What Does Child Protective Services Investigate as Neglect? A Population-Based Study

Abstract

Most child protective services (CPS) investigations involve allegations of neglect. Broad and vague definitions have led to concerns that CPS-investigated neglect is driven by poverty-based material hardship. In a representative sample of 295 neglect investigations in California in 2017, structured data and narrative text fields were used to characterize the types of neglect and concurrent parental risk factors investigated by CPS and to assess the rate and nature of investigated physical neglect, defined as inadequate food, housing, or hygiene. The most common types of neglect were inadequate supervision (44%) and failure to protect (29%), followed by physical neglect (14%). Common risk factors identified in neglect investigations were parental substance use (41%), domestic violence (21%), mental illness (18%), and co-reported physical or sexual abuse (29%). Nearly all investigations of physical neglect (99%) included concerns related to substance use, domestic violence, mental illness, co-reported abuse or an additional neglect allegation (i.e., abandonment). Given concerns identified in neglect investigations, economic supports are likely insufficient without an array of behavioral-health supports.

Each year, approximately 3.2% of U.S. children are the subject of a child protective services (CPS) investigation (U.S. Department of Health and Human Services, 2021), with 27%–37% experiencing an investigation by age 18 (Kim et al., 2017; Putnam-Hornstein, Ahn, et al., 2021). A majority of those investigations involve allegations of neglect. In 2019, neglect was identified in 75% of substantiated maltreatment reports (U.S. Department of Health and Human Services, 2021), a substantial increase from 58% in 1999 (U.S. Department of Health and Human Services, 1999). Similarly, neglect is documented as a factor in 61% of foster care entries (U.S. Department of Health and Human Services, 2020). There is widespread concern that CPS mischaracterizes poverty as neglect, resulting in rates of systems involvement that far exceed children’s risk of serious harm (Dettlaff et al., 2020; Milner & Kelly, 2020; Roberts, 2022). Indeed, the belief that CPS systematically investigates and intervenes for reasons of poverty has been asserted by high-profile media outlets (Connolly et al., 2015; Dewan, 2018), the U.S. Children’s Bureau (Milner & Kelly, 2020), major foundations, and various academic experts (Pelton, 2016; Raz & Sankaran, 2019).

Concerns about the conflation of neglect and poverty are not unmerited. Poverty and other economic conditions are strongly associated with both child maltreatment and CPS involvement, and associations are generally stronger for neglect than abuse (Pelton, 2015; Sedlak et al., 2010). A growing body of research suggests that government programs or policies that reduce poverty or enhance modest incomes, particularly for working parents, may reduce some forms of CPS involvement (Berger et al., 2017; Wildeman & Fallesen, 2017; Raissian & Bullinger, 2017). Addressing poverty and material needs has the potential to reduce child maltreatment, perhaps especially neglect. Yet the mechanisms of these effects are not altogether clear and may include changes in parental behaviors, changes in the impact of parental behaviors on child health and safety, or changes in the detection (reporting and labeling) of child maltreatment (Font & Maguire-Jack, 2020b).

Basic statutory definitions of neglect focus on a parent’s or caregiver’s failure to provide for a child’s basic needs, resulting in harm or imminent risk of harm (Rebbe, 2018). A child’s basic needs include both material needs, such as food and shelter, and nonmaterial needs, such as supervision (Child Welfare Information Gateway, 2016). The material needs aspect of neglect definitions is strongly emphasized in assertions that CPS confuses poverty for neglect, even though unmet material needs—commonly referred to as “physical neglect” or “failure to provide”—account for less than 10% of all CPS investigations (Casanueva et al., 2011).

