Pancreatic cancer incidence and mortality trends in urban Shanghai, China from 1973 to 2017: a joinpoint regression and age-period-cohort analysis

Background and purposeTo provide a comprehensive overview of epidemiological features and temporal trends of pancreatic cancer in urban Shanghai from 1973 to 2017. MethodsData on pancreatic cancer in urban Shanghai were obtained through the Shanghai Cancer Registry and the Vital Statistics System. Joinpoint analysis was used to describe the temporal trends and annual percent changes (APCs) and age-period-cohort analysis were used to estimate the effects of age, period, and birth cohort on pancreatic cancer. ResultsThere were a total of 29,253 cases and 27,105 deaths of pancreatic cancer in urban Shanghai over the 45-year study period. The overall average annual age-standardized incidence and mortality rates were 5.45/100,000 and 5.02/100,000, respectively. Both the incidence and mortality rates demonstrated fluctuating upward trends, with an average annual increase rate of 1.51% (APC = 1.51, P < 0.001) and 1.04% (APC = 1.04, P < 0.001), respectively. The upward trend in incidence was greater for females than for males, while the trend in mortality was seen in both sexes equally and continuously. In recent years (2013-2017), the age-specific incidence rates increased further than before, with statistically significant changes in the 35-year, 45- to 55-year and 70- to 85-year age groups (P < 0.05). The age-specific mortality rates also showed obvious upward trends, which in the 50- to 55-year, and 75- to 85-year age groups increased significantly. The results of the age-period-cohort analysis suggested significant effects of age, period, and cohort on the prevalence of pancreatic cancer. ConclusionThe prevalence of pancreatic cancer, dramatically influenced by socioeconomic development and lifestyles, demonstrated a significant upward trend from 1973 to 2017 in urban Shanghai and underscored the necessity and urgency for additional efforts in primary and secondary prevention measures.