Search bioRxiv⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/504001

Accounting for measurement error to assess the effect of air pollution on omic signals

Ponzi, E.·Vineis, P.·Chung, K. F.·Blangiardo, M.
View original source ↗Export metadata ↓

Abstract

Studies on the effects of air pollution and more generally environmental exposures on health require measurements of pollutants, which are affected by measurement error. This is a cause of bias in the estimation of parameters relevant to the study and can lead to inaccurate conclusions when evaluating associations among pollutants, disease risk and biomarkers. Although the presence of measurement error in such studies has been recognized as a potential problem, it is rarely considered in applications and practical solutions are still lacking.\n\nIn this work, we formulate Bayesian measurement error models and apply them to study the link between air pollution and omic signals. The data we use stem from the \"Oxford Street II Study\", a randomized crossover trial in which 60 volunteers walked for two hours in a traffic-free area (Hyde Park) and in a busy shopping street (Oxford Street) of London. Metabolomic measurements were made in each individual as well as air pollution measurements, in order to investigate the association between short-term exposure to traffic related air pollution and perturbation of metabolic pathways. We implemented error-corrected models in a classical framework and used the flexibility of Bayesian hierarchical models to account for dependencies among omic signals, as well as among different pollutants. Models were implemented using traditional MCMC simulative methods as well as integrated Laplace approximation.\n\nThe inclusion of a classical measurement error term resulted in variable estimates of the association between omic signals and traffic related air pollution (TRAP) measurements, where the direction of the bias was not predictable a priori. The models were successful in including and accounting for different correlation structures, both among omic signals and among different pollutant exposures. In general, more associations were identified when the correlation among omics and among pollutants were modeled, and their number increased when a measurement error term additionally included in the multivariate models (particularly for the associations between metabolomics and NO2).

Source connections

Explore related subjects

epidemiologyPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Ponzi, E., Vineis, P., Chung, K. F., Blangiardo, M.. 2018-12-21. Accounting for measurement error to assess the effect of air pollution on omic signals. https://doi.org/10.1101/504001

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2019

Climatic Factors in Relation to Diarrhea for Informed Public Health Decision-Making: A Novel Methodological Approach

BackgroundDiarrheal disease is one of the leading causes of morbidity and mortality globally, particularly in children under 5 years of age. Factors related to diarrheal disease incidence include infection, malnutrition, and exposure to contaminated water and food. Climate factors also contribute to diarrheal disease. ObjectivesWe aimed to explore the relationship between temperature, precipitation and diarrhea case counts of hospital admissions among vulnerable communities living in a rural setting in South Africa. MethodsWe applied a novel approach of contour analysis to visually examine simultaneous observations in frequencies of anomalously high and low diarrhea case counts occurring in a season and assigning colors to differences that were statistically significant based on chi-squared test results. ResultsThere was a significantly positive difference between high and low groups when there was a lack of rain (0 mm of cumulative rain) for 1 to 2 weeks in winter for children under 5. Diarrhea prevalence was greater among children under 5 years when conditions were hotter than usual during winter and spring. DiscussionDry conditions may lead to increased water storage raising the risks of water contamination. Reduced use of water for personal hygiene and cleaning of outdoor pit latrines affect sanitation quality. Rural communities require adequate and uninterrupted water provision and healthcare providers should raise awareness about potential diarrheal risks especially during the dry season.

epidemiology↗
bioRxiv2019

Prevalence and Risk Factors for Child Labour and Violence against Child issues in Egypt using Bayesian Geospatial Modelling with Multiple Imputation

BackgroundThe incidence of child labour, especially across the developing nations is of global concern. The use of children in employment in developing economies constitutes a major threat to the societies, and concerted efforts are made by the relevant stakeholders towards addressing some of the factors/issues responsible for the menace. Significant risk factors include socio-demographic and economic factors, including poverty, neglect, lack of adequate care, and exposure of children to various grades of violence, parental education status, gender, place of residence, household size, residences type or size, wealth index, parental survivorship and household size. ObjectivesThis study, therefore, focuses on identifying socio-demographic/economic and geospatial factors associated with child labour participation. MethodsWe utilised 2014 Demographic and Health Survey (EDHS) from the Ministry health and Population in Egypt, with the record of 20, 560 never married children aged 5-17years engaging in economic activities, in and out of their home. The data focused on demographic and socio-economic characteristics of household members. Multivariate Bayesian geo-additive models were employed to examine the demographical and socio-economic factors for children working less than 16hrs; between 16 and 45hrs and above 45hrours weekly. ResultsThe results showed that at least 31.6% of the children in the age group from 5-10 were working, 68.5% of children aged 11-17 years engaging in child labour (wage), while 44.7 of the children in the age group from 5-10 were engaged in hazard work. From the multivariate Bayesian geo-additive models, the female children (with male children as reference category) working at least 16hrs (OR: 1.3; with 95% CI: 1.2-1.5) are more likely to be engaged in child labour than those working 16 to 45hours (OR: 1; 95% CI: 0.3-1.5). Children born to women without formal education, under non-hazardous jobs, irrespective of the hours spent at work, were mostly exposed to child labour with following percentages: 52.9%, 56.8%, 62.4%, compared to children of mothers with some levels of education. Finally, children that have experienced psychological aggression and physical punishment are mostly exposed to child labour than those without such experience across the job types and hours spent. ConclusionThis study revealed a significant influence of socio-demographic and economic factors on the children labour and violence against children in Egypt. Poverty neglects, lack of adequate care and exposures of children to various grades of violence are major drivers of child labour across the country. North-eastern region of Egypt has a higher likelihood of child labour than most other regions, while children who live in Delta are more engaged in hazard work.

epidemiology↗
bioRxiv2019

Using hospital network-based surveillance for antimicrobial resistance as a more robust alternative to self-reporting

Hospital performance is often measured using self-reported statistics, such as the incidence of hospital-transmitted micro-organisms or those exhibiting antimicrobial resistance (AMR), encouraging hospitals with high levels to improve their performance. However, hospitals that increase screening efforts will appear to have a higher incidence and perform poorly, undermining comparison between hospitals and disincentivising testing, thus hampering infection control. We propose a surveillance system in which hospitals test patients previously discharged from other hospitals and report observed cases. Using NHS Hospital Episode Statistics data, we analysed patient movements across England and assessed the number of hospitals required to participate in such a reporting scheme to deliver robust estimates of incidence. With over 1.2 million admissions to English hospitals previously discharged from other hospitals annually, even when only a fraction of hospitals (41/155) participate (each screening at least 1000 of these admissions), the proposed surveillance system can estimate incidence across all hospitals. By reporting on other hospitals, the reporting of incidence is separated from the task of improving own performance. Therefore the incentives for increasing performance can be aligned to increase (rather than decrease) screening efforts, thus delivering both more comparable figures on the AMR problems across hospitals and improving infection control efforts.

epidemiology↗