Abstract / Summary
Background: Gallstone disease and pancreatitis are clinically linked, yet population-level estimates for GBD-defined gallbladder and biliary diseases are not a direct surrogate for gallstone pancreatitis risk. The global architecture of co-occurrence and epidemiological discordance between GBD-defined biliary disease and pancreatitis, and the exposure profiles characterizing discordant settings, remain insufficiently defined. Methods: Using Global Burden of Disease (GBD) 2023 data for 204 countries and territories, we mapped 2023 age-standardized incidence rates (ASIRs) of GBD-defined biliary disease and pancreatitis and classified countries into four relative co-occurrence typologies: Synchronized High Burden, Synchronized Low Burden, Biliary Retention, and Pancreatitis Escape. These categories are distribution-derived global burden strata, not clinically validated risk strata. Classification stability was evaluated through Monte Carlo uncertainty propagation, yielding country-specific membership probabilities and maximum probability (maxP). Discordance was further quantified using a continuous distance metric (D), and a Strong Escape subset was identified as the upper quartile within Pancreatitis Escape (Dmedian ≥ 0.988). Exposure profiles were characterized by standardized deviations of summary exposure values (SEVs) across candidate exposure indicators. Secondary analyses examined exposures associated with Synchronized High Burden, model-based exposure prediction contrasts, and future cases under five official IHME demographic scenarios. Results: Synchronized High Burden clustered in North America, Australia, Western-Central Europe, and parts of Latin America, whereas Pancreatitis Escape concentrated in Russia, South Asia, the Middle East, and parts of Southeast Asia. Typology assignments were highly stable under uncertainty propagation. Strong Escape settings exhibited marked heterogeneity in exposure profiles, including predominant smokeless tobacco exposure in South Asia, prominent processed meat consumption and smoking signals in Russia, and elevated unsafe water exposure in Nigeria. These findings are interpreted as hypothesis-generating ecological patterns rather than evidence of individual-level causality. In Synchronized High Burden settings, higher low-density lipoprotein cholesterol and processed meat intake were consistently associated with higher ASIRs of both conditions. Model-based exposure shifts yielded modest prediction contrasts that were similar in an observed joint-donor sensitivity analysis, while decomposition and explicit demographic-scenario analyses indicated that absolute case counts can increase despite stable or declining rates. Conclusion: A reproducible global co-occurrence and discordance structure links GBD-defined biliary disease and pancreatitis and highlights a descriptive Pancreatitis Escape pattern with region-specific exposure profiles. These findings support geographically tailored hypothesis generation regarding non-biliary exposure correlates of pancreatitis and show how alternative demographic trajectories can change projected healthcare demand without establishing a causal effect of population change.