78853 - Environmental Epidemiology

Learning outcomes

At the end of this course students know how to handle statistical methods and models to study the relationship between health and exposure to environmental stressors. In particular, students will learn about: geostatistics for spatial predictions of pollutants; disease mapping in presence of small areas; poisson regression models for counts of disease to study the link between pollution and health; the most popular R packages to analyse spatial data.

Course contents

Module I

Atmospheric pollutants. Correlation between time series of pollutants.

Literature review on pollutant effects on health. Short-term effects versus long-term effects. Response-dose relationships. Health hazards associated to waste management systems.

Assessing exposure to atmospheric pollutants. Direct methods versus indirect methods. Statistical models to assess exposure.

Impact on health evaluation. Quantify risks and impacts. Attributable fractions and conterfactual scenarios. Impact indicators (AC, YLL, YLD, DALYs). Exposure-response functions quantifying impact of particulate matter on mortality.

 

Module II

Statistics for environmental epidemiology. Spatial data. Mapping areal risks, rates and proportions.

Areal data analysis for environmental epidemiology applications. Standardized mortality rates. Internal standardization. Poisson regression for counts of disease. Disease mapping. Smoothing methods: local mean, empirical Bayes. Global clustering of disease risks; Moran index.

Geostatistics tools useful in environmental epidemiology. Stochastic spatial processes. Covariogram. Matern covariance functions. Spatial prediction.

Ecological regression models for studying the relationship between pollution and health.

Readings/Bibliography

Module I:

• Epidemiologia ambientale: Metodi di studio e applicazioni in sanità pubblica. Dean Baker , Fabio Barbone , Rebecca Calderon , Tord Kjellstrom , Harris Pastides (2004). Edizione italiana del testo: Environmental Epidemiology A Textbook on Study Methods and Public Health Applications. WHO-USEPA (WHO/SPE/OEH/99.7). Disponibile all’indirizzo: http://www.arpat.toscana.it/documentazione/catalogo-pubblicazioni-arpat/epidemiologia-ambientale?searchterm=epidemiologia+ambientale
• D. Baker, M.J. Nieuwenhuijsen(eds). Environmental Epidemiology: study methods and application, Oxford University Press, 2008.

Module II:

• Applied Spatial Statistics for Public Health Data (2004). Lance A. Waller, Carol A. Gotway. Wiley
• Peter Diggle, Paulo Ribeiro (2007). "Model-based Geostatistics". Springer

Teaching methods

Frontal lectures; LAB tutorial with RStudio

Assessment methods

Assessment methods

The examination consists of two independent assessments corresponding to Modules 1 and 2.

Module 1 (Cycle 3, 30 hours)

Students will be assessed through a take-home assignment to be completed using Excel.

The assignment will be made available through the Virtuale platform before the end of Module 1. Students must submit their report, also through Virtuale, by the specified deadline (normally before the first summer examination session). Further information regarding the length of the report, submission format, deadlines, and submission procedures will be provided during the course.

Note: Students intending to take any examination session after the first summer session should contact the instructor for Module 1 (Andrea Ranzi) by email to request the take-home assignment, copying the instructor for Module 2 (Massimo Ventrucci). The instructors will communicate the submission deadlines. In any case, the report must be submitted no later than the date of the examination session.

Module 2 (Cycle 4, 30 hours)

Students will be assessed on the basis of:

• a practical assignment (take-home assignment) to be completed using R and normally submitted by the end of the course;
• a written EOL exam, to be taken during one of the official examination sessions (the first available session coincides with the midterm exam).

The final grade for Module 2 is the average of these two assessments.

The written EOL exam is designed to assess students' theoretical knowledge and consists of multiple-choice questions, open-ended questions, and some questions requiring the use of R.

The take-home assignment is designed to assess practical skills. It consists of analysing a dataset using R and preparing a report—in the form of a Word, PDF, or RMarkdown document including both R code and comments—which must be submitted through the Virtuale platform by the specified deadline. Students will work in groups of up to three. The assignment will be distributed during the last week of teaching (the final 6 hours of Module 2). Students will begin working on it during class and complete it at home. Further details on the required analyses and the submission procedure through the Virtuale platform will be provided during the course.

Note: Students who are unable to complete or pass the take-home assignment (because they are not attending the course, are absent during the final week of Module 2, or fail to submit the assignment by the deadline) will be required to complete an additional practical section as part of the written EOL exam. This additional section is intended to assess the same practical skills covered by the take-home assignment.

AI

With regard to the computer-based quiz used for the assessment of learning, the use of AI tools is prohibited. Any use of such tools constitutes a violation of academic integrity.

For the take-home assignments, limited, declared, and non-substantial use of AI tools is permitted for support activities (e.g., summarization and rephrasing). Students must indicate which exercises involved the use of AI support.

Grading

The assessment process aims to evaluate both theoretical knowledge (the ability to describe the concepts and analytical methods covered during the course) and practical skills (the ability to apply the analytical tools correctly and interpret the results appropriately).

The overall evaluation is based on the instructors' assessment of four aspects:

1. the appropriateness of the analyses carried out in relation to the questions posed in the practical assignment (take-home assignment);
2. the clarity of the presentation and discussion of the analyses;
3. the technical correctness of the analyses;
4. the ability to correctly interpret the results.

To pass the course, students must obtain a grade of at least 18/30 in both modules. The final course grade is the average of the grades obtained in the two modules.

 

Teaching tools

All material will be provided through the virtuale platform (login in to https://virtuale.unibo.it/). There you will find slides, lab tutorial in pdf, further readings and material discussed in class. I suggest students who have a laptop to bring it in class. Softwares used in this course are all open source: we will use R (http://www.r-project.org/) and RStudio (https://rstudio.com/).

Office hours

See the website of Massimo Ventrucci

See the website of