Child maltreatment data in the state of New Mexico across space and time

Child maltreatment is a serious public health problem. Previous research demonstrates that child maltreatment clusters in low-income, racially homogenous neighborhoods. Little is known, however, about the structural correlates of spatial risk in small areas such as census tracts. Here we present additional information regarding the data and methods used in the recent article published in Child Abuse & Neglect entitled “Variability and stability in child maltreatment risk across time and space and its association with neighborhood social & housing vulnerability in New Mexico: A Bayesian space-time model” [1]. The present dataset merges child maltreatment data from the New Mexico Department of Public Health with multiple sources of publicly available data to create a novel public health analysis. Bayesian spatio-temporal modeling techniques were used to map the relative risk of substantiated child maltreatment across census tracts in the state, and to elucidate spatial and temporal heterogeneity in risk. The data was initially collected by the New Mexico Children, Youth and Families Department, the state organization that suspected child abuse and neglect cases are reported to and the organization that then substantiates these cases. The data were then sent to the New Mexico Community Data Collaborative, a data analytic organization under the umbrella of the New Mexico Department of Health. The point file consisting of home addresses of substantiated cases of child abuse was then aggregated by census tract, mapped for the entire state of New Mexico and made available to the public for research and analysis by different public health organizations and researchers (including the present researchers). The very purpose of making the data available to the public was to allow deeper investigations into trends and associations with other social determinants of health. This analysis demonstrates the public health importance of data sharing and accessibility.