The Geography of Child Maltreatment: A Spatiotemporal Analysis Using Bayesian Hierarchical Analysis With Integrated Nested Laplace Approximation

The Spatial Scale and Spread of Child Victimization

Previous research shows that large, densely populated urban areas have higher rates of child victimization that have persisted over time. However, few investigations have inquired about the processes that produce and sustain hot and cold spots of child victimization. As a result, the mechanisms that produce the observed spatial clustering of child victimization, and hence "why" harms against children tend to cluster in space, remains unknown. Does the likelihood of being a victim of violence in one location depend on a similar event happening in a nearby location within a specified timeframe? Rather, are child victims of violence more likely to reside in suboptimal neighborhood conditions? This paper aims to present an analytical and theoretical framework for distinguishing between these locational (point) processes to determine whether the empirical spatial patterns undergirding child victimization are more reflective of the "spread" via contagion (i.e., dependency) or whether they are produced by neighborhood structural inequality resulting from spatial heterogeneity. To detect spatial dependence, we applied the inhomogeneous K-function to Los Angeles Medical Examiner data on child homicide victim locations while controlling for regional differences in victimization events (i.e., heterogeneity). Our analysis found strong evidence of spatial clustering in child victimization at small spatial scales but inhibition at larger scales. We further found limited support for the spatiotemporal clustering of child victimization indicative of a contagion effect. Overall, our results support the role of neighborhood structural vulnerability in the underlying mechanisms producing patterns of child victimization across Los Angeles County. We conclude by discussing the policy implications for understanding this spatial patterning in geographical context and for developing effective and targeted preventive interventions.

Application of geospatial analysis in health research: A systematic review of methodological aspects of studies on violence against children and young people

BACKGROUND

Geographical variation exists in violence experienced by children and young people; however, there is limited research applying geospatial techniques to study this variation, and the methodological quality of this body of work is unclear.

OBJECTIVE

This study aimed to review the application of geospatial analysis in research on violence against children (VAC) and evaluate how essential methodological aspects are reported.

METHODS

Twelve databases were searched for studies on VAC using geospatial techniques. Two independent reviewers screened the papers for eligibility. Findings were narratively synthesised.

RESULTS

Sixty studies were included. Six studies estimated the prevalence of VAC and 54 investigated the associations between VAC and covariates. Most studies were conducted in the US (68 %), and the broad definition of 'child maltreatment' (53 %) was the most common form of violence explored. Most studies (83 %) used administrative data, whereas 23 % used an ecological study design to estimate the associations between risk factors and official reports of VAC. Frequentist modelling approaches were used in 54 % of the studies, and 47 % investigated VAC at census tract level. Model fit metrics were reported in 69 % of studies.

CONCLUSIONS

Current knowledge of the geographical distribution of VAC is severely limited because of the reliance on administrative data, which vastly underestimates the prevalence of VAC compared with self-reports and poor reporting of quality characteristics. There is a huge opportunity for applying geospatial methods in VAC research in diverse geographic contexts. Future research must adopt rigorous and standardised approaches to model fitting and validation and make better use of self-reported data.

The neighborhood deprivation gradient and child physical abuse and neglect: A Bayesian spatial model

BACKGROUND

Child abuse and neglect is a public health priority due to its long-term maladaptive consequences. No study in the USA has assessed the nature and magnitude of the social deprivation effect on substantiated child maltreatment risk.

OBJECTIVES

To examine linear and non-linear relationships between area level deprivation and the log-risk of both substantiated physical abuse and neglect while accounting for spatial and heterogeneous random effects.

METHODS

Substantiated child maltreatment and population data (2008-2015) were aggregated to neighborhoods in Bernalillo County, New Mexico. The contribution of area level deprivation to the geographical variation in the log-risks of substantiated child physical abuse and neglect was modeled using Bayesian spatial regression.

