergic" actually means

Acetylcholine is one of the brain's main messengers for memory, attention and learning. It also runs a lot of the plumbing: saliva, sweat, bladder, gut, heart rate, pupils.

A drug is called anticholinergic when it blocks acetylcholine. For a few medicines that is the whole point — bladder antispasmodics work precisely by damping down acetylcholine signalling to the bladder muscle. For many more, it is a side effect that comes along with the drug's real job.

You can usually feel it. Dry mouth, blurry near vision, constipation, difficulty passing urine, drowsiness and a woolly, slow-witted feeling are the classic set. In older adults the cognitive part can be pronounced enough to be mistaken for early dementia — and it lifts when the drug is stopped.

Anticholinergic burden is the cumulative effect of everything you take at once. Someone on three medicines, each mildly anticholinergic, can carry more burden than someone on a single strong one. That is why the useful question is never "is this one drug bad" but "what does my whole list add up to."

The drug groups that carry it

There is no single official list — several competing scales exist, and, as a Cochrane review noted, agreement between them "is often poor." What follows is the broad map, grouped by how much they typically carry.

Commonly rated strong — the ones the evidence concentrates on:

• Older, sedating antihistamines — diphenhydramine (Benadryl and most "PM" and night-time products), chlorphenamine, hydroxyzine, promethazine, dimenhydrinate and meclizine for travel sickness and vertigo
• Bladder / overactive-bladder drugs — oxybutynin, tolterodine, solifenacin, darifenacin
• Older tricyclic antidepressants — amitriptyline, nortriptyline, doxepin, imipramine (often prescribed at low doses for nerve pain, migraine or sleep rather than depression)
• Some antipsychotics — chlorpromazine, clozapine, olanzapine, quetiapine
• Anti-Parkinson's anticholinergics — trihexyphenidyl, benztropine
• Some antispasmodics — dicyclomine, hyoscine/scopolamine

Commonly rated weak or moderate: many cardiovascular drugs (including furosemide, digoxin, warfarin), some opioids (codeine, morphine), prednisolone, ranitidine-class acid drugs, carbamazepine, loperamide, metoclopramide, some SSRIs (notably paroxetine) and some antibiotics.

Usually not anticholinergic at all: newer non-sedating antihistamines (loratadine/Claritin, cetirizine/Zyrtec, fexofenadine/Allegra) were specifically designed to stay out of the brain. Most SSRIs other than paroxetine sit low. Statins are not anticholinergic — and are, if anything, associated with lower dementia risk, not higher.

This is a map, not your medication list. Scales disagree, doses matter, and a drug's rating tells you nothing about whether you should be on it. Use it to prepare a question, not to reach a verdict.

What the largest study found — and the distinction nobody reports

In October 2025, researchers at Karolinska Institutet published the largest analysis on this question in Alzheimer's Research & Therapy: 199,526 people aged 40 and over who developed dementia in Sweden between 2008 and 2017, each matched to a control of the same age, sex and region — nearly 400,000 people in total, using national prescription registers rather than anyone's memory.

Crucially, they separated drugs by potency using the Anticholinergic Cognitive Burden scale, and counted only prescriptions filled at least a year before dementia onset.

For strong anticholinergics, the risk rose with cumulative dose:

• 1–89 defined daily doses — adjusted odds ratio 1.10 (1.08–1.12)
• 90–364 — 1.39 (1.34–1.43)
• 365–1,094 — 1.57 (1.50–1.65)
• 1,095 or more — 1.66 (1.55–1.79)

To reach that top band you would need roughly three years' worth of daily doses of a strong anticholinergic.

For weak anticholinergics, there was no such pattern. The lowest exposure band gave an odds ratio of 1.11 (1.08–1.13) — and the highest gave 1.01 (0.98–1.03), essentially nothing.

That flat, slightly inverted shape is worth pausing on, because it teaches you how to read this kind of research. A genuine drug effect gets stronger with more drug. An association that is present at low doses and vanishes at high ones is behaving like a marker of something else — the kind of person who gets prescribed anything at all, or an early symptom prompting the prescription. The dose-response for strong drugs is what makes that arm of the finding more credible; its absence for weak drugs is what makes that arm less so.

The signal concentrated in three groups: urinary antispasmodics, antihistamines and psychotropic drugs. It was not significant for gastrointestinal drugs or muscle relaxants.

The associations were also larger for vascular dementia and Lewy body dementia than for Alzheimer's disease — which fits, since acetylcholine signalling is especially disrupted in Lewy body disease. (More on how these differ: Alzheimer's vs vascular dementia.)

One caution the authors flag themselves: associations were stronger in people with milder-stage dementia. If the drugs were driving the disease you would expect the opposite. That pattern points, again, towards detection effects and early symptoms doing some of the work.

How strong is the evidence overall?

Two other bodies of work fill in the picture, and they point the same way: a consistent but modest signal from observational data, with real uncertainty attached.

