Background

Child maltreatment is a pervasive public health problem in the United States (US).1 Comprised of acts of commission and omission by a parent or other caregiver (e.g., physical abuse, sexual abuse, and various forms of neglect),2 child maltreatment is a substantial cause of pediatric injury and death. In 2018, nearly 700,000 childhood victims of nonfatal maltreatment were identified, and an estimated 1770 children died.1 The combined human and institutional cost attributed to maltreatment morbidity and mortality in the US is estimated to be $124 billion annually.3

The World Health Organization defines social determinants of health (SDH) as “the conditions in which people are born, grown, work, live, and age, and the wider set of forces and systems shaping the conditions of life.”4 These conditions are shaped by the distribution of resources, and connect facets of the physical, social, and built environment associated with health outcomes.5 Among the most commonly recognized SDH (economic stability, education, neighborhood and built environment, health and healthcare, and social and community context),6 poverty is a major and often overarching factor. Poverty also has been identified as a known risk factor for child maltreatment.7 Thus, identifying how poverty and other SDH are associated with child maltreatment is a necessary step to develop effective interventions for maltreatment prevention and treatment, and mitigating the risk of associated physical and psychological injury.

Not enough is known about the association of SDH with child maltreatment. Four published systematic reviews have included analyses that examined the relationship between a single or two SDH and maltreatment. Two included socioeconomic status,8,9 one included socioeconomic status and parental educational attainment,10 and the fourth included immigration status.11 No published systematic reviews (to our knowledge), however, have examined the spectrum of SDH with respect to their associations with child maltreatment. Therefore, the aim of this study was to conduct a systematic review of the associations of SDH (including poverty, housing insecurity, food insecurity, uninsurance, healthcare access, and transportation) with child maltreatment.

Methods

Inclusion criteria

Studies were selected using the PICOS approach for inclusion and exclusion.12,13 The a priori inclusion criteria for studies were as follows: (1) English-language studies, (2) children 0–18 years old living in the US, (3) peer-reviewed, (4) observational and experimental designs, (5) outcome measures reported for at least one form of maltreatment, and 6) exposure measures for at least one SDH. The exclusion criteria were: (1) specific SDH could not be identified, and (2) conference presentations (e.g., abstracts, posters, or oral presentations).

The outcome of interest was child maltreatment, defined by the Child Abuse Prevention and Treatment Reauthorization Act of 2010 as “at a minimum, any recent act or failure to act on the part of a parent or caretaker, which results in death, serious physical or emotional harm, sexual abuse or exploitation, or an act or failure to act which presents an imminent risk of serious harm.”2 Included studies were assessed for the associations of selected SDH—including poverty, food insecurity, housing instability, parental educational attainment, child uninsurance, transportation barriers, and access barriers to healthcare—with child maltreatment. These SDH were chosen because they are domains hypothesized to be most likely associated with child maltreatment and were addressed in a recently published SDH screening instrument used for testing interventions effective in reducing SDH and improving child and caregiver health.14 Immigration status was not included because of the recent publication of a systematic review examining the association of this SDH with child maltreatment.11

Data sources

Five data sources were searched through March 2020: (1) PubMed, (2) Web of Science Core Collection, (3) SCOPUS, (4) JSTORE, and (5) the Social Intervention Research and Evaluation Network Evidence Library. All searches contained the following terms: (“Child Abuse”[Mesh] OR “child abuse”[tw] OR “child maltreatment”[tw] OR “child mistreatment”[tw] OR “child neglect”[tw]) AND (“Social Determinants of Health”[Mesh] OR “social determinants of health”[tw] OR “social class”). Searches for terms related to specific SDH varied. A sample search strategy (SCOPUS) can be found in Supplementary Table S1 (online).

An effort-to-yield measure of search precision, number needed to read (NNR) was calculated by taking the inverse of the precision of the searches. Precision was calculated by dividing the number of included studies by the number of screened studies, after removal of duplicates. NNR quantifies the number of articles that would be needed to be read before finding one that meets the established inclusion criteria. Dependent on the subject and inclusion criteria, this number provides insights into the time and resources needed for replication, or to conduct a similar study.