RESULTS

Forty-three percent and 46.4 % of the 153 neighborhoods recorded greater risk for either substantiated physical abuse or neglect compared to the county average. The most deprived 20 % of neighborhoods had 71 % and 72 % more cases of substantiated physical abuse and neglect, respectively, than would be expected if the substantiations were randomly distributed throughout the county. Area level deprivation explained 47 % of the variation in substantiated physical abuse and 51 % of the variation in substantiated neglect after controlling for both spatial autocorrelation and heterogeneity.

CONCLUSIONS

Implications from this study can be used to quantify disparities in substantiated child maltreatment attributed to regional differences in social deprivation and to identify priority areas for intervention.

Systematic braiding of Smoke-Free Home SafeCare to address child maltreatment risk and secondhand smoke exposure: findings from a pilot study

BACKGROUND

Exposure to secondhand tobacco smoke (SHS) and child maltreatment are preventable threats to child health. Few evidence-based interventions target both SHS and child maltreatment risk. The purpose of this paper is to describe the systematic braiding process of two evidence-based programs to address child SHS in the home and maltreatment perpetration risk, and present results from the formative work and pilot study.

METHODS

The first 4 steps of the systematic braiding process were completed, including the following: (1) the identification of core elements of both programs, (2) the development of an initial draft of the braided curriculum (Smoke-Free Home SafeCare - SFH-SC), (3) an acceptability and feasibility pilot of SFH-SC with caregivers of young children who reported a smoker living in the home (N = 8), and (4) feedback collection on the braided curriculum from SafeCare Providers (N = 9).

RESULTS

Experts identified common pedagogical and theoretical underpinnings for the two programs and braided Smoke-Free Homes: Some Things Are Better Outside into two SafeCare modules. Caregiver feedback from the pilot demonstrated that participants were engaged with SFH-SC and felt supported and comfortable discussing SHS intervention content with the SFH-SC Provider. Caregiver self-reports indicated a slight increase in smoke-free home rules from baseline to follow-up and a notable reduction in parent stress on the Parent Stress Index of 5.9 points (SD = 10.2). SafeCare Provider feedback following intensive review of the curriculum indicated high feasibility for SFH-SC delivery.

CONCLUSIONS

Parent and Provider findings suggest SFH-SC is a viable intervention that has potential to reduce the public health impact of SHS and child maltreatment for at-risk families.

PROTOCOL

The protocol for the pilot is not published elsewhere; however, the full protocol for the hybrid trial can be found here: https://clinicaltrials.gov/ct2/show/NCT05000632 .

TRIAL REGISTRATION

NCT, NCT05000632. Registered 14 July 2021, there is not a separate registration number for the pilot.

The Impact of COVID-19 on the Quality of Life and Happiness of Care Home Residents in Croatia: A Cross-Sectional Study

Care/nursing homes globally have been severely affected by the COVID-19 pandemic and have disproportionately experienced a high rate of mortality which led to the introduction of strict isolation policies. However, while there are studies on the mortality, epidemiology, staffing challenges, and mismanagement in long-term care homes as a result of COVID-19, there appears to be a paucity of information regarding the Quality of Life (QoL), happiness, and associated well-being of the elderly residents of these homes. Therefore, we examined if COVID-19 affected the happiness level, QoL, and financial condition of long-term care home residents in Croatia. To achieve this, a survey of 308 participants in eight long term care homes was conducted. Descriptive analysis was performed to describe the mean of all responses and the Bayesian Integrated Nested Laplace Approximation (INLA) was used to provide a detailed quantitative analysis of the results. We found that the QoL and happiness of residents remained relatively stable during the COVID-19 pandemic. However, the income level, financial outlook, marital status, and vaccination positivity influenced the QoL and happiness of care home residents to a considerable degree. We recommend that policy makers pay attention to these underlying factors.

Contrasting Methods of Measurement in Spatial Analyses Examining the Alcohol Environment and Child Maltreatment

Child physical abuse is a major public health issue in the United States. Environmental child welfare research has focused on neighborhood characteristics and the influence of alcohol and marijuana establishments. To our knowledge, child welfare studies have singularly examined the outcome in terms of victims, that is, at the level of child population, and have not considered the parent population. Thus, in this exploratory study, we use spatial scan statistics to analyze patterns of child physical abuse at the child and household level, and we use Bayesian hierarchical spatial conditional autoregressive models to determine the relative influence of alcohol availability and other environmental factors. We find that household clusters are nested in child clusters and that controlling for alcohol establishments reduces cluster size. In the Bayesian regression models, alcohol availability increased risk slightly, while neighborhood diversity (measured using Blau's Index) elevated risk considerably. Immediate implications for child welfare agencies are discussed.

