The temporal-spatial distribution of seriously maltreated children. (February 2017)

• Record Type: Journal Article
• Authors: Thurston, Holly; Freisthler, Bridget; Bell, Janice; Tancredi, Daniel; Romano, Patrick S.; Miyamoto, Sheridan; Joseph, Jill G.
• Is Part Of: Spatial and spatio-temporal epidemiology. Volume 20(2017)
• Page Start: 1
• Page End: 8
• Publication Date: 2017-02
• Subjects: Epidemiology -- Statistical methods -- Periodicals; Epidemiology -- Periodicals; 614.4072
• DOI: 10.1016/j.sste.2016.12.004
• Languages: English
• ISSNs: 1877-5845

Abstract

Highlights: While the geographic dispersion of cases relative to controls did not differ, clustering of cases relative to the general population of young children was found in four distinct areas. The clustering accounted for 47 cases spanning 5 years, so the majority of cases were outside these cluster areas. The "hot spots" shifted over time and space. Abstract: This descriptive study utilized Bernoulli and Poisson spatial scan statistical models in SatScan v.9.4 to examine the distribution in space and time of residence of maltreatment cases–operationalized as families with serious maltreatment (resulting in death or hospitalization) of children under 6 years–for the presence of clusters ("hot spots"). In the Poisson model, a population dataset of serious maltreatment cases were non-randomly dispersed in four major areas, with these "hot spots" moving over time and space. Most cases were outside these clusters. In the Bernoulli model, the geographic distribution of a case-control dataset of families with serious maltreatment who were previously investigated by child welfare did not differ compared to controls previously investigated by child welfare with no serious maltreatment. Findings suggest that child fatality prevention efforts such as Back to Sleep and Never Shake a Baby campaigns should continue to be universal efforts, targeted to all parents.