The Decaf Question: What the Evidence Says About Caffeine Restriction in Care Settings

CARE MARKET INSIGHT

Weighing clinical benefit against the fundamental right to choice across older adult, dementia and addiction recovery services.

Walk into many a care home, dementia unit or rehabilitation service and you will find a quiet policy decision has been made on everyone's behalf: the coffee and tea served are decaffeinated, full stop. Often it is well intentioned — staff have seen residents settle, sleep improve, agitation ease. But a blanket switch to decaf is also a restriction of choice, and in regulated care that is never a neutral act. This piece weighs the evidence across three populations and asks the question that matters most: who decided, and on what basis?

Why caffeine matters more as we age

Caffeine is the most widely consumed psychoactive substance in the world, and its pharmacology does not stand still as people grow older. The body metabolises caffeine more slowly with age, and changes in kidney function can prolong its effects — meaning a mid-afternoon cup may linger in an older person's system far longer than it would in a younger adult. The same dose, in other words, is not the same dose.

That slower clearance is the mechanism behind most of the clinical concern. Caffeine lengthens the time taken to fall asleep and reduces deep, slow-wave sleep through its action on adenosine receptors. In a population where roughly a quarter of older adults already fall short of recommended sleep, and where poor sleep is linked to cognitive decline, reduced quality of life and greater caregiver burden, that effect is not trivial.

The evidence in dementia care

Dementia care is where the case for restriction is strongest — though still more modest than is often assumed. A frequently cited Dutch interventional study gradually removed afternoon and evening caffeine from 21 nursing home residents with dementia in a special care unit. It found a statistically significant improvement in sleep scores and, interestingly, in apathy. What it did not find was any significant change in agitation, aggression, irritability or aberrant motor behaviour — the very symptoms staff most often hope decaf will calm.

An earlier observational study of 29 residents in the same kind of setting pointed in a similar direction: the total caffeine consumed across the day, and the amount taken after 6pm, were both associated with how often residents woke during the night. The relationship with apathy ran the other way, suggesting caffeine may have a complex rather than uniformly negative role.

The honest summary is this: the evidence supports timing restrictions — removing caffeine in the afternoon and evening — far more strongly than it supports a total ban. Both studies are small pilots in single units, and their authors explicitly call for larger controlled trials before firm conclusions are drawn. Eliminating the morning cup, which most of the sleep benefit does not depend on, sacrifices a valued pleasure for little measurable gain.

The evidence in older adult services more broadly

Outside dementia, the picture becomes genuinely two-sided, and this is where blanket policies start to look shaky. Moderate caffeine intake is associated with improved alertness, better concentration and a possible protective role in cognitive decline — benefits that matter in a population at risk of daytime drowsiness and under-stimulation.

More striking still is a large cross-sectional study of caffeine and sleep in older adults, which found that older women who abstained from caffeine reported more sleep disturbance and were more than twice as likely to be short sleepers than those who drank it. The authors caution this may reflect reverse causation — women who already sleep poorly may have given up caffeine — but it is a useful corrective to the assumption that less caffeine automatically means better sleep. There is also a hydration dimension: caffeinated drinks still count meaningfully towards fluid intake, and a familiar, enjoyed cup of tea is sometimes the most reliable way to keep a frail older person drinking at all.

The evidence in drug and alcohol recovery

Addiction services are a distinct case, and the rationale here is psychological as much as physiological. Caffeine produces a dopamine response — not large enough to be classed by the American Psychiatric Association as a substance use disorder, though the World Health Organization recognises caffeine dependence as a clinical disorder, and DSM-5 includes caffeine withdrawal and intoxication while listing caffeine use disorder as a condition needing further research.

The practical concerns in early recovery are real. Caffeine can heighten anxiety, and anxiety disorders frequently co-occur with substance use disorders, potentially raising relapse risk. It can also worsen the sleep disturbance that is common when someone has recently stopped using alcohol or drugs. There is a further argument that a structured period without dependence on any mood-altering substance supports the wider goal of establishing healthier patterns. Surveys have long noted that coffee consumption is markedly higher among people in recovery than in the general population — a habit that can itself become compulsive.

