Authors: Huiyu Lin*, Kent State University, Jay Lee, Kent State University; Henan University, China
Topics: Geography and Urban Health, Spatial Analysis & Modeling, Geographic Information Science and Systems
Keywords: Modifiable Spatial and Temporal Units Problem; cancer cluster; spatiotemporal pattern
Session Type: Paper
Start / End Time: 2:00 PM / 3:40 PM
Room: Tyler, Marriott, Mezzanine Level
Presentation File: No File Uploaded
Geographical distribution of cancer and cancer clusters has been increasingly interested in scholars and policy makers. In addition to exploring the spatial heterogeneity of cancer incidences, temporal analysis has been increasingly being incorporated into cancer research. Exploring spatial-temporal patterns of cancer cases provides valuable information for cancer predictions and potential public health actions. In addition, studies have also dedicated to revealing the relationships between the geography of cancer cases and local environmental contaminants and risks. However, for the construction of spatiotemporal cancer patterns, no universal spatiotemporal units are recognized for analysis. Most studies employe spatial or temporal units such as counties and years for analysis. Due to the effects of the recently proposed Modifiable Spatial and Temporal Unit Problem (MSTUP), the inconsistent result may be produced in cancer pattern research. In this research, three cancer types are selected - lung cancer, breast cancer, and prostate cancer - to explore and compare their spatiotemporal patterns in the State of Ohio, from 2010 to 2017. Cancer patterns are identified using Local Indicator of Spatial Autocorrelation (LISA) with finer, different spatiotemporal units. Relationships between the spatiotemporal cancer clusters and local environmental characteristics are examined to highlight differences caused by the MSTUP. This research presented the methodological issue of MSTUP and its effects on the cancer patterns outcome and provides new evidence for cancer research.