Limits...
A flexibly shaped spatial scan statistic for detecting clusters.

Tango T, Takahashi K - Int J Health Geogr (2005)

Bottom Line: The circular spatial scan statistics shows a high level of accuracy in detecting circular clusters exactly.The proposed spatial scan statistic is shown to work well for small to moderate cluster size, up to say 30.For larger cluster sizes, the method is not practically feasible and a more efficient algorithm is needed.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Technology Assessment and Biostatistics, National Institute of Public Health, 3-6 Minami 2 chome Wako, Saitama 351-0197, Japan. tango@niph.go.jp

ABSTRACT

Background: The spatial scan statistic proposed by Kulldorff has been applied to a wide variety of epidemiological studies for cluster detection. This scan statistic, however, uses a circular window to define the potential cluster areas and thus has difficulty in correctly detecting actual noncircular clusters. A recent proposal by Duczmal and Assunção for detecting noncircular clusters is shown to detect a cluster of very irregular shape that is much larger than the true cluster in our experiences.

Methods: We propose a flexibly shaped spatial scan statistic that can detect irregular shaped clusters within relatively small neighborhoods of each region. The performance of the proposed spatial scan statistic is compared to that of Kulldorff's circular spatial scan statistic with Monte Carlo simulation by considering several circular and noncircular hot-spot cluster models. For comparison, we also propose a new bivariate power distribution classified by the number of regions detected as the most likely cluster and the number of hot-spot regions included in the most likely cluster.

Results: The circular spatial scan statistics shows a high level of accuracy in detecting circular clusters exactly. The proposed spatial scan statistic is shown to have good usual powers plus the ability to detect the noncircular hot-spot clusters more accurately than the circular one.

Conclusion: The proposed spatial scan statistic is shown to work well for small to moderate cluster size, up to say 30. For larger cluster sizes, the method is not practically feasible and a more efficient algorithm is needed.

No MeSH data available.


Related in: MedlinePlus

A random sample from cluster model C. Dots describe the centroids of regions with some cases. Circles are drawn only for the regions whose standardized risk ratios are statistically significantly larger than 1 at α = 0.05 and the region number is placed in stead of dot. The radius is set inversely proportional to the tail probability.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC1173134&req=5

Figure 2: A random sample from cluster model C. Dots describe the centroids of regions with some cases. Circles are drawn only for the regions whose standardized risk ratios are statistically significantly larger than 1 at α = 0.05 and the region number is placed in stead of dot. The radius is set inversely proportional to the tail probability.

Mentions: As an illustration, we will apply the circular spatial scan statistic, the flexible spatial scan statistic and Duczmal and Assunção's spatial scan statistic to the disease map shown in Figure 2 which is a random sample of n = 235 cases assuming the cluster model C. Circles are drawn only for the regions whose observed-expected ratio (standardized risk ratio) is statistically significantly larger than 1 at α = 0.05. The radius of the circles is set inversely proportional to the upper tail p-value. The number shown in Figure 2 indicates the region number. Figure 2 obviously suggests the clusters occurring in the area including regions {14, 15, 26, 27, 33}.


A flexibly shaped spatial scan statistic for detecting clusters.

Tango T, Takahashi K - Int J Health Geogr (2005)

A random sample from cluster model C. Dots describe the centroids of regions with some cases. Circles are drawn only for the regions whose standardized risk ratios are statistically significantly larger than 1 at α = 0.05 and the region number is placed in stead of dot. The radius is set inversely proportional to the tail probability.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC1173134&req=5

Figure 2: A random sample from cluster model C. Dots describe the centroids of regions with some cases. Circles are drawn only for the regions whose standardized risk ratios are statistically significantly larger than 1 at α = 0.05 and the region number is placed in stead of dot. The radius is set inversely proportional to the tail probability.
Mentions: As an illustration, we will apply the circular spatial scan statistic, the flexible spatial scan statistic and Duczmal and Assunção's spatial scan statistic to the disease map shown in Figure 2 which is a random sample of n = 235 cases assuming the cluster model C. Circles are drawn only for the regions whose observed-expected ratio (standardized risk ratio) is statistically significantly larger than 1 at α = 0.05. The radius of the circles is set inversely proportional to the upper tail p-value. The number shown in Figure 2 indicates the region number. Figure 2 obviously suggests the clusters occurring in the area including regions {14, 15, 26, 27, 33}.

Bottom Line: The circular spatial scan statistics shows a high level of accuracy in detecting circular clusters exactly.The proposed spatial scan statistic is shown to work well for small to moderate cluster size, up to say 30.For larger cluster sizes, the method is not practically feasible and a more efficient algorithm is needed.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Technology Assessment and Biostatistics, National Institute of Public Health, 3-6 Minami 2 chome Wako, Saitama 351-0197, Japan. tango@niph.go.jp

ABSTRACT

Background: The spatial scan statistic proposed by Kulldorff has been applied to a wide variety of epidemiological studies for cluster detection. This scan statistic, however, uses a circular window to define the potential cluster areas and thus has difficulty in correctly detecting actual noncircular clusters. A recent proposal by Duczmal and Assunção for detecting noncircular clusters is shown to detect a cluster of very irregular shape that is much larger than the true cluster in our experiences.

Methods: We propose a flexibly shaped spatial scan statistic that can detect irregular shaped clusters within relatively small neighborhoods of each region. The performance of the proposed spatial scan statistic is compared to that of Kulldorff's circular spatial scan statistic with Monte Carlo simulation by considering several circular and noncircular hot-spot cluster models. For comparison, we also propose a new bivariate power distribution classified by the number of regions detected as the most likely cluster and the number of hot-spot regions included in the most likely cluster.

Results: The circular spatial scan statistics shows a high level of accuracy in detecting circular clusters exactly. The proposed spatial scan statistic is shown to have good usual powers plus the ability to detect the noncircular hot-spot clusters more accurately than the circular one.

Conclusion: The proposed spatial scan statistic is shown to work well for small to moderate cluster size, up to say 30. For larger cluster sizes, the method is not practically feasible and a more efficient algorithm is needed.

No MeSH data available.


Related in: MedlinePlus