Projections of cancer mortality risks using spatio-temporal P-spline models

A scalable approach for short-term disease forecasting in high spatial resolution areal data

Short-term disease forecasting at specific discrete spatial resolutions has become a high-impact decision-support tool in health planning. However, when the number of areas is very large obtaining predictions can be computationally intensive or even unfeasible using standard spatiotemporal models. The purpose of this paper is to provide a method for short-term predictions in high-dimensional areal data based on a newly proposed "divide-and-conquer" approach. We assess the predictive performance of this method and other classical spatiotemporal models in a validation study that uses cancer mortality data for the 7907 municipalities of continental Spain. The new proposal outperforms traditional models in terms of mean absolute error, root mean square error, and interval score when forecasting cancer mortality 1, 2, and 3 years ahead. Models are implemented in a fully Bayesian framework using the well-known integrated nested Laplace estimation technique.

Using mortality to predict incidence for rare and lethal cancers in very small areas

Incidence and mortality figures are needed to get a comprehensive overview of cancer burden. In many countries, cancer mortality figures are routinely recorded by statistical offices, whereas incidence depends on regional cancer registries. However, due to the complexity of updating cancer registries, incidence numbers become available 3 or 4 years later than mortality figures. It is, therefore, necessary to develop reliable procedures to predict cancer incidence at least until the period when mortality data are available. Most of the methods proposed in the literature are designed to predict total cancer (except nonmelanoma skin cancer) or major cancer sites. However, less frequent lethal cancers, such as brain cancer, are generally excluded from predictions because the scarce number of cases makes it difficult to use univariate models. Our proposal comes to fill this gap and consists of modeling jointly incidence and mortality data using spatio-temporal models with spatial and age shared components. This approach allows for predicting lethal cancers improving the performance of individual models when data are scarce by taking advantage of the high correlation between incidence and mortality. A fully Bayesian approach based on integrated nested Laplace approximations is considered for model fitting and inference. A validation process is also conducted to assess the performance of alternative models. We use the new proposals to predict brain cancer incidence rates by gender and age groups in the health units of Navarre and Basque Country (Spain) during the period 2005-2008.

Multidimensional penalized splines for incidence and mortality-trend analyses and validation of national cancer-incidence estimates

BACKGROUND

Cancer-incidence and mortality-trend analyses require appropriate statistical modelling. In countries without a nationwide cancer registry, an additional issue is estimating national incidence from local-registry data. The objectives of this study were to (i) promote the use of multidimensional penalized splines (MPS) for trend analyses; (ii) estimate the national cancer-incidence trends, using MPS, from only local-registry data; and (iii) propose a validation process of these estimates.

METHODS

We used an MPS model of age and year for trend analyses in France over 1990-2015 with a projection up to 2018. Validation was performed for 22 cancer sites and relied essentially on comparison with reference estimates that used the incidence/health-care ratio over the period 2011-2015. Alternative estimates that used the incidence/mortality ratio were also used to validate the trends.

RESULTS

In the validation assessment, the relative differences of the incidence estimates (2011-2015) with the reference estimates were <5% except for testis cancer in men and < 7% except for larynx cancer in women. Trends could be correctly derived since 1990 despite incomplete histories in some registries. The proposed method was applied to estimate the incidence and mortality trends of female lung cancer and prostate cancer in France.

CONCLUSIONS

The validation process confirmed the validity of the national French estimates; it may be applied in other countries to help in choosing the most appropriate national estimation method according to country-specific contexts. MPS form a powerful statistical tool for trend analyses; they allow trends to vary smoothly with age and are suitable for modelling simple as well as complex trends thanks to penalization. Detailed trend analyses of lung and prostate cancers illustrated the suitability of MPS and the epidemiological interest of such analyses.

A systematic review of Bayesian spatial-temporal models on cancer incidence and mortality

OBJECTIVES

This study aimed to review the types and applications of fully Bayesian (FB) spatial-temporal models and covariates used to study cancer incidence and mortality.

METHODS

This systematic review searched articles published within Medline, Embase, Web-of-Science and Google Scholar between 2014 and 2018.

RESULTS

A total of 38 studies were included in our study. All studies applied Bayesian spatial-temporal models to explore spatial patterns over time, and over half assessed the association with risk factors. Studies used different modelling approaches and prior distributions for spatial, temporal and spatial-temporal interaction effects depending on the nature of data, outcomes and applications. The most common Bayesian spatial-temporal model was a generalized linear mixed model. These models adjusted for covariates at the patient, area or temporal level, and through standardization.

CONCLUSIONS

Few studies (4) modelled patient-level clinical characteristics (11%), and the applications of an FB approach in the forecasting of spatial-temporally aligned cancer data were limited. This review highlighted the need for Bayesian spatial-temporal models to incorporate patient-level prognostic characteristics through the multi-level framework and forecast future cancer incidence and outcomes for cancer prevention and control strategies.

Advances in spatiotemporal models for non-communicable disease surveillance

Surveillance systems are commonly used to provide early warning detection or to assess an impact of an intervention/policy. Traditionally, the methodological and conceptual frameworks for surveillance have been designed for infectious diseases, but the rising burden of non-communicable diseases (NCDs) worldwide suggests a pressing need for surveillance strategies to detect unusual patterns in the data and to help unveil important risk factors in this setting. Surveillance methods need to be able to detect meaningful departures from expectation and exploit dependencies within such data to produce unbiased estimates of risk as well as future forecasts. This has led to the increasing development of a range of space-time methods specifically designed for NCD surveillance. We present an overview of recent advances in spatiotemporal disease surveillance for NCDs, using hierarchically specified models. This provides a coherent framework for modelling complex data structures, dealing with data sparsity, exploiting dependencies between data sources and propagating the inherent uncertainties present in both the data and the modelling process. We then focus on three commonly used models within the Bayesian Hierarchical Model (BHM) framework and, through a simulation study, we compare their performance. We also discuss some challenges faced by researchers when dealing with NCD surveillance, including how to account for false detection and the modifiable areal unit problem. Finally, we consider how to use and interpret the complex models, how model selection may vary depending on the intended user group and how best to communicate results to stakeholders and the general public.