Selection of studies

All studies were stored on a Microsoft Excel document detailing the reasons for inclusion or exclusion.

Data abstraction

A codebook was developed using Microsoft Excel. Variables included study characteristics (year of publication, study design and population size, duration, data sources, and level[s] of analysis), sociodemographic characteristics of the study population (child age, racial composition, and sex), SDH under investigation, child maltreatment type (sexual, physical, psychological, neglect, multiple forms, and other), and measures of study quality.

Study quality

A modified version of the Downs and Black checklist was used to assess study quality (Supplementary Table S2).15 Each item was scored as no (0) or yes (1). The sum of all items ranged from 1 to 8, with higher scores representing a lower risk of bias.

Data synthesis

The criteria for SDH and definitions of child maltreatment varied by study. Therefore, we were unable to combine endpoints in a meta-analysis. Data synthesis at the level of the individual, family, and community were used to analyze included studies.

Study registration

The study protocol was registered with PROSPERO (CRD42020166969).

Results

Study characteristics

Our initial search yielded 3441 studies. After screening by titles and abstracts, 118 met the initial inclusion criteria. Following a full review of 118 studies, 33 were included in the final analysis. The process for selecting included studies is presented in Fig. 1. Search precision was 0.0096 and the NNR was 104. The characteristics of included studies are presented in Table 1. Included studies were published from 1978 to 2020.16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48 Nine studies used national data,16,23,27,30,36,38,42,43,47 and the remaining studies used data from individual states, including 14 from the Midwest,17,18,19,20,25,26,28,29,31,33,39,40,41,45 four from the South,21,22,32,37 four from the Northeast,34,35,46,48 one from the West (California),24 and one from the Pacific (Alaska).44 Of these studies, 5 conducted chart reviews, 7 used cohort study designs, 7 used a cross-sectional design, and 14 conducted ecological analyses. Included studies assessed the relationship between SDH and child maltreatment at the levels of the individual, zip code, county, and census tracts.

Fig. 1: Flow diagram of study selection.
figure 1

aStudies may have been excluded for multiple reasons.

Full size image
Table 1 Characteristics of included studies (n = 33).
Full size table

Study outcomes

Poverty

Twenty-nine studies explored the association of poverty with child maltreatment.16,17,18,19,20,21,22,23,24,25,26,27,28,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,47 Poverty was found to be consistently and strongly associated with maltreatment, with all but three studies identifying a significant association between either familial or community-level poverty and child maltreatment.16,18,21 Across studies, poverty was defined by county,45 neighborhood,41 familial/household income,17,18,19,20,23,28,41,42,43 socioeconomic status,44 poverty rate,21,27,35,40 unemployment,16,17,21,31,32,34,36,40 percentage of families living below the federal poverty level,24,28,31,32,33,36,37,38,39 children living in poverty,17,47 receipt of public assistance,19,25,31,40 composite impoverishment scores,26 and self-reported acute financial challenges.22

In some studies, the relationship between poverty and maltreatment differed by abuse type. For example, one study found that neighborhood poverty was associated with all three forms of child maltreatment, but to different degrees.38 Another study indicated that financial problems were strongly associated with neglect and abandonment, but the association was less pronounced for sexual abuse.21

Associations between poverty and maltreatment varied by race/ethnicity. A study comparing predominantly white and black neighborhoods found that the association between poverty and child maltreatment was strongest in whites.25 Research linking multiple sources of data showed that black children living in poverty were twice as likely to be reported for needs-based neglect than their white counterparts.26 A recent study showed that when income was held constant, white race was strongly associated with both sexual abuse and neglect, and black race was associated with physical abuse.27

Housing instability

Thirteen studies examined the relationship between housing instability and child maltreatment.16,18,20,21,23,26,28,29,32,33,34,40,46 Most studies revealed that housing instability is associated with child maltreatment. Among these studies, the definition of housing stability varied, and included percent vacancy,21,26,32,33,40 rates of foreclosure and delinquency,16,18,34 hazardous living conditions,29 and instability/mobility (>1 move per year).20,23,28 Only one study examined homelessness, performing an analysis of hospital and pediatric ambulatory records of children <18 years old.46 After matching families on income, homeless children were found to have higher rates of maltreatment-related emergency-department (ED) visits and child maltreatment than their nonhomeless counterparts. One study found that displacement due to foreclosure, eviction, or mortgage delinquency was associated with maltreatment investigations.34 Two studies documented that housing instability/mobility (>1 move per year) was associated with child protective service (CPS) reports and maltreatment risk.20,23

