Abstract

BACKGROUND: Previous research has linked higher exposure to air pollution to increased cognitive impairment at older ages. We aimed to extend the existing evidence in this area by incorporating exposures across the life course in addition to measures of cognition and brain structural imaging in participants at midlife to older age. METHODS: For this population-based study, we used data from the Medical Research Council National Survey of Health and Development (NSHD; also known as the 1946 British Birth Cohort) and a neuroimaging substudy of the NSHD known as Insight 46. Participants were recruited after birth in a single week during March, 1946. Our objectives were to assess whether exposure to air pollutants in midlife (age 45-64 years) was associated with poorer processing speed and poorer verbal memory between the ages of 43 years and 69 years, and whether exposures were associated with poorer cognitive state and brain structure outcomes at age 69-71 years. Air pollution exposure data were available for nitrogen dioxide (NO2; ages 45-64 years); particulate matter with diameter less than 10 μm (PM10; ages 55-64 years); and nitrogen oxides (NOx) and particulate matter with diameters less than 2·5 μm (PM2·5) and between 2·5 μm and less than 10 μm (PMcoarse) and particulate matter absorbance (PM2·5abs) as a measure of black carbon absorption (ages 60-64 years), with adjustments for early-life exposures to black smoke and sulphur dioxide. Verbal memory was tested with a 15-item recall task and processing speed with a visual search task at ages 43, 53, 60-64, and 69 years. The Addenbrooke's Cognitive Examination III (ACE-III), a measure of cognitive state, was conducted at age 69 years. Whole-brain, ventricular, hippocampal, and white matter hyperintensity volumes were assessed by MRI at age 69-71 years. Generalised linear models and generalised mixed linear models were used to explore associations between pollution exposure, cognitive measures, and brain structural outcomes, adjusted for sociodemographic factors including smoking status and neighbourhood deprivation. FINDINGS: Between the ages of 43 years and 69 years, we included 1534 NSHD participants in the verbal memory and processing speed analysis. Of 2148 participants who underwent testing during the wave of follow-up in 2015-16, at age 69 years, 1761 were included in the ACE-III analysis. Of the 502 NSHD participants recruited into the Insight 46 substudy, 453 were included in the analysis. Higher exposure to NO2 and PM10 was associated with slower processing speed between the ages of 43 years and 69 years (NO2 β -8·121 [95% CI -10·338 to -5·905 per IQR increase in exposure]; PM10 β -4·518 [-6·680 to -2·357]). Higher exposure to all tested pollutants was associated with lower ACE-III score at age 69 years (eg, NO2 β -0·589 [-0·921 to -0·257]). Higher exposure to NOx was associated with smaller hippocampal volume (β -0·088 [-0·172 to -0·004]) and higher exposure to NO2 and PM10 was associated with larger ventricular volume (NO2 β 2·259 [0·457 to 4·061]; PM10 β 1·841 [0·013 to 3·669]) at age 69-71 years. INTERPRETATION: Acknowledging the probable effects of exposure early in life, higher exposure to nitrogen dioxide, nitrogen oxides, and coarse particulate matter in midlife to older age was associated with poorer cognition, processing speed, and brain structural outcomes, strengthening evidence for the adverse effects of air pollution on brain function in older age. FUNDING: The National Institute for Health and Care Research, the Medical Research Council (MRC), Alzheimer's Research UK, the Alzheimer's Association, MRC Dementias Platform UK, and Brain Research UK.