BACKGROUND: The introduction of non-invasive imaging techniques on a large scale may falsely suggest an increase in incidence of pancreatic cancer. In the present study, it was investigated whether the new diagnostic techniques have improved the documentation of deaths due to pancreatic cancer and thus have an influence on the trend of mortality. MATERIALS AND METHODS: Age-standardized mortality rates for men and women were calculated covering the period 1970 to 1993. Using Spearman's correlation coefficient, the trends during 1970 to 1993 as well as during 1970 to 1978 and 1981 to 1993 were analyzed for statistical significance. Furthermore, the development of mortality during 1973 to 1993 was analyzed in 5-year intervals for various birth cohorts and age groups. RESULTS: Age-standardized mortality rates have increased significantly since 1970, i.e. by 17% in men (p = 0.0021) and by 20% in women (p = 0.001). The observed trend is not linear but characterized by a step-like rise in 1979/1980, coinciding with the introduction of non-invasive imaging techniques on a large scale. CONCLUSIONS: The risk to develop cancer of the pancreas has not significantly changed since 1970. Rather, the documentation of pancreatic cancer on death certificates has improved as a result of the higher diagnostic accuracy, which explains the observed step-like rise in death rates.
BACKGROUND: The introduction of non-invasive imaging techniques on a large scale may falsely suggest an increase in incidence of pancreatic cancer. In the present study, it was investigated whether the new diagnostic techniques have improved the documentation of deaths due to pancreatic cancer and thus have an influence on the trend of mortality. MATERIALS AND METHODS: Age-standardized mortality rates for men and women were calculated covering the period 1970 to 1993. Using Spearman's correlation coefficient, the trends during 1970 to 1993 as well as during 1970 to 1978 and 1981 to 1993 were analyzed for statistical significance. Furthermore, the development of mortality during 1973 to 1993 was analyzed in 5-year intervals for various birth cohorts and age groups. RESULTS: Age-standardized mortality rates have increased significantly since 1970, i.e. by 17% in men (p = 0.0021) and by 20% in women (p = 0.001). The observed trend is not linear but characterized by a step-like rise in 1979/1980, coinciding with the introduction of non-invasive imaging techniques on a large scale. CONCLUSIONS: The risk to develop cancer of the pancreas has not significantly changed since 1970. Rather, the documentation of pancreatic cancer on death certificates has improved as a result of the higher diagnostic accuracy, which explains the observed step-like rise in death rates.