Two studies found no association between housing insecurity and child maltreatment.18,28 In the first, housing instability consisted of an aggregate measure of material hardship, including difficulty paying rent, eviction, or having experienced any utility shutoff in the previous year.18 In the second, housing instability was measured by residential mobility.28

Several studies reported differences in the association between housing stability and child maltreatment type. Two identified an association between the percent of vacant housing in communities and sexual abuse.21,32 Another study found that hazardous housing conditions were associated with neglect, but not physical abuse; a history of housing instability increased the strength of this association.29 One study found that mortgage delinquency was associated with traumatic brain injury and other forms of physical abuse.20

Food insecurity

Just one study examined the relationship between food insecurity and child maltreatment.30 An analysis of a national sample from the Fragile Families and Childhood Wellbeing Study revealed that, compared with food-secure households, food-insecure households experienced increased rates of total parental aggression (7% vs. 20%, respectively). Controlling for maternal characteristics did not attenuate this association.

Parental educational attainment

Eight studies considered the relationship between parental educational attainment and child maltreatment.17,18,20,24,25,32,41,42 The results of most studies indicate that low parental educational attainment is associated with child maltreatment. Parental educational attainment was defined as high-school completion in six studies,17,18,20,32,41,42 maternal education level in one,25 and completion of postsecondary education in the last.24 Two studies found no association.18,24 Notably, one of these studied failed to report victim and perpetrator demographic characteristics (age, sex, or race/ethnicity),18 and the other relied on self-reported data.24

Uninsurance

One study was identified that examined the association of the child lacking health insurance with child maltreatment.48 This study reported that a higher proportion of preadolescent children seen in the ED with suspected sexual child abuse were uninsured, compared with a control group of children seen in the ED with upper-limb fractures, at 52% vs. 1%, respectively. No statistical analyses, however, were conducted, nor is it clear whether there was matching of cases and controls by age, sex, or other relevant characteristics.

Other SDH

The search did not reveal any studies that examined the associations of transportation or access to healthcare with child maltreatment.

Discussion

Multiple studies document that SDH, including poverty, housing instability, food insecurity, low parental educational attainment, and child uninsurance, are significantly associated with child maltreatment. A recent systematic review also concluded that although the immigrant parental status is associated with a lower likelihood of overall child maltreatment, it may be associated with a higher risk of child neglect and neglectful supervision.11 Taken together, these findings suggest that an urgent priority, therefore, should be to routinely screen families for SDH in inpatient and outpatient settings and in CPS, and to address identified SDH with referrals to appropriate services. This screening and referral process could have the potential to not only prevent child maltreatment by reducing or eliminating the SDH before they result in maltreatment, but might also decrease the risk of maltreatment recidivism in families in which maltreatment already has occurred. The American Academy of Pediatrics, American Academy of Family Physicians, and the National Academy of Sciences, Engineering, and Medicine all have endorsed SDH screening and service referral.49,50,51 Several studies document that patients and caregivers are comfortable with completing SDH screening.52,53,54,55,56 Addressing SDH by referral to such services as case managers, social workers, housing vouchers, medical–legal partnerships, and parent mentors, already has been shown to reduce hospitalizations, improve housing quality and stability, enhance economic security, improve healthcare outcomes, insure more uninsured children, increase the quality of care, empower parents, and save money for society,57 thereby holding great promise as interventions that may prove effective in ultimately reducing or preventing child maltreatment.

Poverty was the SDH for which the greatest number of studies documented an association with child maltreatment. Although few studies have investigated the temporal relationship between poverty and child maltreatment,8 there is evidence that families living in poverty are more likely to be reported to CPS for neglect.58 Poverty sequelae, such as inability to feed, clothe, or house a child, overlap with the definition of child neglect, so it is important to distinguish intentional neglect from family challenges related to living in poverty. Differential or alternative response is one CPS approach that addresses maltreatment reports by attending to unmet family needs.59 An analysis of the effectiveness of this form of intervention has shown that families living in poverty benefit most from this approach.60 To date, this response has been implemented at the individual and family levels. Extending differential or alternative response to the community level may be an effective strategy for families living in impoverished neighborhoods, where racial biases in child maltreatment reports and investigations have been identified.