Even when physical neglect is the primary form of neglect investigated by CPS, caution is warranted in assuming that such conditions derive solely from poverty. It is generally believed that parents rarely maliciously deprive their children of food, shelter, or other basic needs to cause harm (Dubowitz & Kobulsky, 2022). Thus, setting aside presumably rare occurrences involving malicious intent, the remaining instances of neglect fall onto a spectrum of involuntary to reckless disregard. Involuntary neglect refers to cases in which a child’s needs are unmet due to poverty alone, such that no voluntary parental acts or omissions contributed to the child’s experience of neglect (unmet basic needs). In several states, involuntary neglect is explicitly exempted from the statutory definition of neglect (Berger & Slack, 2020) and may be exempted in practice or regulation by others. At the other end of this spectrum are scenarios in which parents’ behavior is a proximal cause of the unmet needs, even though the intent of the behavior likely is not to inflict harm on the child. For example, when a parent is incapacitated due to drugs or alcohol and is unable to properly supervise their infant, the parent’s behavior is a proximal cause of unmet needs. Of course, an investigation may be needed to ascertain when a child’s needs were unmet due solely to poverty or involved other contributing factors, such as parental substance abuse (Font & Maguire-Jack, 2020b). Many investigations of neglect, however, may be less clear cut, including when parents face substantial constraints due to limited resources and may make choices in the face of those constraints that result in or exacerbate avoidable risks of harm. Current research has not provided a clear understanding of when unmet material needs in the absence of other family or parental risks lead to investigations of families (Wald, 2022).

Family circumstances commonly noted as contributing to child maltreatment and CPS involvement include parental substance use (Dubowitz et al., 2011; Walsh et al., 2003), parental mental illness (Hammond et al., 2017; Kaplan et al., 2019), and domestic violence (Child Welfare Information Gateway, 2012; Rebbe et al., 2021). These family circumstances are especially concerning for infants and toddlers, who are dependent on their caregivers for basic needs and cannot self-advocate. Substance use creates risk of neglect when caregivers are frequently intoxicated or absent from the home while procuring or using substances (i.e., unable to engage in basic care and supervision) or divert limited resources to purchase substances (i.e., compromising children’s food, housing, or other material needs). Similarly, unmanaged mental illness can manifest in withdrawn, distracted, aggressive, or inconsistent parenting responses and may also affect a parent’s ability to engage in basic care tasks (Oyserman et al., 2000). Last, domestic violence is sometimes categorized as neglectful or emotionally abusive (Child Welfare Information Gateway, 2012; Rebbe et al., 2021), because witnessing violence can induce trauma symptoms similar to those exhibited by victims of abuse (Evans et al., 2008; McTavish et al., 2016). Further, the victimized parent (or both parents, in the case of mutual violence) may be less attentive and emotionally available to their children (Holt et al., 2008). These risks often co-occur, with substance use, emerging as both a precursor and outcome of mental illness, violence, and poverty (Nicholson et al., 2002; Sells et al., 2003; Simon & Brooks, 2017). Further, neglect often co-occurs with abuse, as children are more vulnerable to abuse by others when parental supervision is lacking or compromised, and antecedents of neglect, such as comorbid mental illness and substance use, can also enhance the risk of violent behavior (Volavka & Swanson, 2010). In sum, caregivers’ substance use, mental illness, and domestic violence may be proximal contributors to child neglect—beyond the effects of poverty—thus precipitating a report and CPS investigation.

Research Questions

Because existing CPS data largely categorize neglect into a single category and data on contributing factors such as substance abuse are either unavailable or unreliable, it is difficult to ascertain the concerns that lead to neglect investigations and inform dispositions. Using both structured and narrative text data from a random sample of CPS investigations for allegations of neglect, we explored three questions: (1) What are the most common manifestations or types of neglect reported?; (2) What proportion of CPS investigations classified as neglect document parental risk factors of substance use, mental illness, domestic violence, unmet material needs, custodial discord, or co-reported abuse?; and (3) In what proportion of investigations is physical neglect reported in the absence of an identified risk factor (e.g., substance use)?

The current study focused on how reporters and investigators characterize the primary concerns or nature of suspected maltreatment. Families may face numerous social or economic challenges that a call screener or investigating caseworker does not consider to be relevant to document or chooses not to assess. This study concerned one question: What predominant types of neglect and parental risk factors are reported and investigated under the umbrella of neglect? We assumed that CPS workers document contributing factors and concerns that are the focus of their investigations or feature significantly in their disposition, but we did not assume that all factors that are relevant to understanding the causes or consequences of neglect are documented.

