Introduction
The connection between diseases of the lungs and environmental conditions, such as air quality, is a crucial research question to improve disease prevention. Ecological studies are helpful in addressing such topics but involve the risks of the ecological fallacy or incorrect generalizations. The problem above could be investigated by designing a multi-group ecological study to explore correlations between lung disease prevalence and average Air Quality Index (AQI) scores at the state level.
Designing a Study to Research the Lung Disease and Air Pollution Problem
To design a study in order to explore the link between lung disease and air pollution, it would be possible to follow a four-step process started by identifying the level or unit of analysis. With the assigned problem, giving preference to the city/county level could produce results with limited generalizability, whereas analyzing the U.S. population as a whole would reduce the room for comparison. The level of the state would be preferable, forming over fifty groups to be compared.
Step two would center on identifying the suitable ecological study design among the available options. The three approaches prevalent in ecological research include mixed, time-trend, and multiple-group designs (Aerts et al., 2020; Ferreira et al., 2019). A time-trend design could be challenging to implement as it would require keeping track of lung disease and air pollution dynamics. It could be difficult to locate disorder prevalence data for periods of time shorter than a year for all populations involved in the analysis. A mixed design would involve time-focused and group-focused comparisons. With the selected unit of analysis, the state, this design would lead to extremely effort-intensive endeavors and overcomplicated analytical procedures. In contrast, a multiple-group design would be more suitable to proceed with the chosen unit of analysis and identify whether people in the U.S. states with higher air pollution rates are more likely to develop lung disorders. Thus, the hypothetical study would focus on inter-group comparisons, with each state’s population representing a separate group.
Within the frame of the next stage, the sources of trustworthy information on the hypothetical exposure (air pollution) and outcome (lung disease incidence) will be selected, and the outcome will be further specified. The AQI reported by AirNow (n.d.) measures air pollution levels on a scale from 0 to 500, where values exceeding 50 indicate air conditions that create health risks for pollution-sensitive populations. The AQI is calculated for each U.S. location and state individually with reference to five separate pollutants, including tropospheric ozone, particulate matter, nitrogen dioxide, sulfur dioxide, and carbon monoxide (AirNow, n.d.). The outcome could be presented by lung diseases of different etiology connected to air quality, for instance, asthma, lung cancer, and chronic obstructive pulmonary disease (COPD) (Lipfert & Wyzga, 2019; Mutlu & Peker, 2019). The Centers for Disease Control and Prevention is a reliable source of state-specific data on such conditions (CDC, 2022). Therefore, AQI values and disease prevalence data from the CDC would offer credible exposure and outcome data for separate states.
The final stage of the study planning process would revolve around specifying data sets to be analyzed with reference to correlations. State-specific AQI results may change relatively quickly due to atmospheric instability. Also, it is notable that COPD does not develop immediately after harmful exposures, which is also true for the other two selected conditions (Okui & Park, 2021). To account for these factors, it would be reasonable to measure the exposure component as the average AQI value for each state during one year, for instance, 2022. The suitable outcome data could then be derived from the CDC’s disease prevalence reports for the three conditions that would specify the prevalence rate for each condition in each state as of the end of 2022.
The Ecological Fallacy in the Specified Situation
The ecological inference fallacy might come into play as the researcher’s inability to identify individual cases in which the correlation does not exist. Firstly, if the correlation between pollution levels and lung disease prevalence is strong and positive, one will probably conclude that it is strong in each case. However, it is hypothetically possible that the most polluted states happen to have more citizens prone to lung disease for reasons unrelated to ecology. Secondly, state-specific disease prevalence data does not capture citizens’ inter-state and international migration. For instance, an individual might develop a lung condition after exposure to polluted air in a state with a poor environmental situation and then change the place of residence to seek treatment in another state. This person might eventually get diagnosed with a more serious condition at this second location. Another possible source of the fallacy is international migration from poorer countries. If more immigrants from countries with relatively poor air quality come to U.S. states with the best AQI results, this will artificially increase the state’s disease prevalence rate. Therefore, there can be different barriers to making correct assumptions based on non-individual data.
Conclusion
Finally, ecological research to explore how air pollution affects the exposed population’s respiratory health could be designed as a multi-group study with each U.S. state’s population as a separate group. The analysis of correlations between higher AQI scores and higher asthma, lung cancer, and COPD prevalence rates using data for separate U.S. states could help to add to previous research on the predictors of lung disease. Nevertheless, each state’s population might be heterogeneous in actual pollutant exposure and baseline health, which could give rise to incorrect inferences.
References
Aerts, R., Nemery, B., Bauwelinck, M., Trabelsi, S., Deboosere, P., Van Nieuwenhuyse, A., Nawrot, T. S., & Casas, L. (2020). Residential green space, air pollution, socioeconomic deprivation and cardiovascular medication sales in Belgium: A nationwide ecological study.Science of the Total Environment, 712, 136426. Web.
AirNow. (n.d.). National maps. Web.
Centers for Disease Control and Prevention. (2022). Most recent asthma state or territory data. Web.
Ferreira, L. D. C. M., Nogueira, M. C., Pereira, R. V. D. B., De Farias, W. C. M., Rodrigues, M. M. D. S., Teixeira, M. T. B., & Carvalho, M. S. (2019). Ambient temperature and mortality due to acute myocardial infarction in Brazil: An ecological study of time-series analyses.Scientific Reports, 9(1), 1-10. Web.
Lipfert, F. W., & Wyzga, R. E. (2019). Longitudinal relationships between lung cancer mortality rates, smoking, and ambient air quality: A comprehensive review and analysis.Critical Reviews in Toxicology, 49(9), 790-818. Web.
Mutlu, G. M., & Peker, Y. (2019). Air pollution, asthma, and sleep apnea: New epidemiological links?Annals of the American Thoracic Society, 16(3), 307-308. Web.
Okui, T., & Park, J. (2021). Geographical differences and their associated factors in chronic obstructive pulmonary disease mortality in Japan: An ecological study using nationwide data. International Journal of Environmental Research and Public Health, 18(24), 1-10. Web.