Neighborhood characteristics and violence behind closed doors: The spatial overlap of child maltreatment and intimate partner violence

• Published in: PloS one Vol. 13; no. 6; p. e0198684
• Main Authors: Gracia, Enrique, López-Quílez, Antonio, Marco, Miriam, Lila, Marisol
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 07.06.2018; Public Library of Science (PLoS)
• Subjects: Adolescent; Aggression; Alcohol; Analysis; Bayesian analysis; Census; Censuses; Child; Child abuse; Child abuse & neglect; Child Abuse - statistics & numerical data; Children; Communities; Community research; Criminal investigations; Domestic violence; Earth Sciences; Economic Status - statistics & numerical data; Education; Educational Status; Emigrants and Immigrants - statistics & numerical data; Families & family life; Female; Health risk assessment; Higher education; Humans; Immigrants; Intimate partner violence; Male; Medicine and Health Sciences; Minority groups; Modelling; Neighborhoods; People and Places; Public health; Research and Analysis Methods; Residence Characteristics - statistics & numerical data; Residential areas; Risk; Risk Factors; Social psychology; Social Sciences; Spain; Spatial analysis; Spatial distribution; Spouse Abuse - statistics & numerical data; Stability; Studies; Trauma; Violence; Spain; United States > US; Chicago Illinois
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0198684

In this study, we analyze first whether there is a common spatial distribution of child maltreatment (CM) and intimate partner violence (IPV), and second, whether the risks of CM and IPV are influenced by the same neighborhood characteristics, and if these risks spatially overlap. To this end we used geocoded data of CM referrals (N = 588) and IPV incidents (N = 1450) in the city of Valencia (Spain). As neighborhood proxies, we used 552 census block groups. Neighborhood characteristics analyzed at the aggregated level (census block groups) were: Neighborhood concentrated disadvantage (neighborhood economic status, neighborhood education level, and policing activity), immigrant concentration, and residential instability. A Bayesian joint modeling approach was used to examine the spatial distribution of CM and IPV, and a Bayesian random-effects modeling approach was used to analyze the influence of neighborhood-level characteristics on small-area variations of CM and IPV risks. For CM, 98% of the total between-area variation in risk was captured by a shared spatial component, while for IPV the shared component was 77%. The risks of CM and IPV were higher in neighborhoods characterized by lower levels of economic status and education, and higher levels of policing activity, immigrant concentration, and residential instability. The correlation between the log relative risk of CM and IPV was .85. Most census block groups had either low or high risks in both outcomes (with only 10.5% of the areas with mismatched risks). These results show that certain neighborhood characteristics are associated with an increase in the risk of family violence, regardless of whether this violence is against children or against intimate partners. Identifying these high-risk areas can inform a more integrated community-level response to both types of family violence. Future research should consider a community-level approach to address both types of family violence, as opposed to individual-level intervention addressing each type of violence separately.