Evaluation of the Gini Coefficient in Spatial Scan Statistics for Detecting Irregularly Shaped Clusters

Evaluating the ability of a locally focused culling program in removing chronic wasting disease infected free-ranging white-tailed deer in Illinois, USA, 2003-2020

In northern Illinois, chronic wasting disease (CWD) was first identified in free-ranging white-tailed deer (Odocoileus virginianus; hereafter referred to as "deer") in 2002. To reduce CWD transmission rates in Illinois, wildlife biologists have conducted locally focussed culling of deer since 2003 in areas where CWD has been detected. We used retrospective spatial, temporal and space-time scan statistical models to identify areas and periods where culling removed higher than expected numbers of CWD-positive deer. We included 490 Public Land Survey "sections" (∼2.59 km² ) from 15 northern Illinois counties in which at least one deer tested positive for CWD between 2003 and 2020. A negative binomial regression model compared the proportion of CWD positive cases removed from sections with at least one CWD case detected in the previous years, "local area 1 (L1)," to the proportion of CWD cases in adjacent sections-L2, L3, and L4-designated by their increasing distance from L1. Of the 14,661 deer removed and tested via culling, 325 (2.22 %) were CWD-positive. A single temporal CWD cluster occurred in 2020. Three spatial clusters were identified, with a primary cluster located at the border of Boone and Winnebago counties. Four space-time clusters were identified with a primary cluster in the northern portion of the study area from 2003 to 2005 that overlapped with the spatial cluster. The proportion of CWD cases removed from L1 (3.92, 95% CI, 2.56-6.01) and L2 (2.32, 95% CI, 1.50-3.59) were significantly higher compared to L3. Focussing culling efforts on accessible properties closest to L1 areas results in more CWD-infected deer being removed, which highlights the value of collaborations among landowners, hunters, and wildlife management agencies to control CWD. Continuous evaluation and updating of the culling and surveillance programs are essential to mitigate the health burden of CWD on deer populations in Illinois.

A spatiotemporal data mining study to identify high-risk neighborhoods for out-of-hospital cardiac arrest (OHCA) incidents

Out-of-hospital cardiac arrest (OHCA) is a worldwide health problem. The aim of the study is to utilize the territorial-wide OHCA data of Hong Kong in 2012-2015 to examine its spatiotemporal pattern and high-risk neighborhoods. Three techniques for spatiotemporal data mining (SaTScan's spatial scan statistic, Local Moran's I, and Getis Ord Gi*) were used to extract high-risk neighborhoods of OHCA occurrence and identify local clusters/hotspots. By capitalizing on the strengths of these methods, the results were then triangulated to reveal "truly" high-risk OHCA clusters. The final clusters for all ages and the elderly 65+ groups exhibited relatively similar patterns. All ages groups were mainly distributed in the urbanized neighborhoods throughout Kowloon. More diverse distribution primarily in less accessible areas was observed among the elderly group. All outcomes were further converted into an index for easy interpretation by the general public. Noticing the spatial mismatches between hospitals and ambulance depots (representing supplies) and high-risk neighborhoods (representing demands), this setback should be addressed along with public education and strategic ambulance deployment plan to shorten response time and improve OHCA survival rate. This study offers policymakers and EMS providers essential spatial evidence to assist with emergency healthcare planning and informed decision-making.

Optimizing the maximum reported cluster size in the spatial scan statistic for survival data

Background

The spatial scan statistic is a useful tool for cluster detection analysis in geographical disease surveillance. The method requires users to specify the maximum scanning window size or the maximum reported cluster size (MRCS), which is often set to 50% of the total population. It is important to optimize the maximum reported cluster size, keeping the maximum scanning window size at as large as 50% of the total population, to obtain valid and meaningful results.

Results

We developed a measure, a Gini coefficient, to optimize the maximum reported cluster size for the exponential-based spatial scan statistic. The simulation study showed that the proposed method mostly selected the optimal MRCS, similar to the true cluster size. The detection accuracy was higher for the best chosen MRCS than at the default setting. The application of the method to the Korea Community Health Survey data supported that the proposed method can optimize the MRCS in spatial cluster detection analysis for survival data.

