Extensions to Multivariate Space Time Mixture Modeling of Small Area Cancer Data

Abstract

Oral cavity and pharynx cancer, even when considered together, is a fairly rare disease. Implementation of multivariate modeling with lung and bronchus cancer, as well as melanoma cancer of the skin, could lead to better inference for oral cavity and pharynx cancer. The multivariate structure of these models is accomplished via the use of shared random effects, as well as other multivariate prior distributions. The results in this paper indicate that care should be taken when executing these types of models, and that multivariate mixture models may not always be the ideal option, depending on the data of interest.

Small-area variation of cardiovascular diseases and select risk factors and their association to household and area poverty in South Africa: Capturing emerging trends in South Africa to better target local level interventions

“…A number of multivariate mapping models have been proposed in the literature including the shared spatial component models [24,25] and multivariate conditionally autoregressive (MCAR) models [26][27][28]. These joint mapping models have been used to model and estimate risks of related cancers [29][30][31] childhood illnesses [32] and childhood cancer and diabetes [33,34]. We suppose Y ijk is a binary response corresponding to participant i in district j having a k CVD disease k (j = 1,. .…”

Smoothed Temporal Atlases of Age-Gender All-Cause Mortality in South Africa

“…These issues could be addressed effectively within a spatiotemporal modelling of mortality rates to permit an assessment of the evolution of mortality dependence on both space and time. Moreover, these models help to address the problem of missing and unmeasured ecological determinants of mortality [ 15 , 16 ].…”

“…Currently, there is no research in South Africa, and to the best of our knowledge in sub-Saharan Africa, providing a dynamic image of a country’s burden of disease status from a spatial-temporal point of view. Most age and gender all-cause mortality risks are analyzed independently, either descriptively [ 7 ] or using disease mapping models [ 14 , 16 , 20 , 21 ]). However, several cause-specific mortality risks from certain diseases are age- and gender-dependent, for example, infectious-related deaths are more prevalent among young adults and children, and mortality from non-communicable disease are higher in the elderly [ 22 ].…”

“…However, several cause-specific mortality risks from certain diseases are age- and gender-dependent, for example, infectious-related deaths are more prevalent among young adults and children, and mortality from non-communicable disease are higher in the elderly [ 22 ]. Thus, independent analyses of age and gender mortality risks would not capture common and uncommon risk factor relationships between a number of age-gender mortality rates [ 16 ]. It becomes necessary to isolate common and age-gender-specific geographical patterns of mortality in a single model, which can be achieved by using the recently developed multivariate disease mapping models for multiple disease outcomes [ 3 , 10 , 11 , 13 , 15 , 16 , 23 ].…”

“…Thus, independent analyses of age and gender mortality risks would not capture common and uncommon risk factor relationships between a number of age-gender mortality rates [ 16 ]. It becomes necessary to isolate common and age-gender-specific geographical patterns of mortality in a single model, which can be achieved by using the recently developed multivariate disease mapping models for multiple disease outcomes [ 3 , 10 , 11 , 13 , 15 , 16 , 23 ]. These models will enable us to assess similarities as well as differences between geographical risk patterns among age-gender groups purportedly sharing common risk profiles.…”