Spatial scan statistics can be dangerous
Spatial scan statistics can be dangerous
复制标题
DOI：
10.1177/0962280220930562
复制
发表时间：
2021-01-01
影响因子：
2.3
通讯作者：
Tango, Toshiro
中科院分区：
文献类型：
--
作者：
Tango, Toshiro
空间扫描统计是广泛使用的疾病集群检测工具。特别是Kulldorff沿着SaTScan软件提出的圆形空间扫描统计量已被广泛应用于各种流行病学研究和疾病监测。然而，由于它不能检测非圆形，不规则形状的集群，许多作者提出了非圆形空间扫描统计。首先，还使用了Tango和Takahashi沿着的FlexXScan软件提出的灵活的空间扫描统计。然而，人们似乎没有很好地认识到，这些空间扫描统计，特别是SaTScan，往往会检测到最可能的集群，比真正的集群大得多，通过吸收邻近地区的疾病发生的风险不高。因此，如果研究人员报告了检测到的最可能的集群，这可能会导致对他们的批评，因为一些没有升高风险的地区被包括在检测到的最可能的集群中。在本文中，为了避免检测到这种不良的和误导性的集群，可能会引起社会关注，我们将提出使用限制的似然比提出的探戈，并说明在日本的两种死亡率数据的程序。
Spatial scan statistics are widely used tools for the detection of disease clusters. Especially, the circular spatial scan statistic proposed by Kulldorff along with SaTScan software has been used in a wide variety of epidemiological studies and disease surveillance. However, as it cannot detect non-circular, irregularly shaped clusters, many authors have proposed non-circular spatial scan statistics. Above all, the flexible spatial scan statistic proposed by Tango and Takahashi along with FleXScan software has also been used. However, it does not seem to be well recognized that these spatial scan statistics, especially SaTScan, tend to detect the most likely cluster, much larger than the true cluster by absorbing neighboring regions with nonelevated risk of disease occurrence. Therefore, if researchers reported the detected most likely cluster as they are, it might lead to a criticism to them due to the fact that some regions with nonelevated risk are included in the detected most likely cluster. In this paper, to avoid detecting such undesirable and misleading clusters which might cause a social concern, we shall propose the use of the restricted likelihood ratio proposed by Tango and illustrate the procedure with two kinds of mortality data in Japan.