Conclusions

Using the Gini coefficient in the exponential-based spatial scan statistic can be very helpful for reporting more refined and informative clusters for survival data.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12942-021-00286-w.

Identification of high-risk and low-risk clusters and estimation of the relative risk of acute lymphoblastic leukemia in provinces of Iran during 2006-2014 period: A geo-epidemiological study

BACKGROUND

The present study was conducted to determine the epidemiological status, identify high-risk and low-risk clusters, and estimate the relative risk (RR) of acute lymphoblastic leukemia (ALL) in provinces of Iran.

MATERIALS AND METHODS

This is an ecological study carried out using an Exploratory Multiple-Group design on 3769 children under 15 years of age with ALL from 2006 to 2014. Data analysis was performed using Mann-Whitney U, Global Moran's I and Kuldorff's purely spatial scan statistic tests at a significance level of 0.05.

RESULTS

The average annual incidence rate of ALL during 2006-2014 period was 2.25/100,000 children under 15 years of age. The most likely high-risk cluster with log-likelihood ratio (LLR) =327.47 is located in the southwestern part of Iran with a radius of 294.93 km and a centrality of 30.77 N and 50.83 E, which contained 1276 patients with a RR of 2.56. It includes Fars, Bushehr, Kohgiluyeh and Boyer-Ahmad, Khuzestan and Chahar Mahall and Bakhtiari provinces. On the other hand, the most likely low-risk cluster with 517 patients, and a RR 0.49 and LLR = 227.03 was identified in the northwestern part of Iran with a radius of 270.38 km and a centrality of 37.25 N and 49.49 E. It includes Zanjan, Qazvin, Gilan and East Azerbaijan, Ardabil, Alborz and Tehran provinces.

CONCLUSION

High-risk clusters were observed in Southwestern, central, and eastern Iran, while low-risk clusters were identified in Northern and Western Iran.

Spatial scan statistics for matched case-control data

Spatial scan statistics are widely used for cluster detection analysis in geographical disease surveillance. While this method has been developed for various types of data such as binary, count, and continuous data, spatial scan statistics for matched case-control data, which often arise in spatial epidemiology, have not been considered. We propose spatial scan statistics for matched case-control data. The proposed test statistics consider the correlations between matched pairs. We evaluate the statistical power and cluster detection accuracy of the proposed methods through simulations compared to the Bernoulli-based method. We illustrate the proposed methods using a real data example. The simulation study clearly revealed that the proposed methods had higher power and higher accuracy for detecting spatial clusters for matched case-control data than the Bernoulli-based spatial scan statistic. The cluster detection result of the real data example also appeared to reflect a higher power of the proposed methods. The proposed methods are very useful for spatial cluster detection for matched case-control data.

Identifying high-risk areas for nonfatal opioid overdose: a spatial case-control study using EMS run data

PURPOSE

The objective of our study was to incorporate stricter probable nonfatal opioid overdose case criteria, and advanced epidemiologic approaches to more reliably detect local clustering in nonfatal opioid overdose activity in EMS runs data.

METHODS

Data were obtained using emsCharts for our study area in southwestern Pennsylvania from 2007 to 2018. Cases were identified as emergency medical service (EMS) responses where naloxone was administered, and improvement was noted in patient records between initial and final Glasgow Coma Score. A subsample of all-cause EMS responses sites were used as controls and exact matched to cases on sex and 10-year-age category. Clustering was assessed using difference in Ripley's K function for cases and controls and Kulldorff scan statistics.

RESULTS

Difference in K functions indicated no significant difference in probable nonfatal overdose EMS runs across the study area compared to all-cause EMS runs. However, scan statistics did identify significant local clustering of probable nonfatal overdose EMS runs (maximum likelihood = 16.40, P = 0.0003).

