Decomposition of Moran’s Coefficient to Detect NonMulticollinear, Non-Zero, Eigen-Autocorrelated, Non-Gaussian Coefficients in Colorectal Cancer Estimator Determinants Epidemiologically Sampled in Hillsborough County, Florida

Read full text Summarise with VIRAT
Authors: Katlin Eaton
Cited by: 0
References: 14

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Despite the production of state-level Colorectal Cancer (CRC) incidence statistics, there are currently no precision count variable models to understand localized incidence rates of CRC.This article aims to utilize a predictive county variable model with semi-parametric eigen spatial autocorrelation to map Hillsborough County-level CRC incidence rates using zip code census data.The first is an over-dispersed Poisson regression model that uses a negative binomial model with a non-homogeneously distributed mean to account for outliers.An eigenfunction, eigen decomposition, and spatial filter technique is presented.The dependent variable was the incidence percentage of CRC at the county level, while independent variables included sociodemographic indicators obtained from the U.S. Census Bureau.This study used sociodemographic information at the zip code level in Hillsborough County, Florida, to investigate the geographical aggregation of colorectal cancer cases.Only the white population emerged as a significant predictor in the Poisson regression model, which demonstrated a non-dispersed paradigm.Several non-zero autocorrelated clusters were found across different zip codes in Hillsborough County using a second-order eigenfunction eigen decomposition.A spatial autocorrelation hot and cold spot analysis was conducted.This analysis identified zip codes with the highest and lowest predicted likelihood of CRC incidence.The identified zip code locations were 33578, 33511, and 33647 in southern Hillsborough County in the Brandon and Riverview area.The suggested method found hotspots for colorectal cancer where the white population is the main risk factor, which led to greater hotspot concentrations in Brandon and Riverview.Future studies should encourage routine colorectal cancer screening among individuals in these at-risk locations and investigate the method's applicability at the state level.

Fields of Study & Subjects

Colorectal Cancer Screening and Detection Global Cancer Incidence and Screening Spatial and Panel Data Analysis

Publication details

Journal / source
Annals of Biostatistics & Biometric Applications
Volume
7
Issue
2
Pages
1–11
Publication date
17 February 2026
Language
EN