Cancer mortality projection: disparities, COVID-19, and late diagnosis impact

TL;DR

This study uses Bayesian models to project future cancer mortality in England, highlighting persistent disparities, regional variations, and the impact of COVID-19-related diagnosis delays on lung and breast cancer deaths.

Contribution

It introduces a Bayesian framework for projecting cancer mortality considering socio-economic and regional factors, and assesses COVID-19 pandemic effects and diagnosis delays.

Findings

Socio-economic disparities in lung cancer mortality will persist.

Regional variations in breast cancer mortality are observed up to 2036.

Diagnosis delays during COVID-19 led to increased excess deaths, especially in northern and deprived areas.

Abstract

This paper investigates projection of two major causes of cancer mortality, breast cancer and lung cancer, by using a Bayesian modelling framework. We investigate patterns in 2001-2018 (as baseline) in cause-specific cancer mortality and project these by year of death and various risk factors: age, gender, regions of England, income deprivation quintile, average age-at-diagnosis, and non-smoker prevalence rates. We then assess excess cancer mortality during the COVID-19 pandemic years, and we examine the impact of diagnosis delays on lung cancer mortality across various scenarios. Our findings indicate that socio-economic disparities in lung cancer mortality will persist in the future. Additionally, we observe slight variations in breast cancer mortality across different regions up to 2036. Furthermore, marginal increases in excess deaths from lung and breast cancer are estimated in specific regions of England throughout the pandemic year (2020-2022), contrasting with the national trend. However, the excess lung cancer deaths markedly differ by age, region and deprivation as a result of delays in cancer diagnosis. Specifically, we find a notably higher number of excess deaths in the northern regions of England compared to the southern regions, as well as among individuals living in the most deprived areas compared to those in the least deprived areas.