Academic journal article Geographical Analysis

Exploring Scale-Dependent Correlations between Cancer Mortality Rates Using Factorial Kriging and Population-Weighted Semivariograms

Academic journal article Geographical Analysis

Exploring Scale-Dependent Correlations between Cancer Mortality Rates Using Factorial Kriging and Population-Weighted Semivariograms

Article excerpt

This article presents a geostatistical methodology that accounts for spatially varying population size in the processing of cancer mortality data. The approach proceeds in two steps: (1) spatial patterns are first described and modeled using population-weighted semivariogram estimators, (2) spatial components corresponding to nested structures identified on semivariograms are then estimated and mapped using a variant of factorial kriging. The main benefit over traditional spatial smoothers is that the pattern of spatial variability (i.e., direction-dependent variability, range of correlation, presence of nested scales of variability) is directly incorporated into the computation of weights assigned to surrounding observations. Moreover, besides filtering the noise in the data, the procedure allows the decomposition of the structured component into several spatial components (i.e., local versus regional variability) on the basis of semivariogram models. A simulation study demonstrates that maps of spatial components are closer to the underlying risk maps in terms of prediction errors and provide a better visualization of regional patterns than the original maps of mortality rates or the maps smoothed using weighted linear averages. The proposed approach also attenuates the underestimation of the magnitude of the correlation between various cancer rates resulting from noise attached to the data. This methodology has great potential to explore scale-dependent correlation between risks of developing cancers and to detect clusters at various spatial scales, which should lead to a more accurate representation of geographic variation in cancer risk, and ultimately to a better understanding of causative relationships.

**********

Introduction

Cancer mortality maps are important tools in health research, allowing the identification of spatial patterns, clusters, and disease "hot spots" that often stimulate research to elucidate causative relationships (Jacquez 1998; Rushton, Elmes, and McMaster 2000). For example, an early version of the National Cancer Institute's (NCI's) cancer atlas stimulated research that uncovered associations between snuff dipping and oral cancer (Win et al. 1981), as well as the association between shipyard asbestos exposure and lung cancers (Blot et al. 1978). Analysis of mortality maps for a series of time intervals can also contribute to a greater understanding of temporal trends and can help to pinpoint locations where health policies need to be changed. For example, comparison of maps of mortality rates of cervix uteri cancer from 1950 through 1994 highlighted states that did not follow the national decline because poverty reduced access to health care and to early detection through the Pap smear test in particular (Friedell et al. 1992). Another use of mortality maps is the exploration of multivariate relationships in cancer mortality. A population exposed to a given carcinogen often exhibits excess risk of cancer at several different body sites, that is, ionizing radiation exposure is associated with excess lymphomas, leukemia, and cancers of the thyroid, breast, lung, and other organs (Beebe et al. 1978). Across populations, the incidence of different cancers varies and is related to differences in genetics and exposure to carcinogens. Several studies have demonstrated that multivariate analysis can reveal common risk factors underlying patterns of covariation in site-specific cancer mortalities. For example, in a factor analysis of age-adjusted mortality rates for 15 cancers in 46 countries, Groves et al. (1987) found that lymphoma and cancers of the colon, rectum, lung, and prostate were highly correlated with a first factor the authors associated with smoking and high-fat diets.

Detection of space-time patterns and multivariate relationships is frequently hampered by the presence of noise in mortality data, which is often caused by unreliable extreme relative risks estimated over small areas, such as United States ZIP code areas or census tracts (Mungiole, Pickle, and Hansen Simonson 1999; MacNab and Dean 2002). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.