Abstract

Purpose: This study focused on using Moran’s tests and logistic regression to detect changes in spatial clustering for females and males. Methods: For spatial distribution analysis, an average morbidity rate for a 7-year period was calculated. Medical cases from Taiwan National Health Insurance (NHI) were used as the numerator, and the denominator was the average mid-year population. Spatial analysis techniques, with a morbidity-smoothing coefficient estimate based on the empirical Bayesian method, were incorporated and applied to global and local Moran tests. In addition, we used a logistic regression model to test the characteristics of similarity and dissimilarity between males and females and to formulate the common spatial risk. Results: The mean found by local spatial autocorrelation analysis was used to identify spatial cluster patterns. There is great interest in discovering the relationship between leading health care problems and spatial risk factors. For example, in Taiwan, the geographic distribution of clusters where neoplasms were prevalent was found to closely correspond to the locations in the arseniasis-endemic areas of Southwestern and Northeastern Taiwan, as well as to locations in the Tainan urban area (for females) and clusters in Changhua County and Yunlin County (for males). The high-density populations in urban areas showed carcinogen clusters in Taiwan’s 3 main urban centers (i.e., Taipei, Taichung, and Kaohsiung) for female neoplasms. Conclusion: Cluster mapping helped clarify issues such as the spatial aspects of both the internal and external correlations for leading health care events. This information greatly assists in assessing spatial risk factors, which facilitates the planning of the most advantageous types of health care policies, as well as the implementation of effective health care services.

This work is licensed under a Creative Commons Attribution 4.0 License.