Monitoring the decreasing trend of testicular cancer mortality in Spain during 2005–2019 through a Bayesian approach

Abstract 

Purpose: To assess time trends of testicular cancer (TC) mortality in Spain for period 1985–2019 for age groups 15–74 years old through a Bayesian age–period–cohort (APC) analysis. Methods: A Bayesian age–drift model has been fitted to describe trends. Projections for 2005–2019 have been calculated by means of an autoregressive APC model. Prior precision for these parameters has been selected through evaluation of an adaptive precision parameter and 95% credible intervals (95% CRI) have been obtained for each model parameter. Results: A decrease of −2.41% (95% CRI: −3.65%; −1.13%) per year has been found for TC mortality rates in age groups 15–74 during 1985–2004, whereas mortality showed a lower annual decrease when data was restricted to age groups 15–54 (−1.18%; 95% CRI: −2.60%; −0.31%). During 2005–2019 is expected a decrease of TC mortality of 2.30% per year for men younger than 35, whereas a leveling off for TC mortality rates is expected for men older than 35. Conclusions: A Bayesian approach should be recommended to describe and project time trends for those diseases with low number of cases. Through this model it has been assessed that management of TC and advances in therapy led to decreasing trend of TC mortality during the period 1985–2004, whereas a leveling off for these trends can be considered during 2005–2019 among men older than 35.

Keywords: Testicular cancer, Mortality, Time trends, Bayesian, Projections