Space-Time Statistical Insights about Geographic Variation in Lung Cancer Incidence Rates: Florida, USA, 2000⁻2011

Lung Cancer Prevalence in Virginia: A Spatial Zipcode-Level Analysis via INLA

BACKGROUND

Examining lung cancer (LC) cases in Virginia (VA) is essential due to its significant public health implications. By studying demographic, environmental, and socioeconomic variables, this paper aims to provide insights into the underlying drivers of LC prevalence in the state adjusted for spatial associations at the zipcode level.

METHODS

We model the available VA zipcode-level LC counts via (spatial) Poisson and negative binomial regression models, taking into account missing covariate data, zipcode-level spatial association and allow for overdispersion. Under latent Gaussian Markov Random Field (GMRF) assumptions, our Bayesian hierarchical model powered by Integrated Nested Laplace Approximation (INLA) considers simultaneous (spatial) imputation of all missing covariates through elegant prediction. The spatial random effect across zip codes follows a Conditional Autoregressive (CAR) prior.

RESULTS

Zip codes with elevated smoking indices demonstrated a corresponding increase in LC counts, underscoring the well-established connection between smoking and LC. Additionally, we observed a notable correlation between higher Social Deprivation Index (SDI) scores and increased LC counts, aligning with the prevalent pattern of heightened LC prevalence in regions characterized by lower income and education levels. On the demographic level, our findings indicated higher LC counts in zip codes with larger White and Black populations (with Whites having higher prevalence than Blacks), lower counts in zip codes with higher Hispanic populations (compared to non-Hispanics), and higher prevalence among women compared to men. Furthermore, zip codes with a larger population of elderly people (age ≥ 65 years) exhibited higher LC prevalence, consistent with established national patterns.

CONCLUSIONS

This comprehensive analysis contributes to our understanding of the complex interplay of demographic and socioeconomic factors influencing LC disparities in VA at the zip code level, providing valuable information for targeted public health interventions and resource allocation. Implementation code is available at GitHub.

Explainable artificial intelligence (XAI) for exploring spatial variability of lung and bronchus cancer (LBC) mortality rates in the contiguous USA

Machine learning (ML) has demonstrated promise in predicting mortality; however, understanding spatial variation in risk factor contributions to mortality rate requires explainability. We applied explainable artificial intelligence (XAI) on a stack-ensemble machine learning model framework to explore and visualize the spatial distribution of the contributions of known risk factors to lung and bronchus cancer (LBC) mortality rates in the conterminous United States. We used five base-learners-generalized linear model (GLM), random forest (RF), Gradient boosting machine (GBM), extreme Gradient boosting machine (XGBoost), and Deep Neural Network (DNN) for developing stack-ensemble models. Then we applied several model-agnostic approaches to interpret and visualize the stack ensemble model's output in global and local scales (at the county level). The stack ensemble generally performs better than all the base learners and three spatial regression models. A permutation-based feature importance technique ranked smoking prevalence as the most important predictor, followed by poverty and elevation. However, the impact of these risk factors on LBC mortality rates varies spatially. This is the first study to use ensemble machine learning with explainable algorithms to explore and visualize the spatial heterogeneity of the relationships between LBC mortality and risk factors in the contiguous USA.

Deeper Spatial Statistical Insights into Small Geographic Area Data Uncertainty

Small areas refer to small geographic areas, a more literal meaning of the phrase, as well as small domains (e.g., small sub-populations), a more figurative meaning of the phrase. With post-stratification, even with big data, either case can encounter the problem of small local sample sizes, which tend to inflate local uncertainty and undermine otherwise sound statistical analyses. This condition is the opposite of that afflicting statistical significance in the context of big data. These two definitions can also occur jointly, such as during the standardization of data: small geographic units may contain small populations, which in turn have small counts in various age cohorts. Accordingly, big spatial data can become not-so-big spatial data after post-stratification by geography and, for example, by age cohorts. This situation can be ameliorated to some degree by the large volume of and high velocity of big spatial data. However, the variety of any big spatial data may well exacerbate this situation, compromising veracity in terms of bias, noise, and abnormalities in these data. The purpose of this paper is to establish deeper insights into big spatial data with regard to their uncertainty through one of the hallmarks of georeferenced data, namely spatial autocorrelation, coupled with small geographic areas. Impacts of interest concern the nature, degree, and mixture of spatial autocorrelation. The cancer data employed (from Florida for 2001-2010) represent a data category that is beginning to enter the realm of big spatial data; its volume, velocity, and variety are increasing through the widespread use of digital medical records.

Spatial and Temporal Analysis of Lung Cancer in Shenzhen, 2008-2018

Lung cancer is the most commonly diagnosed cancer in China. The incidence trend and geographical distribution of lung cancer in southern China have not been reported. The present study explored the temporal trend and spatial distribution of lung cancer incidence in Shenzhen from 2008 to 2018. The lung cancer incidence data were obtained from the registered population in the Shenzhen Cancer Registry System between 2008 and 2018. The standardized incidence rates of lung cancer were analyzed by using the joinpoint regression model. The Moran's I method was used for spatial autocorrelation analysis and to further draw a spatial cluster map in Shenzhen. From 2008 to 2018, the average crude incidence rate of lung cancer was 27.1 (1/100,000), with an annual percentage change of 2.7% (p < 0.05). The largest average proportion of histological type of lung cancer was determined as adenocarcinoma (69.1%), and an increasing trend was observed in females, with an average annual percentage change of 14.7%. The spatial autocorrelation analysis indicated some sites in Shenzhen as a high incidence rate spatial clustering area. Understanding the incidence patterns of lung cancer is useful for monitoring and prevention.