Method

Data

Data were extracted from the California Child Welfare Services/Case Management System (CWS/CMS), which falls under the authority of the California Department of Social Services. Data were available through a university–agency data-sharing agreement approved by both state and university institutional review boards. CWS/CMS is the state’s case management system and includes records concerning all referrals of maltreatment and investigation activities. The structured fields (predefined data categories) in CWS/CMS data were used to classify allegations into one of two maltreatment types: neglect or abuse. Neglect is classified as any investigation with an allegation of general neglect, severe neglect, or caregiver absence or incapacity. Abuse is classified as any investigation with an allegation of physical or sexual abuse. The decision to exclude emotional abuse from the abuse category was based on the inability to distinguish domestic violence exposure from other forms of emotional abuse. In California, previous research has shown that domestic violence exposure is categorized under and represents a large portion of emotional abuse reports (Rebbe et al., 2021). Types of neglect (e.g., inadequate supervision, failure to protect) were derived from the state’s standardized hotline assessment tool based on information collected by a CPS worker during the call screening process (see Supplemental Appendix A). Demographic variables (race and ethnicity, age of focal child) were derived from the structured administrative data fields and used only to confirm that the randomly selected study population was representative of the overall population of investigated children.

Narrative data abstracted from unstructured fields associated with screening allegations of maltreatment (i.e., hotline screener narrative) and conducting investigations (i.e., investigative narrative) were used for the content analysis. The screener narrative is documented in a text field by a staff member at the CPS hotline. This field summarizes the reporter’s concerns and includes information regarding the alleged incident (severity, frequency, description of injury), child characteristics (demographics and special circumstances, such as health issues), caregiver characteristics (demographics and special circumstances, such as substance use, criminal record, mental illness), family characteristics (environment, support systems), and abuse characteristics (domestic violence, medical care needed). The investigative narrative is documented by the CPS worker assigned to the investigation and includes summaries of interviews with the reporter, child, and alleged perpetrators; records of social, cultural, or physical factors associated with the family; information on the child’s developmental needs; the child’s level of risk; and history of past CPS involvement, dispositions, and the basis for those dispositions.

Study Population

All CPS investigations that took place in California from January 1 to December 31, 2017, were identified (N = 231,728). If a family had multiple investigations during this 1-year period, the most recent investigation was selected. From this remaining set of unique family-investigations, 500 were randomly selected (shuffled and randomly assigned a number using Stata) and data from the selected referral and subsequent investigation were abstracted from CWS/CMS. Of the 500 reports, 79 were excluded because they did not have a corresponding investigative narrative; all 79 were from 12 counties that were subsequently discovered to not systematically use that text field (i.e., they record information from the investigation in alternative fields in CWS/CMS). Table 1 shows the breakdown of allegations for the statewide population of investigations and the random sample after excluding these 79 reports. There were no statistically significant differences between the overall population and the random sample. For the current study, only investigations that involved allegations of neglect were included (N = 295).

Table 1.

Comparison of Randomly Selected Study Population to Overall State Population of Children Investigated, California, 2017

	Statewide	Sample
	(N = 177,600)	(N = 421)
	%	%
Allegation investigated
Neglect	66.6	70.1
Neglect alone	40.3	39.9
Neglect and physical abuse	8.8	11.4
Neglect and sexual abuse	3.7	4
Neglect and emotional abuse	9.1	9.7
Neglect and multiple types of abuse	4.8	5
Allegation substantiated	13.5	14.7
First allegation	52.2	52.8
Race and ethnicity of selected child
White	22.6	20.2
Black	13.5	11.4
Hispanic	48.6	52.7
Asian or Pacific Islander	4.1	4.8
Missing or other	11.1	10.9
Age of selected child (M)	7.8	8.1

Note. No statistically significant (p < .05) differences were observed across characteristics.

Neglect Types

The type of neglect refers to the specific omission in care (i.e., physical neglect, inadequate supervision, failure to protect) determined to meet statutory guidelines warranting an investigation. Five neglect types were categorized based on the state’s structured hotline screening tool: (1) physical neglect, defined as (a) inadequate food, (b) inadequate or hazardous shelter, and (c) inadequate clothing or hygiene; (2) inadequate supervision, defined as nonattendance to a child despite caregiver presence or inadequate care arrangements for the child; (3) failure to protect, indicated when a caregiver knowingly leaves a child in the care of someone known to abuse children, does not intervene when there is knowledge (or a reasonable expectation that the caregiver should have known) of abuse or neglect, or is aware of exploitation by a third party; (4) caretaker absence or abandonment, defined as a caregiver being unable (e.g., incarceration, hospitalization) or unwilling (e.g., deserted or abandoned the child, forced the child out of the home) to provide care and there is no other safe adult to care for the child; and (5) other, including state-defined indicators of failure to thrive, malnutrition, and endangered health and safety (i.e., a caregiver’s willful negligence in providing for basic needs that caused or created imminent risk of causing serious injury; see Supplemental Appendix A for a complete list.)