Cochrane, 2021. A systematic review of 25 studies covering 968,428 older adults without cognitive problems at the start pooled the data where it could. Adjusted odds of cognitive decline or dementia for those carrying anticholinergic burden came out at 2.63 (1.09–6.29) on the ACB scale, with a cross-scale estimate of 2.16 (1.38–3.38). Those look large — but note how wide the intervals are, and note the authors' own verdict: the overall certainty of the evidence was rated low. Their conclusion, in full: "There is low-certainty evidence that older adults without cognitive impairment who take medications with anticholinergic effects may be at increased risk of cognitive decline or dementia."

Cochrane's pooled data and the Swedish study do not fully agree: Cochrane found a gradient even at the weakest ACB level, where the much larger Swedish analysis found none. That disagreement is unresolved, so treat "weak anticholinergics are fine" as an open question rather than a settled answer.

Two large modern cohorts, 2025. Researchers externally validated an anticholinergic index of 88 drugs in the UK Biobank (125,260 people) and the US All of Us programme (92,047 people), with death handled as a competing risk. Each additional index drug at baseline was associated with higher dementia risk — HR 1.15 (1.09–1.21) in UK Biobank and HR 1.06 (1.04–1.09) in All of Us. Small per drug; additive across a list. The authors are careful: "causal inference is not possible."

Taken together: a modest, repeatedly observed association, strongest for heavy long-term use of the strong drugs, from studies that cannot separate the drug from the reason it was prescribed.

"Should I stop?" — the honest answer

This is the question everyone actually has, and it has a real evidence-based answer that almost no consumer page gives.

No trial has yet shown that reducing anticholinergic burden improves cognition.

A Cochrane review published in 2023 looked specifically for randomised trials of deprescribing — actively reducing or stopping anticholinergic medicines — and measured what happened to people's thinking. It found three trials with 299 participants between them, followed for one to three months. The results were rated very low certainty, and the authors' conclusion is unambiguous: "There is insufficient evidence to reach any conclusions on the effects of anticholinergic burden reduction interventions on cognitive outcomes… The evidence from RCTs was of very low certainty so cannot support or refute the hypothesis that actively reducing or stopping prescription of medications with anticholinergic properties can improve cognitive outcomes in older people."

That is not a reason to shrug. It is a reason to be precise about what is and is not known:

• Known: anticholinergic drugs cause immediate confusion, drowsiness, dry mouth and poor recall, and that effect lifts when the drug is stopped. That alone is a good reason to review a list.
• Observed, not proven: heavy long-term use of strong anticholinergics is associated with more dementia.
• Untested: whether cutting your burden lowers your risk. The trials have not been done.

Anyone telling you confidently that stopping undoes anything is ahead of the evidence — as is anyone telling you a few tablets caused a diagnosis.

How to actually check your own list

Something concrete, and safe:

1. Write down everything — including what you buy yourself. Prescription, over-the-counter, and anything taken "as needed." Over-the-counter sleep and allergy products are the single most common source of hidden anticholinergic burden, and they never appear on a prescription record. That is exactly why the studies above underestimate real exposure.

2. Flag the strong groups. Older sedating antihistamines, bladder drugs, tricyclic antidepressants, antipsychotics, anti-Parkinson's anticholinergics. If nothing on your list falls in those groups, the evidence in this article mostly is not about you.

3. Note duration, not just presence. The dose-response is measured in years of daily use. Three weeks of an antihistamine is a different thing from three years of a bladder drug. If you have used a lot in the past, the useful question is not what has already happened — it is whether the total is still climbing.

4. Book a structured medication review. In the US, pharmacists provide medication therapy management; in the UK this is a structured medication review with your GP practice or community pharmacist. Bring the list and ask three questions: Is each of these still doing a job? Is there a lower-burden alternative for any of them? What would we watch for if we changed one?

5. Know the easy swaps exist — but let a clinician make them. Second-generation antihistamines instead of diphenhydramine, newer bladder drugs (mirabegron works by a different mechanism entirely), a different antidepressant class. Whether a swap makes sense depends on why you are on the drug in the first place.

6. Never stop abruptly on your own. Antidepressants, antipsychotics and sedatives all need tapering. Abrupt discontinuation can cause withdrawal effects worse than the thing you were worried about.

Where this sits among the other risk factors

Anticholinergic burden is not one of the fourteen modifiable risk factors named by the Lancet Commission — see the full list here. That is worth knowing, because it tells you something about proportion. The factors with the strongest evidence behind them are blood pressure, hearing loss, education, smoking, physical inactivity, depression, social isolation, diabetes, obesity, alcohol, head injury, air pollution, vision loss and high LDL cholesterol.

Reviewing your medicines is a sensible, low-cost thing to do, and it may improve how you feel right now. It is not a substitute for the levers with much larger effects behind them.

More on this: does Benadryl cause dementia? · do sleeping pills cause dementia? · normal forgetfulness vs dementia · what is mild cognitive impairment? · can you reduce your dementia risk? · check your dementia risk

How we research and review our content →

This article is educational and is not medical advice, diagnosis, or treatment. The studies cited describe findings in general populations, not any individual. Solenna is not a medical device and does not diagnose, prevent, treat, or cure Alzheimer's disease or any other condition. Risk reduction means lowering probability, not eliminating it; individual results vary and no outcome is guaranteed. Do not start, stop or change any prescribed medication on the basis of this article — take your medication list to a pharmacist or the clinician who prescribed it.