Researchers have developed a new medication-risk tool that can automatically identify older adults whose combination of medicines may put them at greater risk of delirium and serious falls.
Called the Anticholinergic Medication Index, or ACMI, the system calculates the cumulative anticholinergic effect of a patient’s medicines and highlights people who may benefit from a structured medication review.
The tool was developed and tested using health records from nearly 750,000 adults aged over 65 and has already moved beyond the research stage. A clinical version is now integrated into electronic patient-record systems at two NHS Trusts in the UK.
According to the new research published in BMJ Medicine, the aim is not to automatically stop medications but to give doctors and pharmacists a faster way to identify patients whose overall medication burden may be contributing to avoidable harm.
Table of Contents
Why Anticholinergic Medicines Can Be a Problem
Anticholinergic medicines reduce the activity of acetylcholine, a neurotransmitter involved in functions including memory, attention, muscle activity and regulation of several organs.
Some medicines are prescribed specifically because of their anticholinergic effects.
Others have anticholinergic activity as an unintended secondary property.
These drugs may be used to treat conditions involving:
- bladder control;
- gastrointestinal symptoms;
- respiratory disease;
- movement disorders;
- depression and other psychiatric conditions;
- allergies;
- sleep problems.
A single anticholinergic medicine may be entirely appropriate.
The problem can emerge when several medicines with anticholinergic effects are taken together.
This cumulative exposure is known as anticholinergic burden.
Older Adults Are Particularly Vulnerable
Anticholinergic side effects can occur at any age, but older adults are often more susceptible.
Possible effects include:
- confusion;
- delirium;
- dizziness;
- blurred vision;
- dry mouth;
- constipation;
- difficulty urinating;
- impaired balance;
- falls.
Older adults are also more likely to live with several chronic illnesses and therefore take multiple medicines simultaneously.
According to the Health Data Research UK report on ACMI, approximately 50% of older adults in the UK receive at least one medication with anticholinergic effects each year.
That does not mean those prescriptions are inappropriate. Instead, it creates a need to monitor the combined effect of several medicines over time.
What Is the Anticholinergic Medication Index?
ACMI is designed to calculate a patient’s overall anticholinergic medication burden automatically.
The current system incorporates 88 anticholinergic medicines commonly prescribed in UK primary care.
A person taking no medications on the list receives a score of zero, while the theoretical maximum score would be 100 if somebody were prescribed every medicine included in the index.
The important difference between ACMI and many older anticholinergic scales is that it was developed using real-world clinical outcomes, particularly delirium and falls.
| ACMI Detail | Information |
|---|---|
| Full name | Anticholinergic Medication Index |
| Target population | Adults aged 65+ |
| Medicines included | 88 |
| Main risks assessed | Delirium and falls |
| Development and validation population | Nearly 750,000 older adults |
| Primary-care pilot | Bradford |
| Hospital implementation | Two NHS Trusts |
| Main purpose | Prioritise medication review |
Nearly 750,000 Health Records Were Used
The scale of the dataset is one of the major strengths of the project.
Researchers developed the index using de-identified routinely collected healthcare records from Connected West Yorkshire, combining information from primary and secondary care.
The model was then tested separately using data held in the Secure Anonymised Information Linkage, or SAIL, Databank in Wales.
Testing a clinical model in a separate geographical population is important because a prediction system can perform well in the dataset in which it was created but fail elsewhere.
The independent Welsh validation therefore provides stronger evidence that ACMI may be useful beyond a single local health system.
The Tool Predicts One-Year Risk
In a pilot involving GP practices in Bradford, ACMI automatically generated lists of people aged 65 and older who may benefit from medication review.
For each patient, the system could show:
- their ACMI score;
- which medicines contributed to that score;
- their predicted one-year risk of delirium or a serious fall.
Without an automated system, a pharmacist would have to reconstruct this information patient by patient.
The tool therefore attempts to turn a complex medication-safety problem into something that can be reviewed efficiently during routine care.
ACMI Is Already Being Used in NHS Hospitals
Many healthcare prediction models never progress beyond academic publications.
ACMI is unusual because a hospital version has already been deployed.
The Anticholinergic Medication Index Clinical Decision Support System, or ACMICS, is now live within the electronic patient-record systems of two NHS Trusts.
The University of Leeds report on the project states that the system automatically recalculates a patient’s score whenever an anticholinergic medicine is prescribed or stopped.
That allows clinicians to see changes in risk during normal prescribing rather than relying on occasional manual review.
Why Integration Into Medical Records Matters
Clinical decision-support tools are more useful when they fit naturally into doctors’ and pharmacists’ existing workflow.
A separate online calculator requires healthcare professionals to:
- leave the patient record;
- manually enter medications;
- calculate the result;
- return to the prescribing system.
ACMICS instead sits inside the electronic patient record.
The developers observed actual clinical workflows when deciding where the score should appear, rather than assuming how doctors and pharmacists would use it.