Association between child welfare reporting rates and the developmental vulnerability of kindergarten children at the neighborhood level

BACKGROUND

In Canada, more than one in four children are considered vulnerable in at least one domain of development when they enter kindergarten. Recent studies have suggested that this ratio is higher among those who were previously maltreated. However, little is known about this associations at the neighborhood level, although it may be an interesting way to identify risk areas and highlight child welfare system data to prevent public health issues.

OBJECTIVE

Using the census tract as the unit of measurement, this study examines the association between the proportions of vulnerable children in different domains of development upon entering kindergarten, and four indicators of child maltreatment (CM) among 0-5 year olds.

PARTICIPANTS, SETTING AND METHOD

This study is based on the secondary use of data from a survey on the development of kindergartners carried out in 2017, combined with data from child welfare records for that same year. The data have been aggregated on the basis of 759 census tracts located in four health regions of Quebec, Canada.

RESULTS

The results of spatial regression analyses show that all indicators of maltreatment are positively and significantly associated with each indicator of developmental vulnerability. The size of these associations varies according to the indicators used (β = 0.192, p < .05 to β = 1.587, p < .001).

CONCLUSION

The results highlight the link between CM and school readiness, including potential externalities on neighborhood children. Neighborhoods at high risk of maltreatment could help identify areas with high rate of vulnerable children in early childhood.

Accessibility to rabies centers and human rabies post-exposure prophylaxis rates in Cambodia: A Bayesian spatio-temporal analysis to identify optimal locations for future centers

Rabies is endemic in Cambodia. For exposed humans, post-exposure prophylaxis (PEP) is very effective in preventing this otherwise fatal disease. The Institut Pasteur du Cambodge (IPC) in Phnom Penh was the primary distributor of PEP in Cambodia until 2018. Since then, and to increase distribution of PEP, two new centers have been opened by IPC in the provinces of Battambang and Kampong Cham. Data on bitten patients, who sometimes bring the head of the biting animal for rabies analyses, have been recorded by IPC since 2000. However, human cases are not routinely recorded in Cambodia, making it difficult to establish a human burden of disease and generate a risk map of dog bites to inform the selection of future PEP center locations in high-risk areas. Our aim was to assess the impact of accessibility to rabies centers on the yearly rate of PEP patients in the population and generate a risk map to identify the locations where new centers would be the most beneficial to the Cambodian population. To accomplish this, we used spatio-temporal Bayesian regression models with the number of PEP patients as the outcome. The primary exposure variable considered was travel time to the nearest IPC center. Secondary exposure variables consisted of travel time to a provincial capital and urban proportion of the population. Between 2000 and 2016, a total of 293,955 PEP patient records were identified. Our results showed a significant negative association between travel time to IPC and the rate of PEP patients: an increase in one hour travel time from the living location to IPC PEP centers leads to a reduction in PEP rate of 70% to 80%. Five provinces were identified as the most efficient locations for future centers to maximize PEP accessibility: Banteay Meanchey, Siem Reap, Takeo, Kampot and Svay Rieng. Adding a PEP center in every provincial capital would increase the proportion of Cambodians living within 60 minutes of a PEP center from 26.6% to 64.9%, and living within 120 minutes from 52.8% to 93.3%, which could save hundreds of lives annually.

