The FINANCIAL — Exposure to air pollution in childhood is linked to a decline in thinking skills in later life, a recent study suggests.
Researchers found that greater exposure to air pollution at the very start of life was associated with a detrimental effect on people’s cognitive skills up to 60 years later.
The study was funded by Scottish Government Chief Scientist Office, Natural Environment Research Council and Medical Research Council.
The University of Edinburgh led the study, in partnership with UK Centre for Ecology and Hydrology, The University of Manchester. Testing 500 people.
The study tested the general intelligence of more than 500 people aged approximately 70 years using a test they had all completed at the age of 11 years.
The participants then repeated the same test at the ages of 76 and 79 years.
A record of where each person had lived throughout their life was used to estimate the level of air pollution they had experienced in their early years.
The team used statistical models to analyze the relationship between a person’s exposure to air pollution and their thinking skills in later life.
They also considered lifestyle factors, such as:
Findings showed exposure to air pollution in childhood had a small but detectable association with worse cognitive change between the ages of 11 and 70 years.
This study shows it is possible to estimate historical air pollution and explore how this relates to cognitive ability throughout life, researchers say.
Reducing risk for future generations
“For the first time, we have shown the effect that exposure to air pollution very early in life could have on the brain many decades later. This is the first step towards understanding the harmful effects of air pollution on the brain and could help reduce the risk of dementia for future generations.
Researchers say until now it has not been possible to explore the impact of early exposure to air pollution on thinking skills in later life. This is due to a lack of data on air pollution levels before the 1990s when routine monitoring began”, Dr Tom Russ, Director of the Alzheimer Scotland Dementia Research Centre at the University of Edinburgh, said.
For this study researchers used a model called the EMEP4UK atmospheric chemistry transport model to determine pollution levels — known as historical fine particulate matter (PM2.5) concentrations — for the years:
They combined these historical findings with contemporary modelled data from 2001 to estimate life course exposure.
The participants were part of the Lothian Birth Cohort 1936 study, a group of individuals who were born in 1936 and took part in the Scottish Mental Survey of 1947.
Since 1999, researchers have been working with the Lothian Birth Cohorts to chart how a person’s thinking power changes over their lifetime.
The study is published in the Journal of Alzheimer’s Disease and in the Handbook on Air Pollution.
Read the full air pollution study (IOS Press).
Researchers were supported by Age UK (Disconnected Mind project) and Alzheimer Scotland.
A total of 572 LBC1936 participants were included in the present analyses. Their characteristics are summarized in Table 1. Briefly, just under half were female, and had completed more than compulsory education. Just over a quarter had parents from occupational social classes I or II (i.e., less deprived), and about half were smokers at the time of recruitment to the LBC1936. Comparing the 572 LBC1936 participants for whom we had location (and therefore air pollution) data with the 519 participants excluded from these analyses revealed no major differences.
|Includeda||Excludedb||pc||Total LBC1936 sample|
|Age at SMS1947 (mean [SD] years)||10.92 (0.27)||10.96 (0.29)||0.027||10.94 (0.28)|
|Age 11 IQd (mean [SD])||101.6 (15.0)||98.2 (14.9)||<0.001||100.0 (15.0)|
|Parental occupational social class (% class I or II)||27.7||26.3||0.011||27.1|
|Current smoker at baseline (%)||49.3||42.2||0.022||45.9|
aParticipants were included if they had at least one location recorded for each time period. bExcluded participants included 21 with missing location data for at least one time period, 111 who did not respond to the questionnaire requesting lifetime residential history, and 387 who were not approached, mainly because they had died or withdrawn from the study prior to the questionnaire being used in 2014. cp-values from comparisons of included and excluded participants d31 participants were missing age 11 intelligence data.
LBC1936: Lothian Birth Cohort 1936 (N = 1091); SMS1947: Scottish Mental Survey 1947 (N = 70,805, of which the LBC1936 is a subset).
Table 2 shows the average air pollution estimates for the LBC1936 participants and Supplementary Figure 1 shows the distribution of air pollution exposure at each time period. Supplementary Figure 2 shows participants’ PM2.5 exposure changes over time and Supplementary Table 1 shows the correlations between individuals’ PM2.5 exposure ranking at different time points. Rankings varied over time—likely due more to participants moving than the relative ranking of areas changing—but were more closely correlated when closer in time, suggesting it is feasible to explore critical/sensitive time periods. Figure 2 shows the modelled PM2.5 values for Scotland in 1935; the urban centers are clearly visible.
|Year||Mean (sd)||Range||Ntotala||>10 μg/m3 b|
|1935||34.8 (16.0)||5.2–133.0||590||562 (95%)|
|1950||32.4 (12.8)||6.0–113.3||591||578 (98%)|
|1970||17.0 (1.5)||9.5–23.9||585||584 (100%)|
|1980||15.0 (1.5)||7.3–24.0||580||575 (99%)|
|1990||13.4 (1.2)||6.7–21.4||580||579 (100%)|
|2001||7.9 (0.6)||4.8–15.9||591||4 (0.7%)|
a593 participants provided lifetime residential histories; 572 had air pollution data from all time periods and were included in the present analyses; bThe number (%) of participants whose PM2.5exposure exceeded the WHO guidelines of an annual mean of≤10μg/m3.
|(a) Change in IQ between ages 11 and 70 years|
|In utero exposure to air pollution||–0.006 (0.002)||0.03|
|(b) IQ trajectories from age 70 to age 79 years|
|Intercept (average IQ at 70 years)||97.74 (1.38)||Rate of change (in IQ from age 70 to 79 years)||–0.11 (0.31)||0.71|
|Random Intercept Variance||71.12 (5.61)||Random Slopes variance||2.36 (0.31)||<0.001|
|Intercept-slope correlation||–3.02 (0.06)||<0.001|
|In utero exposure to air pollution||0.05 (0.02)||0.06||In utero exposure to air pollution||–0.006 (0.006)||0.36|
Model (a) is adjusted for sex, parental (father’s) occupation, and smoking; Model (b) is adjusted for sex, age 11 IQ, parental (father’s) occupation, and smoking; Coefficients (β) represent the change in IQ and rate of change per 1 μg/m3 increase in PM2.5.
|Level and change in IQ between ages 70, 76, and 79 years|
|IQ||β (SE)||p||β (SE)||p|
|Age 70 IQ||102.14 (1.62)||Rate of change in IQ from age 70–79||–0.14 (0.33)||0.46|
|Pollution 1950||–0.027 (0.04)||0.52||–0.001 (0.006)||0.84|
|Age 70 IQ||105.14 (5.56)||0.21 (1.13)||0.85|
|Pollution 1970||–0.22 (0.04)||0.46||–0.03 (0.06)||0.65|
|Age 70 IQ||96.38 (4.94)||0.84 (1.51)||0.57|
|Pollution 1980||0.32 (0.32)||0.32||–0.07 (0.10)||0.45|
|Age 70 IQ||99.39 (7.34)||1.46 (1.51)||0.33|
|Pollution 1990||0.14 (0.54)||0.79||–0.13 (0.11)||0.24|
|Age 70 IQ||103.21 (8.84)||–0.91(1.86)||0.62|
|Pollution 2001||–0.24 (1.10)||0.82||0.08 (0.23)||0.74|
Models adjusted for sex, age 11 IQ, parental (father’s) occupation, and smoking status; Coefficients (β) represent the change in IQ and rate of change per 1 μg/m3 increase in PM2.5.