Spatial distribution of maxillofacial injuries caused by urban violence: An ecological analysis to identify high‐risk areas

Abstract

Objectives To investigate the spatial and spatial‐temporal distribution of oral and maxillofacial injuries caused by urban violence, as well as to identify underlying disparities at regional level through a geostatistical approach. Methods This was a historical ecological cohort study of trauma cases caused by urban violence using aggregate data from victims assisted in a Brazilian medical‐forensic service between January 2012 and December 2015. The longitudinal patterns of change observed in each geographic area (neighbourhoods) were evaluated using the finite mixture model (FMM). The spatial autocorrelation of events was investigated using the Getis‐Ord Indicator (Gi*) to identify significant hot and cold spatial clusters. With a spatial regression model, it was also found when socioeconomic variables, residential infrastructure and neighbourhood infrastructure were associated with high incidence rates. The significance level was set at P ≤ 0.05. Results The finite mixture model revealed three different patterns of longitudinal trajectory of the incidence of oral and maxillofacial trauma caused by urban violence (TP1 to TP3, P < 0.05). TP1 was characterized by an incidence that remained stable and high over time, comprising 17.4% of the city's neighbourhoods. In TP2, it was observed that the incidence was moderate, with a slightly increasing trend in the last year evaluated, representing around 41.8% of the sample. In contrast, in TP3, it was found that the incidence was relatively low and remained stable over time, accounting for about 40.8% of the sample. The Getis‐Ord (Gi*) statistic identified significant high‐risk clusters in the western (P < 0.05), southern (P < 0.05), and eastern regions (P < 0.05) and low risk in the northern region (P < 0.05). The spatial regression model indicated significant association between areas with unfavourable socioeconomic conditions and higher incidence of events (β = 0.178, SE = 0.046, P < 0.001). Conclusions Clusters demarcating areas with high socio‐spatial vulnerability for urban violence and oral and maxillofacial injuries were identified. The findings highlight the need to improve living conditions in segregated urban areas and develop intersectoral actions to improve living conditions, employment, public safety, social support, health care and prevention.

Spatial-temporal analysis of head injuries at the northwest of Iran 2014–2018

“…(35) Hotspot analysis and Anselin Local Moran's I are advanced statistical techniques that used to determine the spatial patterns of HI in different health studies. (10,12,36,37) Spatial analysis play an important role to visualize the spatial patterns and identify the high-risk areas of HI incidence. Global Moran's I analysis can identify the spatial pattern and existence of spatial autocorrelation in HI pattern.…”

Geographic Information Systems in Spatial Epidemiology

“…Some GIS analytical techniques commonly used in oral epidemiology are[ 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 ]:…”

“…The authors used Getis–Ord indicator (Gi*) to identify significant hot and cold spatial clusters that allowed them to assess spatial autocorrelation of events. [ 10 ]…”

“…Association of dental caries, oropharyngeal cancers, oral and maxillofacial injuries, and periodontal diseases with various environmental factors such as access to public and private healthcare centers, school dental clinics, dentist shortages, areas in need of dental facilities, intervention effects on oral health, and individual and contextual level pathways to the oral health outcomes is being explored and analyzed. [ 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 ]…”

Estudo retrospectivo do trauma maxilofacial decorrente de violência atendido em um pronto socorro hospitalar

“…Fatores como idade e sexo podem influenciar na incidência das diversas etiologias. Nesse tipo de trauma as causas mais comuns referidas são: acidentes de trânsito, agressões físicas e quedas (Macedo et al, 2008;Smith et al, 2012;Pham-Dang et al, 2014;Bernardino et al, 2019;Goedecke et al, 2019;Cohn et al, 2020).…”