Rabies is a fatal viral disease that affects the nervous system. It is endemic in many countries in Africa and Asia where free roaming dogs form a reservoir. Transmission to humans occurs most often through a dog bite. However, post-exposure prophylaxis (PEP), if administered before symptom onset, is highly effective at preventing the disease. In Cambodia, a few number of centers offer PEP, with the Institut Pasteur du Cambodge in Phnom Penh being the main one. These few locations lead to limited accessibility for rural areas distant from Phnom Penh and underestimations of the dog-bite burden and PEP needs. Through statistical modelling, we measured the impact of accessibility on the number of PEP patients and predicted the impact of opening new centers in other locations. We found that travel time was significantly associated with the rate of PEP patients. IPC opened new rabies centers in Battambang and Kampong Cham provinces in 2018 and 2019, respectively, and we identified four provinces where future openings would be the most beneficial: Banteay Meanchey, Siem Reap, Takeo, Kampot and Svay Rieng. This study is part of a broader drive to eradicate rabies in Cambodia by 2030 through increased PEP infrastructure and control measures in the dog population.

Temporal and Spatial Heterogeneity of PM2.5 Related to Meteorological and Socioeconomic Factors across China during 2000-2018

In recent years, air pollution caused by PM_2.5 in China has become increasingly severe. This study applied a Bayesian space-time hierarchy model to reveal the spatiotemporal heterogeneity of the PM_2.5 concentrations in China. In addition, the relationship between meteorological and socioeconomic factors and their interaction with PM_2.5 during 2000-2018 was investigated based on the GeoDetector model. Results suggested that the concentration of PM_2.5 across China first increased and then decreased between 2000 and 2018. Geographically, the North China Plain and the Yangtze River Delta were high PM_2.5 pollution areas, while Northeast and Southwest China are regarded as low-risk areas for PM_2.5 pollution. Meanwhile, in Northern and Southern China, the population density was the most important socioeconomic factor affecting PM_2.5 with q values of 0.62 and 0.66, respectively; the main meteorological factors affecting PM_2.5 were air temperature and vapor pressure, with q values of 0.64 and 0.68, respectively. These results are conducive to our in-depth understanding of the status of PM_2.5 pollution in China and provide an important reference for the future direction of PM_2.5 pollution control.

Child Maltreatment, School Bonds, and Adult Violence: A Serial Mediation Model

Physically abused youth are vulnerable to experiencing difficulties across multiple domains of school functioning. Most of the literature examining the relationship between child physical abuse (CPA) and adult violence has focused narrowly on academic outcomes rather than taking a broader view that explores the processes undergirding school engagement and connections. The present study adopted Connell's model of school engagement, connectedness and outcomes within a social bond framework to examine (a) the link between CPA and school social bonds, (b) the link between CPA and adult violence persistence, and (c) the mediational (parallel, serial) effects of school bonds (engagement, connection, and achievement) on violence perpetration in adulthood. Consistent with previous research, results indicated that children who experience physical abuse have poorer academic performance, which, in turn, is related to future violent trajectories. We further found that the relationship between CPA and violence persistence is mediated by a process of bonding to school that begins with being actively engaged in school activities and ends with higher levels of academic achievement. In particular, some of the "school achievement" effect found in previous research operates through behavioral and emotional manifestations and may be partly explained through physically abused children's lessened ability to be engaged with and connected to school activities. We conclude with a discussion of the policy implications stemming from our findings.

A spatiotemporal analysis of the impact of COVID-19 on child abuse and neglect in the city of Los Angeles, California

BACKGROUND

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2; COVID-19) has created an urgent need to identify child abuse and neglect (CAN) and efficiently allocate resources to improve the coordination of responses during a public health crisis.

OBJECTIVE

To provide unique insights into the spatial and temporal distribution of CAN in relation to COVID-19 outcomes and identify areas where CAN has increased or decreased during the pandemic.

PARTICIPANTS

Children under 18 years old reported to the Los Angeles Police Department for CAN.

SETTING

CAN incidents in the city of Los Angeles.

METHODS

Negative binomial regression was used to explore associations between the implementation of social distancing protocols and reported CAN during COVID-19. Spatiotemporal analysis identified locations of emerging hot and cold spots during the pandemic. Associations between neighborhood structural factors (e.g., school absenteeism, poverty, unemployment, housing insecurity and birth assets) and hot and cold spot patterns were explored.