The study results underscore several unanswered questions regarding the association between SDH and child maltreatment. First, it is unclear whether transportation barriers or impaired access to healthcare are associated with child maltreatment, given that no studies were identified on these SDH. Second, because the definitions for each SDH varied considerably within and across studies (especially for poverty), it is unclear whether more consistent SDH definitions would yield different findings. Third, because males as caregivers and heads of household were under-represented and often excluded from some study populations,20,23,25,33 an unanswered question is whether there are associations of paternal educational attainment and other male-caregiver SDHs with child maltreatment. Although single mothers have been identified as an at-risk population for maltreatment perpetration, it is equally important to examine the role that men play in maltreatment. In a previous analysis, the first author identified men as the predominant perpetrator in 58% of cases of fatal maltreatment in the US.61 Results of our study emphasize the need for research inclusive of male caregivers, to identify and mitigate risk factors before they escalate to maltreatment fatalities. Fourth, most studies focusing on sexual abuse were primarily limited to female populations,32 despite evidence that male children also are victims of sexual abuse. There is an urgent need to investigate how SDH perpetuate or protect against sexual abuse in male children, so that prevention efforts can be tailored by sex. Finally, because most studies combined maltreatment into one aggregate category, an unanswered question is what are the associations of SDH with specific maltreatment categories. It has been posited that each maltreatment type has a unique etiology, and lumping these types into one category likely attenuates the ability to identify meaningful associations. Although few studies in this systematic review disaggregated by maltreatment categories, those that did found significant differences in maltreatment risk according to the SDH examined.

Based on the study findings, a research agenda is proposed to address key issues regarding the association of SDH with child maltreatment. Research is needed to address the aforementioned identified research gaps, including studies on transportation barriers, impaired access to healthcare, consistently defined SDH, SDH for male caregivers, and the associations of SDH with specific maltreatment categories and male victims of sexual abuse. Studies are needed to determine whether there is a direct association between the number of SDH and the risk of maltreatment, and whether the presence of multiple SDH can synergistically increase maltreatment risk. Research is urgently needed to determine whether SDH screening and referral to appropriate services result in SDH reduction and elimination as well as decreases in or the prevention of child maltreatment and maltreatment recidivism.

Limitations and strengths

Certain study limitations should be noted. First, as with all systematic reviews, the quality of this analysis is limited by the scientific rigor of included studies. Second, studies were selected based on the search criteria. It is possible that relevant literature was missed because of the heterogeneity of terms used to describe the various SDH and child maltreatment. Third, many included studies were cross-sectional or ecological, preventing the ability to draw conclusions about the temporal relationship between SDH and child maltreatment. Fourth, many data sources for the included studies used administrative data derived from CPS. In most instances, these records only included reports of maltreatment that were screened in and accepted for either an investigation or alternative response. As a result, these data sources likely exclude many cases of maltreatment, given evidence demonstrating equivalent risk of incidence and recurrence between maltreatment reports and substantiations.28,38

Conclusions

SDH, including poverty, parental educational attainment, housing instability, food insecurity, and uninsurance, are associated with child maltreatment. These findings suggest that an urgent priority should be routinely screening families for SDH, with referrals to appropriate services, a process that could have the potential to prevent both child maltreatment and subsequent recidivism. Unanswered questions include whether SDH are associated with specific maltreatment categories and male victims of sexual abuse, and whether transportation barriers, impaired access to healthcare, consistently defined SDH, and SDH for male caregivers are associated with child maltreatment. A proposed research agenda includes addressing these unanswered questions; determining whether there is a direct association between the number of SDH and the risk of maltreatment, and whether the presence of multiple SDH can synergistically increase maltreatment risk; and investigations on whether SDH screening and referral to appropriate services result in SDH reduction and elimination, as well as decreases in or the prevention of child maltreatment and maltreatment recidivism.