Flexible and structured survival model for a simultaneous estimation of non-linear and non-proportional effects and complex interactions between continuous variables: Performance of this multidimensional penalized spline approach in net survival trend analysis

Cancer survival trend analyses are essential to describe accurately the way medical practices impact patients' survival according to the year of diagnosis. To this end, survival models should be able to account simultaneously for non-linear and non-proportional effects and for complex interactions between continuous variables. However, in the statistical literature, there is no consensus yet on how to build such models that should be flexible but still provide smooth estimates of survival. In this article, we tackle this challenge by smoothing the complex hypersurface (time since diagnosis, age at diagnosis, year of diagnosis, and mortality hazard) using a multidimensional penalized spline built from the tensor product of the marginal bases of time, age, and year. Considering this penalized survival model as a Poisson model, we assess the performance of this approach in estimating the net survival with a comprehensive simulation study that reflects simple and complex realistic survival trends. The bias was generally small and the root mean squared error was good and often similar to that of the true model that generated the data. This parametric approach offers many advantages and interesting prospects (such as forecasting) that make it an attractive and efficient tool for survival trend analyses.

Two-level spatially structured models in spatio-temporal disease mapping

This work focuses on extending some classical spatio-temporal models in disease mapping. The objective is to present a family of flexible models to analyze real data naturally organized in two different levels of spatial aggregation like municipalities within health areas or provinces, or counties within states. Model fitting and inference will be carried out using integrated nested Laplace approximations. The performance of the new models compared to models including a single spatial random effect is assessed by simulation. Results show good behavior of the proposed two-level spatially structured models in terms of several criteria. Brain cancer mortality data in the municipalities of two regions in Spain will be analyzed using the new model proposals. It will be shown that a model with two-level spatial random effects overcomes the usual single-level models.

Brain and central nervous system cancer incidence in navarre (Spain), 1973-2008 and projections for 2014

Different studies have pointed out Navarre as one of the regions of Spain with the highest incidence rates of brain and other central nervous system (CNS) cancer. Trend analysis for cancer incidence rates for long periods of time, might help determining risk factors as well as, assessing prevention actions involved in this disease. The objective of this study was to describe the incidence of brain and CNS cancer using data from the population-based cancer registry of Navarre, (Spain) during the period 1973-2008 and provide forecast figures up to-2014. Crude and age-standardized (world population) incidence rates of brain cancer per 100,000 person-years were calculated by the direct method separately by gender, area (Pamplona and others), and age-groups. Penalized splines for smoothing rates in the temporal dimensions were applied in order to estimate and forecast cancer incidence rates. Age-adjusted incidence rates showed an increase over the study and forecast periods in both sexes more marked in women than in men. Higher incidence rates were observed in men compared with women but the differences became smaller with time. The increase was due to the rise of rates in the oldest age groups since the rates for younger age groups remained stable or decreased over time. As the entire aetiology of brain and other CNS cancer is not still clear, keep promoting healthful lifestyles for cancer primary prevention among the whole population is necessary.

On fitting spatio-temporal disease mapping models using approximate Bayesian inference

Spatio-temporal disease mapping comprises a wide range of models used to describe the distribution of a disease in space and its evolution in time. These models have been commonly formulated within a hierarchical Bayesian framework with two main approaches: an empirical Bayes (EB) and a fully Bayes (FB) approach. The EB approach provides point estimates of the parameters relying on the well-known penalized quasi-likelihood (PQL) technique. The FB approach provides the posterior distribution of the target parameters. These marginal distributions are not usually available in closed form and common estimation procedures are based on Markov chain Monte Carlo (MCMC) methods. However, the spatio-temporal models used in disease mapping are often very complex and MCMC methods may lead to large Monte Carlo errors and a huge computation time if the dimension of the data at hand is large. To circumvent these potential inconveniences, a new technique called integrated nested Laplace approximations (INLA), based on nested Laplace approximations, has been proposed for Bayesian inference in latent Gaussian models. In this paper, we show how to fit different spatio-temporal models for disease mapping with INLA using the Leroux CAR prior for the spatial component, and we compare it with PQL via a simulation study. The spatio-temporal distribution of male brain cancer mortality in Spain during the period 1986-2010 is also analysed.

Evaluating space-time models for short-term cancer mortality risk predictions in small areas

Current cancer mortality data are available with a delay of roughly three years due to the administrative procedure necessary to create the registries. Therefore, health agencies rely on forecast cancer deaths. In this context, statistical procedures providing mortality/incidence risk predictions for different regions or health areas are very useful. These predictions are essential for defining priorities for cancer prevention and treatment. The main objective of this work is to evaluate the predictive performance of alternative spatio-temporal models for short-term cancer risk/counts prediction in small areas. All the models analyzed here are presented under a general-mixed model framework, providing a unified structure of presentation and facilitating the use of similar tools for computing the prediction mean squared error. Prostate cancer mortality data are used to illustrate the behavior of the different models in Spanish provinces.