Risk Factor Content Analysis

A content analysis of all 295 neglect investigations was conducted to identify parental risk factors present in the hotline and investigative narrative fields. A codebook was constructed to document systematically the presence or absence of four parental risk factors based on previous research: substance use, mental illness, domestic violence, and unmet material needs. The narrative coding of unmet material needs augmented information from the hotline screening tool on physical neglect because the narratives may have included information newly revealed during the investigation or contextual information reported by the referral source that did not factor into the screening decision.

The codebook was created and used to classify each risk using the process outlined by MacQueen et al. (1998), which involves team-based codebook development and coding. Three researchers reviewed and coded 10 randomly selected narratives, then compared coding results to ensure consistency in categorization of risks. Following this initial review, the research team added one risk factor that appeared frequently in the narratives: custodial discord. Custodial discord was distinct from domestic violence and defined as any conflict or fear of conflict, verbal or physical, between parents or caregivers who do not reside in the same household. Throughout the content analysis phase, the codebook was modestly revised, with categories and definitions refined to reflect new patterns emerging in the narratives (White & Marsh, 2006). Supplemental Appendix B reflects the adopted definition and provides examples of each contributing factor, risk, and concern included in the codebook. During the content analysis process, open-ended notes were taken by the research team to capture the context of investigations that did not fall in one or more of these categories. The final parental risk factors presented in this study from the content analysis are substance use, mental illness, domestic violence, unmet material needs, and custodial discord. We additionally examined co-reported child physical or sexual abuse as risk factors motivating an investigation based on structured data capturing allegation type codes.

Descriptive Analysis

First, we computed frequencies for each type of neglect: (1) physical neglect; (2) inadequate supervision; (3) failure to protect; (4) absence or abandonment; and (5) other. Likewise, we examined the frequencies of parental risk factors abstracted through the content analysis: substance use, mental illness, domestic violence, unmet material needs, and custodial discord, in addition to co-reported abuse allegations. Second, we documented the rate at which every parental risk factor was documented by neglect type (e.g., the proportion of reports indicating both inadequate supervision and substance use).

Results

Table 2 depicts the types of neglect and frequency of risk factors and concerns documented for all neglect investigations (N = 295) and the subset of neglect-only investigations (n = 168; i.e., those with no other co-reported physical or sexual abuse allegations). Inadequate supervision was the most frequent type of neglect (44% of all investigations for neglect and 54% of neglect-only investigations), followed by failure to protect (29% and 19%, respectively). Physical neglect was reported in 40 investigations overall (14% and 20%, respectively). In 28 of the 40 (70%) investigations of physical neglect, one or more additional type of neglect was co-reported. In 22 of 40 (55%) physical neglect was co-reported with inadequate supervision.

Table 2.

Summary of Neglect Types and Parental Risks Identified.

	All Investigations with a Neglect Allegation		Neglect-Only Investigations
	(N = 295)		(N = 168)
	%	n	%	n
Neglect type documented at hotline
Physical neglect	14	40	20	33
Inadequate supervision	44	131	54	90
Failure to protect	29	85	19	32
Absence or abandonment	6	18	7	12
Other	17	51	24	40
Parental risk factors discussed in case narratives
Substance use	41	122	49	83
Mental illness	18	52	17	28
Domestic violence	21	61	9	15
Unmet material needs	11	33	17	28
Custodial discord	14	40	9	15
Co-reported abuse (physical, sexual)	29	86

Note. Categories for neglect types and risk factors are not mutually exclusive.

Parental substance use was the most common risk factor identified, observed in 41% of all neglect investigations and 49% of neglect-only investigations. For all neglect investigations, parental mental illness was documented in 18%, domestic violence in 21%, custodial discord in 14%, and unmet material needs in 11%. Concurrent abuse was observed in 29% of all investigations for neglect.

