Spatial Clustering and Local Risk Factors of Chronic Obstructive Pulmonary Disease (COPD)

Chronic obstructive pulmonary disease (COPD) mortality has been steadily increasing in Taiwan since 2009. In order to understand where the hotspot areas are and what the local risk factors are, we integrated an ecological and a case-control study. We used a two-stage approach to identify hotspots and explore the possible risk factors for developing COPD. The first stage used the annual township COPD mortality from 2000 to 2012 and applied the retrospective space-time scan statistic to calculate the local relative risks in each township. In the second stage, we conducted a case-control study, recruiting 200 patients from one local hospital within the one identified hotspot area located in southern Taiwan. Logistic regression was applied for analyzing the personal risk factors of COPD. The univariate analyses showed that higher percentages of aborigines, patients with tuberculosis (TB) history, and those with smoking history had COPD (p < 0.05). After controlling for demographic variables, aboriginal status (adjusted odds ratios (AORs): 3.01, 95% CI: 1.52-5.93) and smoking history (AORs: 2.64, 95% CI: 1.46-4.76) were still the two significant risk factors. This two-stage approach might be beneficial to examine and cross-validate the findings from an aggregate to an individual scale, and can be easily extended to other chronic diseases.