Can treating sleep apnoea reduce my dementia risk?

Not yet established by randomised controlled trial. The Monash Healthy Brain Project found an association between OSA and higher dementia risk scores in adults aged 40–70, but observational studies cannot prove that treating OSA prevents dementia. What is well-established is that CPAP therapy improves sleep quality, reduces daytime sleepiness, and lowers blood pressure — all of which matter for brain health independently. If you think you might have sleep apnoea, a GP-referred sleep study is the practical starting point.

Sleep apnoea in midlife raises dementia risk — Monash data

What just happened

New research from Monash University has added evidence to a question clinicians have been circling for years: is obstructive sleep apnoea a modifiable risk factor for dementia, or simply a comorbidity that travels with the same people?

The Monash Healthy Brain Project, published in Alzheimer’s & Dementia in June 2026, recruited 2,795 cognitively unimpaired Australians between the ages of 40 and 70. Approximately 7% reported a diagnosis of obstructive sleep apnoea (OSA). Those with OSA performed measurably worse on standardised memory tests and carried substantially higher composite dementia risk scores than the rest of the cohort.

Critically, the researchers found the association held even after controlling for APOE ε4 allele status — the genetic variant most strongly linked to Alzheimer’s disease. OSA was not simply a proxy for genetic predisposition.

Lead author Gabriel Abdelmessih put the implication plainly: “Sleep apnoea is common, frequently undiagnosed, and highly treatable, yet it is not often considered in discussions about dementia risk.”

The both-and

The confounders are real — and still don’t let OSA off the hook

The researchers were careful to note an important caveat: people with OSA are considerably more likely to also have obesity, hypertension, and elevated cholesterol. Each of these is an independent risk factor for dementia. Some proportion of the dementia risk signal in this study may reflect that cluster of metabolic and cardiovascular problems, rather than OSA acting directly on the brain.

That is not a reason to dismiss the finding. It is a reason to think about it differently.

If OSA, metabolic syndrome, and cardiovascular risk frequently cluster together — and the evidence strongly suggests they do — then OSA becomes a useful clinical entry point into that cluster. A sleep study can identify it. CPAP therapy or mandibular advancement devices can treat it. In a patient whose hypertension is being managed but whose sleep is fragmented by forty or sixty arousals a night, the OSA may be the load-bearing piece the workup has missed.

The Monash team called explicitly for randomised controlled trials to investigate whether targeting OSA alongside its vascular comorbidities can slow cognitive decline. That evidence does not yet exist at scale. The maybe verdict is about that gap — not about whether the question is worth taking seriously.

The women in this picture are not visible yet

OSA is still largely imagined as a condition affecting overweight middle-aged men who snore loud enough to wake the neighbours. The clinical literature has been correcting this framing for over a decade. The corrections have not fully reached the consultation room.

Women with OSA present differently. The textbook triad — loud snoring, witnessed apnoeas, excessive daytime sleepiness — is more characteristic of male presentations. Women are more likely to report fatigue, low mood, morning headaches, unrefreshing sleep, and frequent waking. These symptoms overlap closely with perimenopause, iron deficiency, thyroid dysfunction, and depression — all of which are more prevalent in women in their 40s, and all of which get investigated first. Often, repeatedly.

The consequence is that women are diagnosed with OSA significantly later than men, after longer diagnostic delays. In the context of a study suggesting OSA is tied to accelerated dementia risk scores precisely in the 40-to-70 window, that delayed diagnosis is worth naming directly.

2 cents

The evidence on OSA and dementia is not settled enough to make a definitive recommendation: treat your sleep apnoea to protect your brain. That conclusion is premature, and framing it that way would overstate what this study shows.

What is reasonable — and what the Monash data supports — is putting OSA on the list of things to investigate in a midlife patient who is fatigued, cognitively foggy, sleeping badly, or asking about dementia risk. Particularly if that patient is a woman whose symptoms have been attributed to hormones or stress for years.

A STOP-BANG questionnaire or Epworth Sleepiness Scale at a general practice appointment takes a few minutes. A home sleep study, if the screening is positive, is far less disruptive than it used to be. The question is low-risk to ask. The potential benefit — identifying a common, treatable condition in the decade that arguably matters most for brain health — is worth asking it.

Verdict: maybe — biologically plausible, clinically actionable now, but causal certainty waits on randomised trials.

Sources cited

RACGP newsGP — Study probes sleep apnoea link to dementia risk. https://www1.racgp.org.au/newsgp/clinical/study-probes-sleep-apnoea-link-to-dementia-risk
Abdelmessih G et al. Alzheimer’s & Dementia, June 2026. doi:10.1002/alz.71553. https://doi.org/10.1002/alz.71553