Mediation Analysis to Investigate Differences in Prostate Cancer Diagnosis Stage Through Environmental Risk Factors
Nubaira Rizvi1, Randy Hamilton1, Xiao-Cheng Wu2
1Biostatistics and Data Science, School of Public Health, LSU Health-New Orleans, New Orleans, LA 70112, USA.
None:
Prostate Cancer (PCa) is the most commonly diagnosed cancer and the second leading cause of cancer death among men. In Louisiana (LA), Black men are disproportionately diagnosed at later stages compared to White men. This study explores environmental risk factors as potential intermediate variables linking race to cancer diagnosis stage. The Louisiana Tumor Registry data included 24,647 male patients diagnosed with PCa in LA between 2010 and 2018. Among them, 15,875 (64.40%) were Caucasian American (CA) and 8772 (35.59%) African American (AA). Mediation analysis using multiple additive regression trees (MART) identified possible intermediate variables that potentially explain the observed disparity. The study found that individual characteristics and environmental factors jointly explained 84% (95% CI: 44.1%, 94.6%) and 18.6% (95% CI: 7.3%, 53.7%) of the observed racial disparity in PCa stage at diagnosis, respectively. Individual factors included BMI (35.9%), marital status (28.5%), CDI (8.2%), female-headed households (2.3%), comorbidity (3.9%), and insurance status (6.3%). Environmental contributors included cancer risk due to air toxicity exposure (7.2%), asthma prevalence (6.6%), acetaldehyde levels (2.1%), railroad proximity (2.1%), walkability (0.3%), and ozone level (-0.1%). Environmental factors jointly played a significant role in the observed racial disparity. The factors such as air toxicity, acetaldehyde levels, and asthma prevalence highlight the need to address industrial pollutants to reduce the differences.
Related Concept Videos
Hazard Ratio
For example, in a clinical trial...
Statistical Methods for Analyzing Epidemiological Data
Comparing the Survival Analysis of Two or More Groups
Cancer Survival Analysis


