select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to study investigating the association between different types of anticholinergic drugs and risk of dementia

A new study, published in the BMJ, estimates the association between duration and level of exposure to different classes of anticholinergic drugs and subsequent incident dementia.

A briefing accompanied this roundup.


Prof Peter Passmore, Professor of Ageing and Geriatric Medicine, Queen’s University Belfast (QUB), said:

“The cholinergic system is known to be critical for cognitive function and some treatments licensed for Alzheimer’s dementia increase cholinergic activity. There have been concerns about medicines with anticholinergic activity as in theory they can reduce cognitive function and several studies have shown that medicines with anticholinergic activity adversely affect cognition and can be associated with an increased risk of dementia.

“This is a cross sectional study using the UK GP database to estimate the association between the duration and level of exposure to different classes of anticholinergic drugs and subsequent incident dementia.  This study of a large number of patients to study anticholinergic drug classes in depth has used a specific score for anticholinergic burden (ACB).  The methods seem appropriate and suggest that certain medications with anticholinergic effects are associated with risk of dementia. Specifically mentioned are exposure to antidepressant, urological, and antiparkinson drugs rated as having higher anticholinergic activity (ACB score of 3).

“The specific antidepressants are amitriptyline, dosulepin, and paroxetine, urologicals predominantly oxybutynin and tolterodine and anticholinergic drugs used in Parkinson’s disease. Amitriptyline and dosulepin are very old antidepressants which have declined in usage. Paroxetine is a more modern agent which would be used more frequently. Amitriptyline is used more widely in older people in small doses and often as an unlicensed medication which is not recommended. Oxybutinin is not recommended by NICE for use in frail older people but can nevertheless be widely used in younger adults and tolterodine is also commonly prescribed. The use of anticholinergic medication in Parkinson’s disease has really fallen away due to guidelines.

“The study findings are really about the need to study specific drug classes rather than total anticholinergic burden which is a reasonable conclusion. In terms of clinical practice the message would remain about caution with medications with anticholinergic effects, particularly if several are used. Cumulative exposure remains an issue so the total length of time that people take these medications is important. The use of older antidepressants such as amitriptyline and dosulepin should be discouraged and more modern alternatives prescribed. The adverse effects of oxybutynin on cognition are well known and this study adds to those concerns. There are less data on tolterodine but the message would be to monitor long term usage of medications in this class and consider alternative medication.”


Prof Les Iversen, retired Professor of Pharmacology and former Chair of the Advisory Council on the Misuse of Drugs (ACMD), said:

“My first impression from a read of the paper is favourable. The authors have conducted a careful analysis of more than 50,000 cases, with rigorous statistical analysis.  However the effect of anticholinergic therapy is relatively small (odds ratio 1.1 – 1.2) and establishing an association does not prove a causal link. Nevertheless, the paper may act as a useful guide for future research and clinical practice.”


Dr Parastou Donyai, Associate Professor of Social and Cognitive Pharmacy, University of Reading, said:

“This paper ties in well with current advice for doctors to be careful when using drugs with anticholinergic effects in older people, those who are frail and others with numerous health conditions. The warning to not use anticholinergic drugs in Parkinson’s disease is also important. But the headline claim that some drugs, especially antidepressants, can cause dementia up to 20 years later should be looked at closely. This type of study imagines that patients actually take their drugs as they were prescribed for them. But we know from other research that people with long-term health conditions really only take their medication as prescribed around half of the time – the other half, people either take more or less of their medication or not at all. This issue of ‘non-adherence’ is common in people with heart conditions, stomach problems and mental health illnesses including depression and we want to be careful not to add to the problem. So my first comment is to remember that the paper really looked at the ‘prescribing’ of drugs rather than the ‘taking’ of these medications. This brings me to my second comment which is that anyone worried about taking their antidepressant as a result of this paper should speak to their doctor or pharmacist first and not suddenly stop their medication.”


MHRA Spokesperson said:

“Patient safety is our highest priority and we continuously monitor the safety of all medicines on the UK market.  These findings will be carefully evaluated, as with any study, to determine whether they have any implications for the safe use of anticholinergic medicines.”


Prof Martin Rossor, NIHR National Director for Dementia Research, University College Hospitals, & Professor of Clinical Neurology, UCL, said:

“Anticholinergic drugs interfere with the brain transmitter acetylcholine that is involved in memory function and is reduced in Alzheimer’s disease, the commonest cause of dementia

“It has been known for a long time that use of an anticholinergic drug can impair memory and attention whilst the drug is being taken. This study is important in that it demonstrates that those individuals who are taking drugs with anticholinergic actions are at a higher risk of developing a dementia many years later (from 4 to 20 years later)

“It is important to be cautious about associations as they do not prove causation. The authors discuss the difficulties of interpretation and one important note of caution is that subtle changes occur many years before a diagnosis of a dementia. For example depression is common before the onset of dementia and so those individuals receiving an antidepressant drug with anticholinergic effects may already have a disease such as Alzheimer’s disease that will lead to dementia ten years later.

“It is also important to recognise that at an individual level the associated increase risk is very small. In the future it will be important to identify whether the risk relates only to those drugs that cross the blood brain barrier. Recent drugs used for bladder symptoms that have anticholinergic activity have much less effect on the brain than older drugs.”


