Light at night exposure and risk of dementia conversion from mild cognitive impairment in a Northern Italy population.

Background: A few studies have suggested that light at night (LAN) exposure, i.e. lighting during night hours, may increase dementia risk. We evaluated such association in a cohort of subjects diagnosed with mild cognitive impairment (MCI).

Methods: We recruited study participants between 2008 and 2014 at the Cognitive Neurology Clinic of Modena Hospital, Northern Italy and followed them for conversion to dementia up to 2021. We collected their residential history and we assessed outdoor artificial LAN exposure at subjects' residences using satellite imagery data available from the Visible Infrared Imaging Radiometer Suite (VIIRS) for the period 2014-2022. We assessed the relation between LAN exposure and cerebrospinal fluid biomarkers. We used a Cox-proportional hazards model to compute the hazard ratio (HR) of dementia with 95% confidence interval (CI) according to increasing LAN exposure through linear, categorical, and non-linear restricted-cubic spline models, adjusting by relevant confounders.

Results: Out of 53 recruited subjects, 34 converted to dementia of any type and 26 converted to Alzheimer's dementia. Higher levels of LAN were positively associated with biomarkers of tau pathology, as well as with lower concentrations of amyloid β assessed at baseline. LAN exposure was positively associated with dementia conversion using linear regression model (HR 1.04, 95% CI 1.01-1.07 for 1-unit increase). Using as reference the lowest tertile, subjects at both intermediate and highest tertiles of LAN exposure showed increased risk of dementia conversion (HRs 2.53, 95% CI 0.99-6.50, and 3.61, 95% CI 1.34-9.74). In spline regression analysis, the risk linearly increased for conversion to both any dementia and Alzheimer's dementia above 30 nW/cm/sr of LAN exposure. Adding potential confounders including traffic-related particulate matter, smoking status, chronic diseases, and apolipoprotein E status to the multivariable model, or removing cases with dementia onset within the first year of follow-up did not substantially alter the results.

Conclusion: Our findings suggest that outdoor artificial LAN may increase dementia conversion, especially above 30 nW/cm/sr, although the limited sample size suggests caution in the interpretation of the results, to be confirmed in larger investigations.