RESULTS

There was a statistically significant decline in reports of CAN during the COVID-19 pandemic but no significant trends following the implementation of social distancing measures (e.g. safer at home orders, school closures). Compared to consecutive cold spots, severe housing burden, the number of assets children have at birth, poverty, school absenteeism and labor force participation were significantly associated with new and intensifying hotspots of CAN during the COVID-19 pandemic.

CONCLUSIONS

Our findings reinforce the utility of developing intervention strategies that minimize harm to children by targeting resources to specific challenges facing families enduring the COVID-19 experience.

The Spatial Overlap of Police Calls Reporting Street-Level and Behind-Closed-Doors Crime: A Bayesian Modeling Approach

Traditionally, intimate-partner violence has been considered a special type of crime that occurs behind closed doors, with different characteristics from street-level crime. The aim of this study is to analyze the spatial overlap of police calls reporting street-level and behind-closed-doors crime. We analyzed geocoded police calls in the 552 census-block groups of the city of Valencia, Spain, related to street-level crime (N = 26,624) and to intimate-partner violence against women (N = 11,673). A Bayesian joint model was run to analyze the spatial overlap. In addition, two Bayesian hierarchical models controlled for different neighborhood characteristics to analyze the relative risks. Results showed that 66.5% of the total between-area variation in risk of reporting street-level crime was captured by a shared spatial component, while for reporting IPVAW the shared component was 91.1%. The log relative risks showed a correlation of 0.53, with 73.6% of the census-block groups having either low or high values in both outcomes, and 26.4% of the areas with mismatched risks. Maps of the shared component and the relative risks are shown to detect spatial differences. These results suggest that although there are some spatial differences between police calls reporting street-level and behind-closed-doors crime, there is also a shared distribution that should be considered to inform better-targeted police interventions.

Spatiotemporal heterogeneity of PM2.5 and its relationship with urbanization in North China from 2000 to 2017

Fine particulate matter (PM_2.5) pollution is becoming an increasing global concern due to rapid urbanization and socioeconomic development, especially in North China. Although North China experiences poor air quality and high PM_2.5 concentrations, their spatial heterogeneity and relationship with the relative spatial risks of air pollution have not been explored. Therefore, in this study, the temporal variation trends (slope values) of the PM_2.5 concentrations in North China from 2000 to 2017 were first quantified using the unitary linear regression model, and the Bayesian space-time hierarchy model was introduced to characterize their spatiotemporal heterogeneity. The spatial lag model was then used to examine the determinant power of urbanization and other socioeconomic factors. Additionally, the correlation between the spatial relative risks (probability of a region becoming more/less polluted relative to the average PM_2.5 concentrations of the study area), and the temporal variation trends of the PM_2.5 concentrations were quantified using the bivariate local indicators of spatial association model. The results showed that the PM_2.5 concentrations increased during 2000-2017, and peaked in 2007 and 2013. Spatially, the cities at high risk of PM_2.5 pollution were mainly clustered in southeastern Hebei, northern Henan, and western Shandong where the slope values were low, as demonstrated by the value of Moran's I (-0.56). Moreover, urbanization and road density were both positively correlated with PM_2.5 pollution, while the proportion of tertiary industry was negatively correlated. Furthermore, a notable increasing trend was observed in some cities, such as Tianjin, Zaozhuang, Qingdao, and Xinyang. These findings can contribute to the development of effective policies from the perspective of rapid urbanization to relieve and reduce PM_2.5 pollution.

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

BACKGROUND

Modeling the spatio-temporal characteristics of substantiated child maltreatment risk has significant implications for child welfare policy.

OBJECTIVE

This study quantifies the spatiotemporal risk of child abuse and neglect in New Mexico at the census tract level over 9 years, identifies areas of increased risk, and evaluates the role of multiple measures of social and housing insecurity on substantiated child maltreatment referrals.

PARTICIPANTS AND SETTING

Child maltreatment substantiation data across 499 census tracts from 2007 to 2015 were obtained from the New Mexico Department of Public Health.

METHODS

Substantiated referral counts were analyzed within census tracts with Bayesian hierarchical space-time models using Laplace approximation. Standardized incidence ratios, spatial risk, and probability exceedances were calculated and mapped.