Prof Stephen Evans, Professor of Pharmacoepidemiology, London School of Hygiene and Tropical Medicine (LSHTM), said:

“This is a carefully conducted and analysed observational study of the association between prescription of various drugs and a note of dementia in a GP’s record. The authors are very careful to note some of the limitations of the study, but some of their wording suggests that they are convinced of a causal effect. While this is possible, other interpretations are also consistent with the data. Even if there is a causal effect, the strength of the effect is rather small.

“A number of the associations do not show a pattern of increasing risk that would be expected if the association were causal. Some associations appear with only a few prescriptions for a drug and do not show a marked increase with increasing numbers of prescriptions. Virtually all of the associations are diminished when adjustment is made for factors that differ between users and non-users of a drug that are not due to the drug itself. Some factors, as the authors note, will be unmeasured, and adjustment for these seems likely to reduce the strength of the associations even further.

“The “diagnosis” of dementia is a note in a GP record rather than a formal diagnosis made by a neurologist. The data in the GP records are not always well recorded, and it is possible that if they were well recorded, the associations could be stronger. Issuing of prescriptions does not mean that patients, especially elderly ones, actually take them, which is another potential source of error.

“The theory of “Anti-cholinergic Burden” (ACB) scale seems to be shown by this study to be not very robust and varying between drug classes. The slightly larger associations with antidepressant and antiparkinson drugs may be explained by being given for conditions possibly associated with dementia, but urological drugs at first sight are less likely to have that explanation. However it also seems possible that even those with urological problems may also have neurological causes.

“All patients should not be given unnecessary drugs, but this paper should not be used to stop drugs that are having benefits in individual patients for the conditions for which the drugs are prescribed.”


Prof Rob Howard, Professor of Old Age Psychiatry, UCL, said:

“This is a large and very well-conducted epidemiological study showing an 11% increase in the incidence of dementia in people who have been treated with drugs that include as a side-effect, strong action against the neurotransmitter acetylcholine, in the preceding 4 to 20 years. Acetylcholine is the brain transmitter that is most severely deficient in the brains of people Alzheimer’s disease and current treatments for dementia work by boosting levels of this. We already know that anticholinergic drugs can precipitate the presentation of people with early dementia, but this is the first convincing demonstration that the use of the drugs many years before dementia diagnosis might increase risk. The results are interesting but it is too early to conclude that use of these drugs “causes” later dementia and people should talk to their doctors before stopping or changing and if the medications implicated. It is possible that use of some of these drugs may have actually been to treat the very earliest symptoms of Alzheimer’s disease, which can be associated with low mood and lower urinary tract infections, many years before the development of dementia.”


Prof Clive Ballard, Professor of Age-Related Diseases, University of Exeter Medical School, said:

“This is a very large and well conducted study demonstrating a clear association between the use of drugs with anticholinergic side effects and the risk of dementia/cognitive decline.  Whilst this association has been previously reported in a number of studies showing impact on both cognition and disease pathology, the current work adds to our knowledge both through the size of the study and more importantly by demonstrating an association with anticholinergic exposure many years before the onset of dementia.

“The important thing is that even individual drugs which only have very modest anticholinergic effects, when taken in combination with other drugs can lead to a combined anticholinergic burden that may have a significant impact on cognition –highlighting the importance of care medication review.”


Dr Carol Routledge, Director of Research at Alzheimer’s Research UK, said:

“This well conducted study adds to existing evidence of a link between certain anticholinergic drugs and an increased risk of dementia. By taking advantage of high-quality medical information held in GP records, researchers were able to analyse data from hundreds of thousands of people and drill down into important details, such as when people started to take these drugs, the strength of their anticholinergic effect, and which conditions they were being prescribed for.

“While the results don’t paint a straightforward picture, they do reveal a link between dementia risk and the use of certain drugs that may be prescribed to treat depression, urological conditions or the symptoms of Parkinson’s disease. Not everyone with the conditions highlighted in the study will be prescribed these particular drugs, and the researchers did not find a link between dementia risk and anticholinergic drugs used to treat allergies, stomach problems or muscle spasms.

“While there are important advantages to research that uses the rich data held in medical records, we know that less than 70% of people with dementia receive a formal diagnosis, and this study may not tell the whole story. The study didn’t investigate what might cause this link between anticholinergics and dementia risk, and researchers will need to build on these findings in future studies.

“Anticholinergics can have many beneficial effects which doctors need to weigh against any potential side effects. Anybody concerned about their current medication should speak to a doctor before stopping a course of treatment.”


* ‘Anticholinergic drugs and risk of dementia: case-control study’ by Richardson et al. published in The BMJ on Wednesday 25 April.


All our previous output on this subject can be seen at this weblink:


Declared interests

Prof Peter Passmore: “Have received honoraria from Pfizer and Astellas.”

Prof Les Iversen: No conflicts of interest.

Dr Parastou Donyai: No conflicts of interest

Prof Martin Rossor: “No conflicts except I am the NIHR National Director for Dementia Research and I am also supervising some research on cognitive side effects of drugs.”

Prof Stephen Evans: “No conflicts of interest. I am an independent expert member of the EU drug safety committee.”

Prof Rob Howard: No conflicts of interest.

Prof Clive Ballard: “No specific COIs related to this work.”

Dr Carol Routledge: No relevant conflicts.

No other conflicts received.

in this section

filter RoundUps by year

search by tag