Acting out dreams: what REM sleep behaviour disorder predicts

Physically acting out dreams during sleep — punching, kicking, shouting or leaping from bed — is one of the strongest early signs medicine has for Parkinson's disease, dementia with Lewy bodies and multiple system atrophy [s1]. In the largest study of the condition, 73.5% of patients developed one of these disorders within 12 years of follow-up, converting at roughly 6.3% per year [s1]. That is a risk marker and a research opportunity, not a diagnosis or a sentence: it identifies people who are, on average, in a prodromal state, and there is as yet no proven treatment to change what follows [s1][s2].

What the disorder is

In normal REM sleep — the stage where most vivid dreaming happens — the body's skeletal muscles are effectively paralysed, so a dreamer stays still. In REM sleep behaviour disorder (RBD) that paralysis fails, and the sleeper enacts the dream [s1]. The clinical version that carries the neurodegenerative risk is "idiopathic" RBD, meaning it is confirmed by an overnight sleep study showing loss of REM muscle atonia and is not explained by another cause such as certain antidepressants [s1]. It is distinct from ordinary restless nights, sleep-talking, or the trauma-linked nocturnal behaviours described in a disputed post-traumatic parasomnia, which lack this polysomnographic signature.

The evidence

The central study is a 2019 multicentre analysis in Brain from the International RBD Study Group, pooling prospective follow-up from 24 centres [s1]. It recruited 1,280 patients with polysomnographically confirmed idiopathic RBD and no parkinsonism or dementia at baseline; their average age was 66.3 and 82.5% were male, and they were followed for an average of 4.6 years, ranging out to 19 [s1]. The overall conversion rate to an overt neurodegenerative syndrome was 6.3% per year, reaching 73.5% by 12 years [s1].

The same study asked what, measured at baseline, foretold faster conversion. Abnormal quantitative motor testing carried a hazard ratio of 3.16, an abnormal clinical motor examination 3.03, an impaired sense of smell 2.62, mild cognitive impairment 1.91 to 2.37, and an abnormal dopamine-transporter (DAT) brain scan 1.98; erectile dysfunction, constipation, colour-vision abnormalities and loss of REM atonia were also predictive [s1]. Several things people might assume matter did not: sex, daytime sleepiness, insomnia, restless legs, sleep apnoea, urinary or orthostatic symptoms, depression and anxiety showed no significant predictive value [s1].

A 2018 systematic review and meta-analysis in Sleep Medicine Reviews pooling longitudinal studies reached a compatible, and if anything starker, long-run picture: the estimated risk of developing a neurodegenerative disease was 33.5% at five years, 82.4% at 10.5 years and 96.6% at 14 years [s2]. Across the studies the average conversion rate was 31.95% after a mean follow-up of 4.75 years, and of those who converted, 43% developed Parkinson's disease and 25% dementia with Lewy bodies [s2]. The authors put the long-term risk at more than 90% [s2].

What the numbers do and do not mean

These are among the highest predictive figures in prodromal neurology, which is exactly why RBD has become a magnet for research into treatments that might slow neurodegeneration before symptoms appear — the Brain study estimated that a definitive neuroprotective trial would need only 142 to 366 patients per arm, a tractable number [s1]. But the cohorts are drawn from specialised sleep centres, are heavily male, and describe idiopathic RBD specifically; they do not speak to a restless night in the general population, and RBD can also arise secondarily from medications or other conditions, which behave differently [s1][s2]. The long-run figures also reflect people who had already been referred and diagnosed, not everyone who ever kicks in their sleep.

Crucially, "prodromal" is a statement about populations followed over time, not a fixed timetable for any one person; conversion rates rise with the years of follow-up precisely because the risk is spread across a long horizon [s1][s2]. And no therapy has yet been shown to interrupt that trajectory — the predictive strength is currently useful mainly for research and for informed clinical follow-up, not for a cure [s2].

What it means

Acting out dreams is a specific, recognised sleep disorder that a clinician can confirm with an overnight study, and it is worth taking seriously rather than dismissing — but the appropriate response is medical assessment, not alarm or self-diagnosis from a single restless night. The finding that so many other sleep complaints carry no predictive weight is itself reassuring for the far larger number of people whose disrupted nights have ordinary causes [s1]. For another line of research into early Parkinson's risk markers, see what an exhaled-breath signal may flag. None of this is medical advice.

Sources

• Risk and predictors of dementia and parkinsonism in idiopathic REM sleep behaviour disorder: a multicentre study — Brain, 2019-02-20
• The risk of neurodegeneration in REM sleep behavior disorder: A systematic review and meta-analysis of longitudinal studies — Sleep Medicine Reviews, 2018-11-08