Non-Technical Explanation of Domain Modeling

[salihakalender]

🧠 Non-Technical Explanation of Domain Modeling

In this project, we studied how air pollution affects human health across different countries. To explore this, we grouped countries based on which air pollutant—such as PM2.5, NO₂, or O₃—was most dominant over a five-year period (2015–2019).

This grouping made it easier to compare health outcomes—like asthma, heart disease, or stroke—between regions with different pollution profiles. Our goal is to understand whether some types of pollutants are more strongly linked to specific health burdens than others.

We chose this approach because it simplifies a large and messy dataset into meaningful categories, allowing us to explore patterns across countries. While data limitations—such as missing or inconsistent records—prevented us from including every country, we selected 7–8 countries with complete, reliable data to begin our analysis.

Later in the project, we also plan to explore whether long-term exposure to specific pollutants might have contributed to higher COVID-19 death rates in some countries, especially where pollution may have weakened lung and heart health before the pandemic began.

To add further context, we are considering the Socio-demographic Index (SDI) of each country. This allows us to see whether development level may influence the relationship between pollution and health outcomes—since access to healthcare, infrastructure, and policy responses vary widely around the world.

Ultimately, this work aims to go beyond numbers—to highlight the real-world impact of air pollution on public health. By identifying which pollutants dominate in different regions and how they relate to specific health outcomes, we hope to support more targeted, evidence-based policies. Clean air is not just an environmental issue—it’s a public health priority that affects millions of lives every day.

[SiSaR-Pal]

Milestone 2: Data Collection – Non-Technical Explanation of Domain Modeling Draft

Project Overview

In this project, we studied how air pollution affects human health across different countries. To explore this, we grouped countries based on which air pollutant such as PM₂.₅, NO₂, or O₃ was most dominant over a five-year period (2015–2019).

This grouping made it easier to compare health outcomes like asthma, heart disease, or stroke between regions with different pollution profiles. Our goal is to understand whether some types of pollutants are more strongly linked to specific health burdens than others.

We chose this approach because it simplifies a large and complex dataset into meaningful categories, allowing us to explore patterns across countries. While data limitations such as missing or inconsistent records prevented us from including every country, we selected 7–8 countries with complete, reliable data to begin our analysis.

To add further context, we are considering the Socio-demographic Index (SDI) of each country. This allows us to see whether development level may influence the relationship between pollution and health outcomes, since access to healthcare, infrastructure, and policy responses vary widely around the world.

Later in the project, we also plan to explore the correlation between heatwaves and air quality and pollution.

---

Data Sources and Modeling Choices

• Air Quality Data: We obtained robust data for air pollution levels (PM₂.₅, NO₂, O₃) from international databases, focusing on years 2015–2019.
• Health Outcome Data: Data on health impacts (asthma, heart disease, stroke) was much harder to obtain. Issues included missing values, inconsistent reporting standards, and outdated records. We prioritized countries with the most complete and recent data.
• Socio-demographic Index (SDI): We included SDI to provide socioeconomic context for each country, recognizing that development level can affect both exposure and health outcomes.

---

Limitations and Challenges

• Incomplete Health Data: Many countries lacked up-to-date or complete health outcome records, limiting our sample size and possibly affecting the representativeness of our findings.
• Temporal Mismatches: Some health data did not align perfectly with our pollution data timeframe, making it harder to draw causal links.
• Diagnostic Variability: Differences in how countries diagnose and report health conditions (like asthma or stroke) introduce comparability challenges.
• Data Gaps for Heatwaves: Data linking heatwaves and air quality is not yet integrated; this will be addressed in a later phase.

---

Rationale and Future Directions

By grouping countries according to dominant pollutants and comparing health outcomes within these groups, we hope to uncover whether certain pollutants are more strongly associated with specific health burdens. Including SDI helps us account for the role of socioeconomic factors in shaping these relationships.

Our approach balances the need for meaningful comparisons with the realities of real-world data limitations. In the next phase, we plan to expand our analysis to include heatwave data and, as more data becomes available, increase the number of countries in our study.

Ultimately, our work aims to highlight the real-world impact of air pollution on public health and support more targeted, evidence-based policies. Clean air is not just an environmental issue it’s a public health priority that affects millions of lives every day.

[salihakalender]

Thank you, Said, for your contributions. Based on the data, we have redirected our project, and I’ll revise the non-technical explanation using some of the input you shared as well.

[salihakalender]

I've revised the non-technical explanation of our domain modeling to reflect the dataset we agreed on.

🧠 Non-Technical Explanation of Domain Modeling

In this project, we are studying how long-term air pollution—especially PM2.5—affects human health across different countries. We decided to use a country-level PM2.5 concentration dataset along with respiratory and cardiovascular disease burden data from the Global Burden of Disease (GBD) study.

For now, we are focusing on 25 diverse countries from different regions and development levels, based on the availability of reliable data. We plan to look at trends in PM2.5 levels over time and see how they relate to health outcomes like asthma, heart disease, and stroke. Later, we may expand the analysis to include more countries for broader comparisons and stronger statistical results.

Our approach allows us to explore the lag effect—the idea that long-term exposure to air pollution can lead to health problems years later. This helps us understand not just what’s happening now, but what pollution over the last decade might have caused.

As an additional layer, we’ll also look at whether countries with higher long-term air pollution exposure experienced higher COVID-19 death rates, possibly because pollution had already weakened lung or heart health before the pandemic.

We are also considering the Socio-demographic Index (SDI) of each country. This helps us explore how development level—like access to healthcare and public policy—might affect the link between pollution and health.

In short, our goal is to turn complex environmental and health data into clear insights. By finding patterns across countries, we hope this project can support more targeted and effective public health policies. Clean air is not just about the environment—it’s about protecting people’s lives.