What comes before Parkinson's, and how many years before

Almost every article on early Parkinson's is a list. Ten signs, unordered, each described in a sentence. The format implies the signs are interchangeable and roughly simultaneous.

They are not. They arrive in a sequence, and the sequence spans decades.

A study that watched it happen

Between 2004 and 2016, researchers recruited 154 people with polysomnography- proven idiopathic REM sleep behaviour disorder — a condition where people physically act out their dreams — and followed them annually. REM sleep behaviour disorder is one of the strongest known predictors of later Parkinson's, so this was a group with an unusually high chance of converting.

Fifty-five of them did, to parkinsonism or dementia. Because the researchers had been measuring olfaction, colour vision, cognition, autonomic function and motor performance every year, they could then look backwards from each person's diagnosis and ask when each feature had actually started.

The order

Smell goes first. "Olfactory loss was first to develop, with predicted onset

20 years before phenoconversion." Two decades.

Then, 10–16 years ahead: impaired colour vision, constipation, and erectile dysfunction.

7–9 years ahead: slight urinary dysfunction and subtle cognitive decline.

7–11 years ahead, motor things begin to change, though mildly — "altered handwriting, turning in bed, walking, salivation, speech, and facial expression". These are the changes a person might notice and attribute to age.

5–7 years ahead: abnormalities detectable on motor examination.

Last, the symptoms everyone associates with Parkinson's. Bradykinesia — slowness — appeared first among the cardinal motor features, about 5–6 years before diagnosis. Then rigidity at about 3 years. Then tremor, at about 2 years.

Tremor is the sign in the public imagination, the thing that makes people think of Parkinson's at all. In this cohort it was close to the last thing to arrive, roughly eighteen years after the loss of smell.

Why this doesn't mean what it looks like it means

Read the list above and the natural next thought is: I have constipation and my sense of smell isn't what it was. Two things make that inference much weaker than it feels.

This was not the general population. Every participant had confirmed REM sleep behaviour disorder — a group already at very high risk. Timelines measured in an enriched cohort do not transfer directly to someone without that condition. The sequence is likely informative; the probabilities are not.

The individual signs are common and mostly mean nothing. Constipation is near-universal. Smell fades with age, with infection, with smoking. A separate meta-analysis of risk factors put constipation at a relative risk of 2.34 for later Parkinson's diagnosis. That is a real signal — but it doubles a small baseline, and doubling a small number leaves a small number.

That same meta-analysis found the strongest associations were a first-degree relative with Parkinson's (odds ratio 3.23), any relative with it (4.45), any relative with tremor (2.74), constipation (2.34), and lack of smoking history (0.44 for current versus never smokers, meaning smokers were diagnosed less often). Note that the strongest predictors are family history — not symptoms at all.

What it is actually good for

Not self-diagnosis. Two things instead.

It explains why Parkinson's is not caught early. By the time the diagnostic features are present, the process has been running for perhaps twenty years. Any treatment aimed at protecting neurons is arriving very late in a story that started long before anyone was looking. This is why prodromal research matters so much to people designing trials.

It reframes what the disease is. A condition that begins with the loss of smell two decades before a tremor is not really a movement disorder that happens to have other symptoms. The movement problems are a late chapter.

If something has genuinely changed for you — particularly if you act out dreams in your sleep, which is the feature this study was built around — that is worth raising with a doctor. Not because a checklist matched, but because it is information a clinician can put next to everything else about you. A list of ten signs cannot do that, and neither can this article.