That could make routine adoption considerably easier.
The Tool Does Not Automatically Tell Doctors to Stop Medication
A high ACMI score is not an instruction to discontinue treatment.
This distinction is crucial.
Many anticholinergic medicines provide substantial clinical benefits, and stopping some drugs suddenly can itself be dangerous.
Instead, a high score is designed to prompt a structured review.
Clinicians can consider:
- why the medicine was originally prescribed;
- whether it is still required;
- whether the dose remains appropriate;
- whether a less anticholinergic alternative exists;
- whether several drugs are producing cumulative effects;
- the patient’s overall health and treatment priorities.
The appropriate outcome may be to reduce, replace or stop a medicine, or to continue treatment unchanged if the benefits outweigh the risks.
Why Delirium Is a Serious Concern
Delirium is a sudden change in attention, awareness and thinking.
It can develop over hours or days and is particularly common among frail older adults during hospitalization.
Triggers can include:
- infections;
- dehydration;
- surgery;
- pain;
- unfamiliar surroundings;
- medication effects.
Acetylcholine plays an important role in attention and cognitive function, which helps explain why medicines that block cholinergic signaling can contribute to delirium in susceptible individuals.
Delirium is not the same as dementia, although people with dementia may be particularly vulnerable to it.
Falls Can Have Long-Lasting Consequences
Medication-related dizziness, blurred vision, sedation or confusion may also contribute to falls.
For an older adult, a fall can lead to much more than a short-term injury.
Possible consequences include:
- hip fractures;
- head injuries;
- hospitalization;
- reduced mobility;
- loss of independence;
- fear of falling again.
The research team is now piloting ACMI in geriatric medicine and orthopaedic hip-fracture wards to examine whether the system could be expanded more widely across hospitals.
Why Cumulative Medication Burden Can Be Missed
One challenge is that anticholinergic exposure can accumulate across medicines prescribed for completely unrelated conditions.
A patient might receive one medicine for bladder symptoms, another for depression and another for an allergy.
Each prescription considered individually may look reasonable.
But their combined anticholinergic effect may become clinically important.
Different specialists may also prescribe medications independently, making cumulative burden less obvious unless the full medication list is reviewed.
ACMI is designed specifically to expose that combined risk.
The Tool Is Intended to Support Deprescribing, Not Replace Clinical Judgment
“Deprescribing” means systematically reviewing medications and reducing or discontinuing those that are no longer necessary or whose harms outweigh their benefits.
It has become increasingly important as populations age and polypharmacy becomes more common.
ACMI could help identify the patients most likely to benefit from such a review.
But the final decision remains clinical.
Factors such as frailty, previous falls, cognition, mobility, current symptoms and patient preferences all need to be considered alongside a numerical score.
Does Anticholinergic Medication Cause Dementia?
Research has found associations between long-term anticholinergic exposure and dementia, but causality remains less certain.
This distinction matters because people taking anticholinergic medicines often have underlying illnesses that may themselves affect cognitive health.
ACMI is specifically designed around delirium and fall risk, not as a dementia-prediction test.
It should therefore not be presented as a tool for identifying people who will develop dementia.
A High ACMI Score Does Not Mean a Fall Is Certain
Like any clinical prediction model, ACMI estimates probability.
An older adult with a high score may never experience delirium or a fall.
Likewise, somebody with a low medication-related score may still fall because of:
- muscle weakness;
- neurological disease;
- poor eyesight;
- environmental hazards;
- low blood pressure;
- infection;
- other medications.
The value of ACMI lies in identifying one potentially modifiable component of risk.
Medication exposure can sometimes be adjusted, unlike age or many chronic diseases.
Researchers Still Need to Prove It Reduces Harm
The model has been developed, externally tested and implemented in clinical systems.
The next important question is whether using it actually improves patient outcomes.
Researchers will need to determine whether routine ACMI-guided medication reviews lead to:
- fewer falls;
- fewer episodes of delirium;
- fewer hospital admissions;
- safer medication regimens;
- improved quality of life.
Current implementation studies are examining how feasible the approach is across hospital settings.
Until those outcome data are available, it is more accurate to say that ACMI identifies patients at elevated risk and supports medication review, rather than claiming it has already been proven to prevent falls.
Why the Tool Could Matter
Medication-related harm is one of the few risk factors for falls and delirium that healthcare systems can potentially modify proactively.
The major advance of ACMI is therefore not simply another scoring system.
It combines three elements:
large real-world health datasets + clinically relevant risk prediction + direct electronic-record integration.
That could allow medication safety checks to occur automatically as part of routine prescribing.
Instead of waiting until an older patient is admitted after a fall or develops delirium, doctors and pharmacists may be able to identify excessive anticholinergic burden earlier and review whether every medication is still necessary.
For older adults taking multiple medicines, that could turn a hidden cumulative risk into something clinicians can see and potentially act on, before serious harm occurs.
