Introduction: Certain metabolic parameters increase the risk of esophageal cancer. This study investigated the association between the variability in metabolic parameters and esophageal cancer incidence using large nationally representative data. Methods: Using the health checkup and claims data provided by the Korean National Health Insurance Service (NHIS), we included 8,376,233 subjects who underwent NHIS-provided health checkups between 2009 and 2010 (index year) and two or more health checkups within five years before the index year. Hazard ratios (HRs) and 95% confidence intervals (CIs) for esophageal cancer were obtained using Cox proportional hazards models according to the quartiles of variability of each metabolic parameter: fasting blood glucose (FBG), weight, systolic blood pressure (SBP), and total cholesterol (TC) as well as a cumulative number of high-variability parameters. Results: A total of 6,455 cases of esophageal cancer occurred during a mean (±SD) follow-up of 8.8 (±1.1) years. The following metabolic parameters were used, with an adjusted HR and 95% CI: FBG (1.11, 1.03-1.18), weight (1.15, 1.07-1.23), SBP (1.08, 1.01-1.16), and TC (1.23, 1.15-1.32). The risk of esophageal cancer was higher in the highest quartile of variability than the lower quartiles. The risk of esophageal cancer gradually increased with a greater number of high-variability parameters: 1.08 (1.02-1.15), 1.22 (1.14-1.31), and 1.33 (1.21-1.46) for 1, 2, and 3-4 high-variability parameters (vs. none). Conclusions: A high variability of metabolic parameters was associated with an increased esophageal cancer risk. Further studies are needed to replicate our findings in other populations.
Keywords: claims data; esophageal cancer; metabolic parameter; nationwide study; variability independent of the mean.