Twenty-three percent of investigations did not include any parental risk factors coded in this study. Concerns documented in these investigations typically involved a child’s delinquency, self-harm, threats to others, or running away; unmet health care needs (e.g., adolescent expressed suicidality and parents declined emergency response team access); unintentional child death or serious injury implicating parental supervision or judgement; and risks created by a nonparent household member or person engaged in caregiving duties. Risks involving nonparent individuals were defined as substance use, severe mental illness, or domestic violence (e.g., mother and uncle, who both live with the child, had a physical altercation).

Table 3 presents types of neglect by parental risk factors documented through the content analysis and co-reported abuse. Substance use was the most frequently documented risk factor for each type of neglect. For example, 55% of investigations that involved physical neglect also involved substance use, as did 49% of investigations for inadequate supervision and 54% of absence or abandonment investigations. One in three (31%) investigations for failure to protect indicated the presence of domestic violence and 40% indicated co-reported abuse. Parental mental illness was identified in approximately one in five neglect investigations, regardless of the type of neglect. Not surprisingly, unmet material needs were documented in 43% of neglect investigations with physical neglect allegations. In four investigations (1% of all investigations), physical neglect was alleged in the absence of other neglect types or parental risk factors.

Table 3.

Types of Neglect by Identified Parental Risks.

	Documented Parental Risks
Neglect Type	Substance Use	Domestic Violence	Mental Illness	Unmet Material Needs	Custodial Discord	Co-Reported Abuse
	% (n)	% (n)	% (n)	% (n)	% (n)	% (n)
Physical neglect	55 (22)	23 (9)	18 (7)	43 (17)	18 (7)	5 (2)
Inadequate supervision	49 (64)	14 (8)	21 (28)	15 (19)	14 (18)	23 (30)
Failure to protect	39 (33)	31 (26)	19 (16)	4 (3)	17 (14)	40 (34)
Absence or abandonment	54 (13)	11 (2)	22 (4)	17 (3)	0 (0)	22 (4)
Other	51 (26)	8 (4)	22 (11)	6 (3)	6 (3)	14 (7)

To focus on the top four risks, substance use, mental illness, domestic violence, and co-reported abuse were combined and examined across each type of neglect. As illustrated in Figure 1, parental substance use, mental illness, domestic violence, and co-reported abuse were present in more than three quarters of all investigations, regardless of neglect type. Eighty-three percent of investigations for failure to protect, 78% of investigations of neglect due to absence or abandonment, 76% of investigations for inadequate supervision, and 75% of investigations for physical neglect included concerns related to parental substance use, domestic violence, mental illness, or co-reported abuse. Overall, 75% of all investigations of neglect included one or more of the top four risks.

Figure 1.

Proportion of each neglect type that occurred with identified parental risk. Note: SU = substance use; MI = mental illness; DV = domestic violence; abuse = co-reported physical or sexual abuse.

Discussion

In the current study, we examined a random sample of CPS investigations, with a detailed analysis of the 70% of cases that involved an allegation of neglect. Drawing on structured codes capturing neglect type and unstructured narrative text fields that documented parental risk factors, we reached three conclusions about neglect investigations. First, neglect allegations, especially allegations of inadequate supervision or physical neglect, typically involve concerns related to parental substance use, mental illness, and domestic violence. Also notable was the share of investigations that included co-reported physical or sexual abuse for at least one child in the report.

Second, concerns related to physical neglect were infrequently reported at the hotline (14% of investigations) or documented during the investigation (11%). This is largely consistent with prior research documenting one in 10 allegations related to failure to provide (Casanueva et al., 2011). Whether in the hotline screening tool or narratives, almost none of the neglect investigations only documented physical neglect or concerns about unmet material needs. This may be somewhat surprising, given that many families investigated for neglect are living in poverty (Dolan et al., 2011). We reiterate that we did not attempt to estimate the percentage of neglect investigations that involved families that are impoverished, experience material hardship, or have unmet material needs. Indeed, it seems evident from decades of data that poverty and financial hardship are common among families that experience CPS involvement (Berger & Waldfogel, 2011; Dolan et al., 2011; Font & Maguire-Jack, 2020a; Lindsey, 1991; Sedlak et al., 2010), especially those reported for neglect. Yet our study suggests that unmet material needs are rarely a concern documented as a basis for investigation. Certainly, it is possible that hotline and investigative caseworkers are discouraged from documenting unmet material needs or financial hardships due to the limited capacity of CPS to respond to those problems. This may also indicate that CPS underappreciates the extent to which poverty constrains parenting choices and induces parental stress; if so, CPS may be under-referring families to appropriate voluntary support services that would enhance employment or access to public benefits. However, 99% of investigations documented risk factors or concerns for the child other than or in addition to unmet material needs. This does not discount the possibility that some families are referred to CPS due to material hardship without other indications of harm or parental negligence but (appropriately) screened out. Because our study focused only on referrals that led to an investigation for neglect, our sample does not include those referrals.

