Search

We propose a methodology to quantify the impact on breast cancer mortality of diagnostic delays caused by public health measures introduced as a response to the COVID-19 pandemic. These measures affected cancer pathways by halting cancer screening, delaying diagnostic tests, and reducing the numbers of patients starting treatment. We introduce a semi-Markov model, to quantify the impact of the pandemic based on publicly available population data for women age 65{89 years in England and relevant medical literature. We quantify age-specific excess deaths, for a period up to 5 years, along with years of life expectancy lost and change in cancer mortality by cancer stage. Our analysis suggests a 3-6% increase in breast cancer deaths, corresponding to more than 40 extra deaths, per 100,000 women, after age 65 years old over 5 years, and a 4-6% increase in registrations of advanced (Stage 4) breast cancer. Our modelling approach exhibits consistent results in sensitivity analyses, providing a model that can account for changes in breast cancer diagnostic and treatment services.

In recent years, improvements in all-cause mortality rates and life expectancies for males and females in England and Wales have slowed down. In this paper, cause-specific mortality data for England and Wales from 2001 to 2018 are used to investigate the cause-specific contributions to the slowdown in improvements. Cause-specific death counts in England and Wales are modelled using negative binomial regression and a breakpoint in the linear temporal trend in log mortality rates is investigated. Cause-specific scenarios are generated, where the post-breakpoint temporal trends for certain causes are reverted to pre-breakpoint rates and the effect of these changes on age-standardised mortality rates and period life expectancies is explored. These scenarios are used to quantify cause-specific contributions to the mortality improvement slowdown. Reductions in improvements at older ages in circulatory system diseases, as well as the worsening of mortality rates due to mental and behavioural disorders and nervous system diseases, provide the greatest contributions to the reduction of improvements in age-standardised mortality rates and period life expectancies. Future period life expectancies scenarios are also generated, where cause-specific mortality rate trends are assumed to either persist or be reverted. In the majority of scenarios, the reversion of cause-specific mortality trends in a single cause of death results in the worsening of period life expectancies at birth and age 65 for both males and females. This work enhances the understanding of cause-specific contributions to the slowdown in all-cause mortality rate improvements from 2001 to 2018, while also providing insights into causes of death that are drivers of life expectancy improvements. The findings can be of benefit to researchers, policy-makers and insurance professionals.

Parameter estimation and associated uncertainty quantification is an important problem in dynamical systems characterized by ordinary differential equation (ODE) models that are often nonlinear. Typically, such models have analytically intractable trajectories which result in likelihoods and posterior distributions that are similarly intractable. Bayesian inference for ODE systems via simulation methods require numerical approximations to produce inference with high accuracy at a cost of heavy computational power and slow convergence. At the same time, Artificial Neural Networks (ANN) offer tractability that can be utilized to construct an approximate but tractable likelihood and posterior distribution. In this paper we propose a hybrid approach, where Laplace-based Bayesian inference is combined with an ANN architecture for obtaining approximations to the ODE trajectories as a function of the unknown initial values and system parameters. Suitable choices of a collocation grid and customized loss functions are proposed to fine tune the ODE trajectories and Laplace approximation. The effectiveness of our proposed methods is demonstrated using an epidemiological system with non-analytical solutions, the Susceptible-Infectious-Removed (SIR) model for infectious diseases, based on simulated and real-life influenza datasets. The novelty and attractiveness of our proposed approach include (i) a new development of Bayesian inference using ANN architectures for ODE based dynamical systems, and (ii) a computationally fast posterior inference by avoiding convergence issues of benchmark Markov Chain Monte Carlo methods. These two features establish the developed approach as an accurate alternative to traditional Bayesian computational methods, with improved computational cost., Comment: 39 pages, 9 figures, 7 tables

One of the most important issues in the critical assessment of spatio-temporal stochastic models for epidemics is the selection of the transmission kernel used to represent the relationship between infectious challenge and spatial separation of infected and susceptible hosts. As the design of control strategies is often based on an assessment of the distance over which transmission can realistically occur and estimation of this distance is very sensitive to the choice of kernel function, it is important that models used to inform control strategies can be scrutinised in the light of observation in order to elicit possible evidence against the selected kernel function. While a range of approaches to model criticism are in existence, the field remains one in which the need for further research is recognised. In this paper, building on earlier contributions by the authors, we introduce a new approach to assessing the validity of spatial kernels - the latent likelihood ratio tests - and compare its capacity to detect model misspecification with that of tests based on the use of infection-link residuals. We demonstrate that the new approach, which combines Bayesian and frequentist ideas by treating the statistical decision maker as a complex entity, can be used to formulate tests with greater power than infection-link residuals to detect kernel misspecification particularly when the degree of misspecification is modest. This new approach avoids the use of a fully Bayesian approach which may introduce undesirable complications related to computational complexity and prior sensitivity.