Five new studies aim to improve exposure assessment

There are many challenges in conducting epidemiological studies of long-term exposure to air pollutants given that concentrations vary so widely over space and time. One difficulty is how to accurately determine the exposures of individuals to air pollution. Another is how to quantify the influence of errors in measuring exposure on health-risk estimates. HEI is embarking on studies to enhance exposure assessment that were funded under request for applications (RFA) 19-1, "Applying Novel Approaches to Improve Long-Term Exposure Assessment of Outdoor Air Pollution for Health Studies." These five new studies are summarized below.

Strategies for Enhanced Exposure Assessment

Three studies will focus on combining novel methods for measuring air pollution and diverse exposure assessment approaches to improve exposure assignment in well-established cohorts.

Scott Weichenthal of McGill University and colleagues will evaluate health impacts of long-term exposures to traffic-related air pollution using exposure estimates from fixed-site and mobile measurement campaigns, as well as deep learning models, in Toronto and Montreal, Canada. They will compare exposure estimates generated by these models to present-day and historical measurements, and to each other. They plan to estimate concentration–response relationships for nonaccidental and cause-specific mortality in the Canadian Census Health and Environment Cohort (CanCHEC), and evaluate how the magnitudes and shapes of those relationships are influenced by different exposure models.

Gerard Hoek of Utrecht University and colleagues will prepare maps of modeled annual average air pollution across the Netherlands, validate the maps using new measurements from over 100 sites, and evaluate the performance of several exposure models. For example, they plan to use measurements from low-cost sensors, mobile monitoring, and a national network of air pollution monitors. The investigators will conduct cross-comparisons to evaluate how different exposure assessment methods compare in their ability to predict long-term pollutant concentrations, with a particular focus on spatial variability of pollutants. Furthermore, they will apply the various models to three cohorts in the Netherlands to evaluate how they influence health effect estimates in epidemiological studies.

Kees de Hoogh of the Swiss Tropical and Public Health Institute and colleagues plan to improve our understanding of the contribution of individual mobility in air pollution exposure estimates. They will use location tracking on a mobile phone application for 2,000 individuals in the Netherlands and Switzerland. These data, together with available data on air pollutant concentrations, will then be used to estimate long-term hourly exposure estimates. Subsequently, these exposure estimates will be compared against exposures estimated using home addresses. The team will then apply their findings to three major cohorts: the Study on Air Pollution and Lung Disease in Adults (SAPALDIA) in Switzerland, participants in the European Prospective Investigation into Cancer and Nutrition Netherlands (EPIC-NL), and the Swiss National Cohort. They plan to evaluate whether the accuracy of health effects estimates for respiratory diseases can be improved using the enhanced exposure estimates they will develop. Through this study, the researchers aim to generate a conceptual framework and novel methods for exposure assessment taking human mobility into account, and produce open-source software for conducting such analyses.

Quantifying Influence of Exposure Error

Two studies will test the added value of incrementally more complex statistical modeling approaches to improving exposure assessment and apply their findings to estimating health effects in epidemiological studies.

Klea Katsouyanni of King’s College London will lead a team to investigate the consequences of measurement error on estimates of health effects of long-term exposure to outdoor air pollution in London by developing increasingly sophisticated exposure models. They plan to compare exposure models that account for mobility, are based on exposure estimates at the residential address, and are based on concentrations measured at the nearest air pollution monitor. The air pollution surfaces will include outputs from several types of air pollution models (a chemical transport model, land use regression, and machine learning) and combinations of these models. Finally, exposures will be applied to the London segment of the UK Biobank study to evaluate associations with asthma, chronic obstructive pulmonary disease, myocardial infarction, stroke incidence, and mortality.

Lianne Sheppard of the University of Washington and colleagues will compare and contrast scientific and logistical benefits of different approaches to air pollution exposure assessment. The investigators will leverage large air pollution datasets obtained from low-cost sensors, mobile monitoring, and passive samplers. They will apply the exposure assessment approaches to determine associations with cognitive decline and dementia incidence in an ongoing cohort study, Adult Changes in Thought Air Pollution (ACT-AP). In particular, the investigators plan to use statistical techniques to assess the bias and precision of health effect estimates. They also plan to compare the value of specific information by including increased time and costs spent on more sophisticated exposure assessment activities. They hope this will guide future studies in efficient selection of exposure assessment methods.

* * *

Contact Allison Patton or Pallavi Pant for more information.