Regardless, poverty may be an important and intervenable contributor to neglect. Poverty has a bidirectional causal connection to various aspects of psychosocial and behavioral functioning (Haushofer & Fehr, 2014; Ridley et al., 2020) and thus, economic supports may be a necessary component of a more comprehensive intervention. Notwithstanding the demonstrated benefits of economic supports for other aspects of child well-being (Parolin et al., 2021), they may be insufficient to address the complexity of concerns present in an alleged neglect case. That is, most neglect investigations involved substance use, mental illness, domestic violence, or co-reported abuse (with the remaining 23% involving challenging or concerning family contexts, such as risks due to other household members, the child’s mental and behavioral health, or parental abandonment), and economic supports alone are unlikely to resolve parental health conditions that manifest as neglect. Many studies linking economic supports to child maltreatment were either aggregate (state-level) analyses from which it is challenging to draw conclusions about the specific nature of changes in risk of harm or focused on economic supports that involve at least one parent in the labor force, such as the earned income tax credit (Berger et al., 2017), child support (Cancian et al., 2013), or minimum wage (Raissian & Bullinger, 2017). Because factors like substance abuse, unmanaged mental illness, and violence are not conducive to stable employment (Ringbom et al., 2022; Sherba et al., 2018) and many parents investigated by CPS do not have stable or full-time employment (Dolan et al., 2011; Font & Potter, 2019), it is questionable whether such policies, in isolation, will reduce neglect in those family contexts. Further research is needed to ascertain the mechanisms through which packaging economic supports with health- and behavior-focused interventions may reduce conditions that lead to CPS involvement.

Given that 44% of investigations were for allegations related to inadequate supervision, it is possible that this concurrent omission in care superseded unmet material needs as a focus of investigation, perhaps due to a perception that it is more voluntary or avoidable than unmet material needs. Nevertheless, 76% of reports screened in for inadequate supervision also included parental substance use, mental illness, domestic violence, or co-occurring sexual or physical abuse, indicating that some form of intervention or support is likely needed to reduce harm or risk of harm. The extent to which increased financial supports can reduce the incidence of inadequate supervision remains unknown. However, economically disadvantaged families are likely more vulnerable to allegations of inadequate supervision because they cannot draw on financial resources to mitigate or compensate for parental behaviors that may otherwise impose risk. For example, a family that has a fenced-in yard in a suburb can leave a child unattended for a few minutes with minimal risk of harm or CPS contact initiated by concerned passersby, whereas a parent utilizing an open greenspace near a busy intersection cannot. A more thorough understanding of the specific contexts underlying inadequate supervision complaints is needed to illuminate the types of interventions most appropriate for enhancing child safety.

Given that 28%–37% of children in the United States experience a CPS investigation by age 18 (Kim et al., 2017; Putnam-Hornstein, Ahn, et al., 2021) a majority of which involve neglect—there is reasonable concern that CPS is involved with too many families that could be served through other systems. A better understanding of the nature of and concerns underlying neglect allegations, such as that provided in this study, can identify opportunities to reduce the reach of CPS through differential pathways that are both preventive and reactive in nature. The lack of quality and accessible mental health and substance abuse treatment providers—particularly for Medicaid-eligible or uninsured persons—reinforces the contention that our current systems are ill equipped to prevent and treat these proximal contributors to neglect (Cunningham et al., 2006). Avenues such as warmlines (helplines similar to the 2-1-1 system but designed to connect at-risk families with family-strengthening resources) may provide mandated reporters and concerned community members with an alternative to CPS hotlines by providing a way to connect families in crisis with supports and services before reaching the threshold for CPS intervention (Casey Family Programs, 2020).