RESULTS

Multiple neighborhood structural factors were associated with an increased risk of substantiated child maltreatment, including the eviction rate (Incidence Density Ratio [IDR] = 1.09 [95 % CrI = 1.01-1.12]), rent burden (IDR = 1.11 [95 % CrI = 1.01-1.13]), urban tracts (IDR = 1.36 [95 % CrI = 1.05-1.77]), food desert tracts (IDR = 1.21 [95 % CrI = 1.04-1.41]), low income tracts (IDR = 1.27 [95 % CrI = 1.09-1.49]), percent of households with no vehicle access ([IDR] = 1.27 [95 % CrI = .247-6.47]), and percent of persons with a disability (IDR = 1.05 [95 % CrI = 1.03-1.06]). The racial/ethnic diversity ratio, however, was associated with lower incidence of child maltreatment allegation risk (IDR = .988 [95 % CrI = .982-.995]).

CONCLUSIONS

Population-based child abuse and neglect prevention and intervention efforts should be aided by the characteristics of neighborhoods that demonstrate strong spatial patterns of household and housing vulnerability, particularly in low income, racially segregated neighborhoods.

Neighborhood inequality in the prevalence of reported and substantiated child maltreatment

BACKGROUND

Prior research documents spatial concentration in the incidence of child maltreatment reported to and confirmed by Child Protective Services (CPS), but without estimates of the prevalence of such reports, the extent of CPS contact in different communities is unknown.

OBJECTIVE

To estimate the prevalence of CPS reports during early childhood and substantiated investigations during childhood for children living in different types of neighborhoods.

PARTICIPANTS AND SETTING

Children who experienced CPS reports and substantiated investigations in Connecticut.

METHODS

This study uses synthetic cohort life tables to estimate the cumulative risk of CPS reports before age five and substantiated CPS investigations before age 18, by neighborhood poverty rate and neighborhood racial composition.

RESULTS

The analysis reveals substantial stratification in the prevalence of CPS contact by the demographic characteristics of children's residential neighborhoods. For example, while 7% of children in low-poverty neighborhoods (under 10% poor) experience a substantiated CPS investigation at some point during childhood at 2014 and 2015 rates, this risk more than doubles to 17% for their peers in moderate-poverty neighborhoods (10-20% poor) and more than triples to 26% for their peers in high-poverty neighborhoods (over 20% poor). Similar trends emerge when examining CPS reports in early childhood as well as when comparing neighborhoods with different proportions of White residents.

CONCLUSIONS

CPS reports and substantiated investigations are a widespread and disproportionately experienced life event for children in poor neighborhoods and children in non-White neighborhoods.

A Secondary Spatial Analysis of Gun Violence near Boston Schools: a Public Health Approach

School neighborhood violence continues to be a major public health problem among urban students. A large body of research addresses violence at school; however, fewer studies have explored concentrations of violence in areas proximal to schools. This study aimed to quantify the concentration of shootings near schools to elucidate the place-based dynamics that may be focal points for violence prevention. Geocoded databases of shooting and school locations were used to examine locational patterns of firearm shootings and elementary, middle, and high schools in Boston, Massachusetts. Analyses utilized spatial statistics for point pattern data including distance matrix and K function methodology to quantify the degree of spatial dependence of shootings around schools. Results suggested that between 2012 and 2015, there were 678 shooting incidents in Boston; the average density was 5.1 per square kilometer. The nearest neighbor index (NNI = 0.335 km, p < .001, O = 0.95 km, E = 0.28 km) and G function analysis revealed a clustered pattern of gun shooting incidents indicative of a spatially non-random process. The mean and median distance from any school to the nearest shooting location was 0.35 and 0.33 km, respectively. A majority (56%, 74/133) of schools in Boston had at least one shooting incident within 400 m, a distance that would take about 5 min to walk if traveling by foot. The bivariate K function indicated that a significantly greater number of shootings were clustered within short distances from schools than would be expected under a null hypothesis of no spatial dependence. Implications for students attending schools in racially homogenous neighborhoods across all income levels are discussed.