To answer this question you would need to find the source of the information. I found the study mentioned in the article in the Toronto Star (Chen, et al. 2017); and while studying risks of dementia, Parkinson's disease and multiple sclerosis, with emphasis mine, they said:
In this population-based cohort study, we assembled two population-based cohorts including all adults aged 20–50 years (about 4·4 million; multiple sclerosis cohort) and all adults aged 55–85 years (about 2·2 million; dementia or Parkinson's disease cohort) who resided in Ontario, Canada on April 1, 2001. Eligible patients were free of these neurological diseases, Ontario residents for 5 years or longer, and Canadian-born.
Between 2001, and 2012, they identified
- 243,611 incident cases of dementia,
- 31,577 cases of Parkinson's disease, and
- 9,247 cases of multiple sclerosis.
They provided adjusted hazard ratios in the study - a comparison of the effect of different variables on survival or other outcomes that develop over time (Dawson, 2008). See also https://www.sciencedirect.com/topics/medicine-and-dentistry/hazard-ratio
The adjusted hazard ratio of incident dementia was
- 1·07 for people living less than 50 m from a major traffic road (95% CI 1·06–1·08),
- 1·04 (1·02–1·05) for 50–100 m,
- 1·02 (1·01–1·03) for 101–200 m, and
- 1·00 (0·99–1·01) for 201–300 m
versus further than 300 m (p for trend=0·0349).
No association was found with Parkinson's disease or multiple sclerosis.
Based on the averages used in the study, statistically speaking you will need to be living at least 200 metres (just over 218.5 yards) away from any major road to be relatively free of risk.
Chen, H., Kwong, J. C., Copes, R., Tu, K., Villeneuve, P. J., Van Donkelaar, A., ... & Wilton, A. S. (2017). Living near major roads and the incidence of dementia, Parkinson's disease, and multiple sclerosis: a population-based cohort study. The Lancet, 389(10070), 718-726.
Dawson, G. F. (2008). Measures of Effect. In: Easy Interpretation of Biostatistics: The Vital Link to Applying Evidence in Medical Decisions. Philadelphia, PA: Saunders
DOI: 10.1016/B978-1-4160-3142-0.50027-4 (Free Preview)