The 1 in 10 neglect investigations that involved custodial discord—hostile disagreements between non-cohabiting parents—present a unique challenge for CPS. With increased rates of nonmarital births and high rates of divorce (Wildsmith et al., 2018), it is not surprising that custody disagreements spill over to CPS. Some of these investigations involve malicious reporting (e.g., false accusations of abuse or neglect by one parent against another), but CPS cannot refuse to investigate allegations of abuse without clear evidence of false reporting. An additional complication is research clearly pointing to a child’s increased risk of serious (including fatal) abuse at the hands of unrelated paramours (Schnitzer & Ewigman, 2005). Targeted services provided outside of CPS, such as mediation or other custody-related services through family courts, may provide an alternative pathway for a subset of investigations. Increased cross-reporting and communication between CPS hotlines and family courts overseeing child custody orders may also benefit all involved parties.

Current legal definitions of neglect focus on the manifestation of harm to a child (unmet material, supervision, or protection needs). However, preventing or alleviating harm due to neglect requires a detailed understanding of both how and why a child’s needs were unmet. Current federal data (National Child Abuse and Neglect Data System) on risk factors are so unreliable that they provide more confusion than clarity; state data systems (from which federal data are drawn) are similarly limited (Font, 2020; Seay, 2015). The lack of a clear accounting of how and why neglect manifests contributes to three interrelated problems. First, reasonable—though uncorroborated by this study’s analysis—narratives about CPS confusing or conflating poverty-induced material hardship with neglect have taken hold (Connolly et al., 2015). These narratives may encourage disinvestment in CPS and lead to investments in programming that fails to address, prevent, or treat parental conditions directly tied to maltreatment. For example, current discourse on CPS reform emphasizes reallocation of (already limited) child welfare funding toward unconditional cash assistance, with relatively limited emphasis on the need for effective substance abuse treatment programs (Roberts, 2020). Second, the lack of clear documentation and conceptualization of neglect provides parents with little guidance about the legal boundaries of minimally adequate care or activities or behavioral changes needed to avoid CPS involvement or retain or regain custody of their children following CPS involvement. Third, the amalgamation of diverse risk factors and manifestations of harm into a single category of neglect reduces the capacity of researchers to identify the circumstances in which interventions are more or less likely to be effective in preventing maltreatment or mitigating its effects on children’s development. When neglect is the basis for a CPS intervention, services generally target the parents and attempt to improve their capacity to provide adequate care. However, current data are limited in their capacity to ascertain the parent’s culpability in the risk or occurrence of neglect, even though that nexus—between parental behavior and risk or harm to a child—is ostensibly what determines substantiation and service provision decisions. This study identified that most neglect investigations involved concerns that implicate parental agency—such as substance use, violence in the home, and protection of a child from harm by others—that once occurring are unlikely to be ameliorated by material supports alone.

Limitations

We note several limitations of our analysis. First, the analysis was based on a single state, and given jurisdictional differences (Rebbe, 2018), it is possible that other states receive more reports that focus on unmet material needs alone, are less likely to screen out those reports, or place a greater emphasis on those concerns during the course of the investigation. Second, it is possible that implicit or explicit biases related to poverty informed the decision to report or the actions ultimately taken by the investigator, even if they never recorded such concerns. Our study did not rule out the possibility of bias based on poverty in how CPS workers assessed or characterized parenting behaviors (e.g., substance use). Third, this study was exploratory in nature and did not assess how the nature of documented risk factors varied by racial and ethnic groups, age, or other child and family attributes. Future research may consider whether automated or machine-learning approaches provide a reliable and less laborious option for coding narrative data in a larger sample.

Conclusion

Results of the current study reinforce that cross-system strategies are needed to address the complex needs of families reported to CPS for neglect. Yet they also suggest that children reported for neglect overwhelmingly do not only suffer from conditions of unmet material needs. Documented parental substance use, mental illness, domestic violence, inadequate supervision and failure to protect indicate that economic strategies alone are likely insufficient for child safety and reinforce the challenges faced by CPS when screening maltreatment reports. Importantly, these parental struggles appear to have been exacerbated by the COVID-19 pandemic, highlighting the relevance of these findings (Czeisler et al., 2020). Low rates of uptake of and retention in voluntary services create challenges for the expansion of prevention services or non-CPS interventions (Putnam-Hornstein, Prindle, et al., 2021).

ORCID iDs

Lindsey Palmer https://orcid.org/0000-0001-5633-9976

Sarah Font https://orcid.org/0000-0002-4022-5517