MSU Statisticians Identify Florida Cancer Cluster
Higher-than-expected rates of pediatric cancers have been identified in the Miami metro area and an area west of the Everglades in a series of five statistical analyses conducted for Statistics and Public Policy, an open-access, online public-policy journal published recently by the American Statistical Association (ASA).
The anomalous rates were detected by five research teams, including one from Michigan State University. Each team used different epidemiological and statistical methodology on a data set spanning the period 2000-2010 provided by Florida Association of Pediatric Tumor Programs (FAPTP). The FAPTP is a consortium of diagnosis and treatment centers that consolidates data for incorporation into the Florida Cancer Data System, part of the National Program of Cancer Registries of the U.S. Centers for Disease Control and Prevention.
The research groups applied different analytical approaches to achieve the same goal: detect spatio-temporal pediatric cancer clusters. The analyses used familiar methods–scan statistics, classification and hierarchical Bayesian modeling–as well as some ideas new to disease clustering: wombling and machine learning.
The MSU team of Chae Young Lim, Tapabrata Maiti and Zhen Zhang— all from the Department of Statistics and Probability — applied a spatial clustering method categorizing zip code tabulation areas into geographically contiguous groups sharing a common risk level of childhood cancer rates in Florida for the period 2000-2010. They identified high-risk areas consistent with the previous studies using the data for the period 2000 to 2007, but produced more detailed maps identifying a significantly higher risk area compared to the other regions in the Southern Florida.
The MSU team’s work took seven months to complete and was based in part on a new algorithm from Zhang’s thesis. Read the full published paper.
During their respective analyses of the FAPTP data, the research groups found several suggestive results. For instance, each approach identified local areas in which the observed pediatric cancer rate is statistically significantly higher than the rate expected given the number of people at risk. While the precise areas of high reported risk differ between methods, the groups identified a few common results that are not identical, but overlap.
Some families in the affected areas have blamed a nearby defense contractor for polluting ground water. In 2010, the state Department of Environmental Protection declared the water safe. Yet last year the magazine, The Nation, produced an investigative story finding a number of similar cases of brain tumors in children.
The results of the various analyses and a summary report have been presented to local authorities for their review and appropriate response.
Contributor: Tapabrata Maiti
The above story is based on materials provided by American Statistical Association.
1. Lance A. Waller. Discussion: Statistical Cluster Detection, Epidemiologic Interpretation, and Public Health Policy. Statistics and Public Policy, 2015; 2 (1): 1 DOI: 10.1080/2330443X.2015.1026621
American Statistical Association. (2015, April 20). Higher-than-expected pediatric cancer rates identified in two Florida areas. ScienceDaily. Retrieved May 8, 2015 from www.sciencedaily.com/releases/2015/04/150420120311.htm