MS left clues years early. They still couldn't predict it.

Cohort study. 113,876 people. biennial questionnaires from 1989 (data to 2017-2019); MS cases with first symptom 1989-2009; up to ~2 decades of data before first symptom.

People were watched for years. Nobody was told what to do. A clue, not proof.

Who: US female registered nurses aged 25-42 at recruitment in 1989 (mean age 34.4 at baseline), predominantly white and highly educated; 336 developed definite/probable MS (mean age at first symptom 42.3, at diagnosis 48.8)

• Less exercise~15 yrsbefore 1st symptom
• Migraine (vs 19.3%)22.8%at first symptom
• Predicting MS early≈chancenear coin-flip

More to know, from our read of the paper:

• Observational and descriptive: trajectories are loess curves of z-scores / excess prevalence versus the whole cohort; the text gives almost no confidence intervals or tests for the pre-symptom differences, so timings ('~10', '~15 years before') are read from figures as described in the Results
• Only 336 MS cases: modest power (authors' own limitation)
• Prediction models built from these signs did NOT predict MS before the first symptom (AUC 0.479-0.567): the signs are non-specific and cannot be used to predict MS. Never suggest viewers can spot MS from migraine, bladder leaks or less exercise
• Abstract lists depression among conditions more prevalent before MS, but the Results say depression rates were similar to the cohort until the first symptom and rose markedly only afterwards. Do not claim depression came before MS
• Women only, mostly white, highly educated nurses; older at onset than typical MS (first symptom 42.3, diagnosis 48.8 vs typical late 20s-early 30s) and diagnosed in the pre-MRI/pre-McDonald era (~6.5 years from first symptom to diagnosis), so may not generalise

Researchers followed 113,876 US nurses, all women, with health surveys every two years; 336 later developed MS. Those who got MS exercised less from about 15 years before their first symptom, and migraine, bladder leaks and restless legs became more common from about 10 years before. At the first symptom, 22.8% had migraine versus 19.3% of the whole group. But a model using these signs could not predict MS before symptoms began, so they are too common to be warning signs on their own. Observational: linked, not proven.

We read the abstract and the full text of this paper and checked every number on this page against it (16 facts, checked on 2026-10-29). The study design is named above. Cohort and other observational studies are described as linked, not proven. Found an error? Email theis@onestudyaday.com.