CONCLUSIONS

Results highlight relevance of EMS data to detect community-level overdose activity and promote reliable use through stricter case definition criteria and advanced methodological approaches. Techniques examined have the potential to improve targeted delivery of neighborhood-level public health response activities using a near real-time data source.

Inequalities in the geographic distribution of hospital beds and doctors in traditional Chinese medicine from 2004 to 2014

Objectives

This study identifies inequities in the provincial-level geographical distribution of traditional Chinese Medicine (TCM) hospital beds and doctors in China from 2004 to 2014. This provides policy implications of the optimal allocation of TCM health care resources.

Methods

Our study used province level data on TCM hospital beds and doctors from 2004 to 2014. These data were obtained from the China TCM Yearbook 2004–2014 and the China Statistical Yearbook 2004–2014.Global and local spatial autocorrelation was performed by using Moran’s index and the local Moran’s index to describe the spatial distribution of TCM hospital beds (doctors) as well as their density. A Gini coefficient was used to estimate inequalities in the geographic distribution of TCM hospital beds (doctors) based on their density. Correlations of the Gini coefficients between TCM hospital beds and doctors were calculated by Pearson correlation analysis.

Results

All indicators of TCM hospital beds and doctor density have increased over the past 11 years. The number of TCM hospital beds per 10,000 populations increased the fastest. Geographical clustering was not obvious in the density distribution of TCM hospital beds or doctors, as no significant spatial autocorrelation was found. Gini coefficients showed that from 2004 to 2014 the distribution of TCM hospital beds per 10,000 population and doctors per 10,000 populations were equitable between different regions. A large gap existed in the distribution inequality of TCM hospital beds (doctors) per square kilometer among different regions.

Conclusion

Targeted health policy with equitable distribution of TCM hospital beds (doctors) per square kilometer and the balance and coordination of related resources should be a priority in shaping China’s healthcare system reform.

Spatial clustering of metal and metalloid mixtures in unregulated water sources on the Navajo Nation - Arizona, New Mexico, and Utah, USA

Contaminant mixtures are identified regularly in public and private drinking water supplies throughout the United States; however, the complex and often correlated nature of mixtures makes identification of relevant combinations challenging. This study employed a Bayesian clustering method to identify subgroups of water sources with similar metal and metalloid profiles. Additionally, a spatial scan statistic assessed spatial clustering of these subgroups and a human health metric was applied to investigate potential for human toxicity. These methods were applied to a dataset comprised of metal and metalloid measurements from unregulated water sources located on the Navajo Nation, in the southwest United States. Results indicated distinct subgroups of water sources with similar contaminant profiles and that some of these subgroups were spatially clustered. Several profiles had metal and metalloid concentrations that may have potential for human toxicity including arsenic, uranium, lead, manganese, and selenium. This approach may be useful for identifying mixtures in water sources, spatially evaluating the clusters, and help inform toxicological research investigating mixtures.

Optimizing the maximum reported cluster size in the spatial scan statistic for ordinal data

The spatial scan statistic is an important tool for spatial cluster detection. There have been numerous studies on scanning window shapes. However, little research has been done on the maximum scanning window size or maximum reported cluster size. Recently, Han et al. proposed to use the Gini coefficient to optimize the maximum reported cluster size. However, the method has been developed and evaluated only for the Poisson model. We adopt the Gini coefficient to be applicable to the spatial scan statistic for ordinal data to determine the optimal maximum reported cluster size. Through a simulation study and application to a real data example, we evaluate the performance of the proposed approach. With some sophisticated modification, the Gini coefficient can be effectively employed for the ordinal model. The Gini coefficient most often picked the optimal maximum reported cluster sizes that were the same as or smaller than the true cluster sizes with very high accuracy. It seems that we can obtain a more refined collection of clusters by using the Gini coefficient. The Gini coefficient developed specifically for the ordinal model can be useful for optimizing the maximum reported cluster size for ordinal data and helpful for properly and informatively discovering cluster patterns.