Purpose: This study constructs a lung cancer risk index (LCRI) that incorporates many modifiable risk factors using an easily reproducible and adaptable method that relies on publicly available data.
Methods: We used meta-analysis followed by Analytic Hierarchy Process (AHP) to generate a lung cancer risk index (LCRI) that incorporates seven modifiable risk factors (active smoking, indoor air pollution, occupational exposure, alcohol consumption, secondhand smoke exposure, outdoor air pollution, and radon exposure) for lung cancer. Using county-level population data, we then performed a case study in which we tailored the LCRI for use in the state of Illinois (LCRIIL).
Results: For both the LCRI and the LCRIIL, active smoking had the highest weights (46.1% and 70%, respectively), whereas radon had the lowest weights (3.0% and 5.7%, respectively). The weights for alcohol consumption were 7.8% and 14.7% for the LCRI and the LCRIIL, respectively, and were 3.8% and 0.95% for outdoor air pollution. Three variables were only included in the LCRI: indoor air pollution (18.5%), occupational exposure (13.2%), and secondhand smoke exposure (7.6%). The Consistency Ratio (CR) was well below the 0.1 cut point. The LCRIIL was moderate though significantly correlated with age-adjusted lung cancer incidence (r = 0.449, P < 0.05) and mortality rates (r = 0.495, P < 0.05).
Conclusion: This study presents an index that incorporates multiple modifiable risk factors for lung cancer into one composite score. Since the LCRI allows data comprising the composite score to vary based on the location of interest, this measurement tool can be used for any geographic location where population-based data for individual risk factors exist. Researchers, policymakers, and public health professionals may utilize this framework to determine areas that are most in need of lung cancer-related interventions and resources.
Keywords: Analytic hierarchy processes; Lung cancer; Meta-analysis; Risk factors; Risk index.
© 2022. The Author(s).