Even here, though, the better evidence-informed practice is usually mindful reduction, particularly in the earliest and most vulnerable phase, rather than a permanent prohibition imposed without discussion. The goal is the individual's recovery, not abstinence from caffeine as an end in itself — and a service that removes choice reflexively risks recreating the very powerlessness recovery seeks to undo.

Weighing it up: the balance sheet

Across all three populations the same tension recurs. The clinical benefits are real but modest and largely tied to timing; the costs to autonomy, dignity and pleasure are immediate and felt daily. The table below summarises the trade-off.

Potential benefits of restriction	Costs and risks of restriction
Improved sleep onset and deep sleep, especially where caffeine is removed in the afternoon and evening	Loss of a valued daily pleasure and routine that supports wellbeing and quality of life
Reduced night-time waking in some residents with dementia	Removal of the alertness and concentration benefits of moderate caffeine, risking under-stimulation
Lower anxiety and relapse risk in early addiction recovery	Potential reduction in fluid intake if familiar drinks are withdrawn, raising dehydration risk
A consistent, simple policy that is easy for staff to apply	A blanket policy applied to people for whom caffeine causes no harm — a restriction without justification
Possible reduction in apathy in some dementia residents (evidence mixed)	Risk of regulatory non-compliance where choice is restricted without individual assessment

The choice dimension: what the Standards require

In Scotland this is not merely an ethical preference; it is a regulatory one. The Health and Social Care Standards are explicitly human-rights based. Standard 1.3 is unambiguous: where a person's independence, control and choice are restricted, this must comply with relevant legislation, be justified, kept to a minimum and carried out sensitively. A service-wide decaf-only policy applied to everyone, regardless of individual need, is by definition not kept to a minimum.

The wellbeing standards reinforce this. People should be able to choose suitably presented meals and snacks, participate in menu planning, and have their personal preferences respected. A resident who has drunk strong tea every morning for sixty years has a legitimate expectation that this continues unless there is a specific, assessed reason it should not. Where capacity is in question, the Adults with Incapacity (Scotland) Act 2000 framework — least restrictive intervention, benefit to the individual — points in exactly the same direction.

The risk for providers is twofold. A blanket restriction is hard to defend at inspection because it cannot show individualised justification. And it quietly erodes the person-centred culture that good care depends on, substituting institutional convenience for the resident's own voice.

A proportionate way forward

None of this argues against ever restricting caffeine. It argues against doing so by default. A defensible, inspection-ready approach looks like this:

•       Make caffeine a matter of individual assessment and personal plan, not a house rule — record the person's preferences, any clinical rationale, and their (or their proxy's) views.

•       Where there is a clinical case, prefer timing restrictions (decaf from early afternoon) over total elimination, mirroring where the evidence is actually strongest.

•       Always offer genuine choice — keep both caffeinated and decaffeinated options available rather than removing one entirely.

•       For people in early addiction recovery, frame reduction as a supported, time-limited part of the recovery plan, discussed with the individual, not an imposed prohibition.

•       Document the justification, review it regularly, and ensure any restriction is the least restrictive option that meets the assessed need.

•       Protect hydration — if a caffeinated drink is the one a person will reliably accept, that fact carries real weight.

In short

The evidence offers modest, timing-specific support for reducing caffeine in dementia care and early addiction recovery, a genuinely mixed picture in older adult services generally, and almost no support anywhere for an indiscriminate, permanent ban. Set against that is a clear, rights-based duty to preserve choice and to justify any restriction individually. The well-run service is not the one that has quietly switched everyone to decaf. It is the one that can explain, for each person, why their cup contains what it does — and can show the person had a say in it.

Mac Research and Consultancy Limited

Specialist adult social care consultancy across Scotland and England — regulatory compliance, quality assurance, mock inspections and policy development. We help services build defensible, person-centred policy that stands up to inspection.

www.macresearchandconsultancy.co.uk