A flexible parametric approach to examining spatial variation in relative survival

Cramb, Susanna; Mengersen, Kerrie; Lambert, Paul C.; Ryan, Louise; Baade, Peter D.

doi:10.1002/sim.7071

Cited by 14 publications

(10 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A major field of research in cancer epidemiology is the assessment of spatial patterns of cancer. Substantial research has focused on assessing spatial patterns of cancer in different parts of the world [16][17][18][19][20][21][22][23]. The common sources of spatially referenced cancer data used for spatial analysis of cancer are population-based cancer registries; those are supplemented with additional population data, health survey data along with environmental and remote sensing data [24].…”

Section: Introductionmentioning

confidence: 99%

Augmenting disease maps: a Bayesian meta-analysis approach

et al. 2020

Self Cite

View full text Add to dashboard Cite

Analysis of spatial patterns of disease is a significant field of research. However, access to unit-level disease data can be difficult for privacy and other reasons. As a consequence, estimates of interest are often published at the small area level as disease maps. This motivates the development of methods for analysis of these ecological estimates directly. Such analyses can widen the scope of research by drawing more insights from published disease maps or atlases. The present study proposes a hierarchical Bayesian meta-analysis model that analyses the point and interval estimates from an online atlas. The proposed model is illustrated by modelling the published cancer incidence estimates available as part of the online Australian Cancer Atlas (ACA). The proposed model aims to reveal patterns of cancer incidence for the 20 cancers included in ACA in major cities, regional and remote areas. The model results are validated using the observed areal data created from unit-level data on cancer incidence in each of 2148 small areas. It is found that the meta-analysis models can generate similar patterns of cancer incidence based on urban/rural status of small areas compared with those already known or revealed by the analysis of observed data. The proposed approach can be generalized to other online disease maps and atlases.

show abstract

Section: Introductionmentioning

confidence: 99%

Augmenting disease maps: a Bayesian meta-analysis approach

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…The number of events was assumed to follow a Poisson distribution, and two random effects were included in the model: a spatially structured random effect for local smoothing and an unstructured random effect global smoothing; Hennerfeind 9 proposed a Bayesian geoadditive relative survival model using penalized Psplines to model the log-baseline effect as well as the nonlinear and time-varying effects of covariates. Spatial and normal random effects were also included in the model formulation; Cramb et al 10 introduced a Bayesian flexible parametric model which extends a frequentist flexible parametric model on the log cumulative excess hazard scale using restricted cubic splines 20 by adding spatially structured random effects.…”

Section: Discussionmentioning

confidence: 99%

“…This applies regardless of the framework of inference, whether frequentist or Bayesian, although inferences for excess hazard models have mainly been based on the frequentist maximisation of the likelihood function. Very few options are available for inferences within the Bayesian framework [7][8][9][10] , in particular none describing the process of deriving a posterior distribution of net survival.…”

Section: Introductionmentioning

confidence: 99%

Flexible Bayesian excess hazard models using low-rank thin plate splines

Quaresma

Carpenter

Rachet

2019

Stat Methods Med Res

View full text Add to dashboard Cite

Excess hazard models became the preferred modelling tool in population-based cancer survival research. In this setting, the model is commonly formulated as the additive decomposition of the overall hazard into two components: the excess hazard due to the cancer of interest and the population hazard due to all other causes of death. We introduce a flexible Bayesian regression model for the log-excess hazard where the baseline log-excess hazard and any non-linear effects of covariates are modelled using low-rank thin plate splines. Using this type of splines will ensure that the log-likelihood function retains tractability not requiring numerical integration. We demonstrate how to derive posterior distributions for the excess hazard and for net survival, a population-level measure of cancer survival that can be derived from excess hazard models. We illustrate the proposed model using survival data for patients diagnosed with colon cancer during 2009 in London, England.

show abstract

“…This model was fitted within the class of generalized linear mixed models by using maximum likelihood methods. In a Bayesian framework, a similar type of model with spatially correlated random effects has been used to assessing spatial variation in cancer survival by Fairley et al, Cramb et al, Saez et al, and Kang et al Recently, Cramb et al extended a flexible parametric model to estimation of spatial variation in excess mortality. The disadvantage of these Bayesian approaches is that they are computationally intensive because the models are fitted by using Markov chain Monte Carlo (MCMC) methods.…”

Section: Introductionmentioning

confidence: 99%

“…The INLA approach can be easily applied by using the R‐INLA package (www.r‐inla.org) to a wide variety of statistical models including generalized linear mixed models. In the estimation of excess mortality and net survival, the INLA approach has been mentioned as a possible solution to overcome the computational challenges related to the MCMC approaches, but to our knowledge, it has not been employed for the estimation in practice.…”

Section: Introductionmentioning

confidence: 99%

Estimating multilevel regional variation in excess mortality of cancer patients using integrated nested Laplace approximation

Seppä

Rue

Hakulinen

et al. 2018

Statistics in Medicine

View full text Add to dashboard Cite

Models of excess mortality with random effects were used to estimate regional variation in relative or net survival of cancer patients. Statistical inference for these models based on the Markov chain Monte Carlo (MCMC) methods is computationally intensive and, therefore, not feasible for routine analyses of cancer register data. This study assessed the performance of the integrated nested Laplace approximation (INLA) in monitoring regional variation in cancer survival. Poisson regression model of excess mortality including both spatially correlated and unstructured random effects was fitted to the data of patients diagnosed with ovarian and breast cancer in Finland during 1955‐2014 with follow up from 1960 through 2014 by using the period approach with five‐year calendar time windows. We estimated standard deviations associated with variation (i) between hospital districts and (ii) between municipalities within hospital districts. Posterior estimates based on the INLA approach were compared to those based on the MCMC simulation. The estimates of the variation parameters were similar between the two approaches. Variation within hospital districts dominated in the total variation between municipalities. In 2000‐2014, the proportion of the average variation within hospital districts was 68% (95% posterior interval: 35%‐93%) and 82% (60%‐98%) out of the total variation in ovarian and breast cancer, respectively. In the estimation of regional variation, the INLA approach was accurate, fast, and easy to implement by using the R‐INLA package.

show abstract

A flexible parametric approach to examining spatial variation in relative survival

Cited by 14 publications

References 32 publications

Augmenting disease maps: a Bayesian meta-analysis approach

Augmenting disease maps: a Bayesian meta-analysis approach

Flexible Bayesian excess hazard models using low-rank thin plate splines

Estimating multilevel regional variation in excess mortality of cancer patients using integrated nested Laplace approximation

Contact Info

Product

Resources

About