Clinical and Epidemiological Characteristics of Suspected Child Maltreatment Reported After Caregiver Emergency Department Visits in Japan

ABSTRACT

Aim

Child maltreatment is a major public health concern, yet identification in emergency department settings remains challenging. Caregiver‐focused approaches may help identify families at risk. We describe cases reported to a hospital‐based Child Protection Team following caregiver emergency department visits in Japan.

Methods

We conducted a single‐center retrospective observational study at an urban tertiary hospital. We included caregiver emergency department visits between January 2022 and December 2024 when the presenting adult was 18 years or older, the visit was reported to the Child Protection Team regarding at least one child under 18 years in that adult's care, and the team subsequently judged that involvement was required. We extracted medical record variables and used descriptive analyses.

Results

We included 79 index caregivers and 101 emergency department visits. A documented history of mental health conditions was present in 53% of caregivers. Common diagnoses included drug intoxication, alcohol‐related disorders, somatic symptom‐related or dissociation‐related diagnoses, injuries related to domestic violence, and self‐inflicted injuries. Neglect was the most common reason for notification. Fifty‐seven percent of caregivers had prior involvement with the Child Protection Team or the Council for Children Requiring Care and Support, and 30% required new registration or temporary child protection after the team review.

Conclusions

This study describes a selected group of caregiver emergency department visits that were reported to a hospital‐based Child Protection Team and subsequently judged to require team involvement. Further studies with defined caregiver denominators and comparison groups are needed to evaluate caregiver‐focused emergency department assessment pathways.

Caregiver emergency department visits that triggered Child Protection Team notification frequently involved mental health and substance‐related conditions, domestic violence‐related injuries, and self‐harm. These findings indicate that caregiver visits can provide opportunities to identify suspected child maltreatment and initiate family‐focused linkage to support services, even when children do not present for care.

1. Background

Child maltreatment is commonly defined as a broad set of harmful conditions affecting children, including physical, psychological, and sexual abuse; neglect; exploitation; and failures to provide appropriate care [1]. It is recognized as a major public health issue with long‐term consequences for children's physical, psychological, and social well‐being. In Japan, official statistics published by the Ministry of Health, Labour and Welfare (MHLW) classify reported cases into categories such as physical harm, psychological harm, sexual harm, and neglect. The number of reported cases has continued to increase, reaching approximately 225,000 in 2023. Although healthcare professionals in Japan are legally required to report suspected child maltreatment, reports originating from medical institutions accounted for only 1.9% of all notifications in 2023 [2]. The MHLW has identified insufficient reporting from healthcare settings as a persistent structural challenge.

Emergency departments (EDs) frequently encounter children with a wide spectrum of medical conditions and therefore represent important points of contact with families at risk of maltreatment [3]. Nevertheless, identifying maltreatment in ED settings remains challenging. Previous studies have described multiple barriers to identifying and responding to suspected child maltreatment in the ED, including the busy and time‐pressured ED environment, limited access to background information, variability in clinicians' experience, cognitive bias related to caregivers' characteristics, and concerns about damaging relationships or potential retaliation [4, 5]. As a result, opportunities for early identification of maltreatment may be missed.

In addition to child‐focused assessments, caregiver‐focused screening approaches have received increasing attention. For example, in pediatric primary care settings, the Safe Environment for Every Kid (SEEK) model screens caregivers for psychosocial risk factors such as depression, substance use, domestic violence (DV), and severe stress and provides structured interventions [6]. It has been reported to reduce psychological maltreatment and some forms of physical abuse. In ED settings, the Hague Protocol was developed in the Netherlands to identify children at high risk for maltreatment based on the characteristics of adults presenting to the ED with their own medical problems [7]. This protocol targets caregivers presenting with factors such as DV, suicide attempts, severe mental health conditions, and substance misuse and systematically assesses whether they have children under 18 years.

Although international efforts to implement caregiver‐focused screening approaches are increasing, to our knowledge, no studies from Japan have described suspected child maltreatment identified through caregiver ED visits. Accordingly, the present study aimed to describe the clinical and epidemiological characteristics of cases in which child maltreatment was suspected and reported to the hospital‐based Child Protection Team (CPT) following caregiver ED visits.

2. Methods

2.1. Study Design and Setting

This was a single‐center retrospective observational study with descriptive analyses conducted at Hyogo Prefectural Amagasaki General Medical Center, a community‐based tertiary emergency hospital in Japan. The ED receives approximately 30,000 patients annually, of whom approximately 35% are children under 18 years. The hospital operates a hospital‐based CPT, a multidisciplinary team responsible for the assessment and management of suspected child maltreatment. At our institution, when healthcare professionals working in the ED consider that child maltreatment or child safety concerns cannot be ruled out, they complete an institutionally standardized CPT notification form developed with reference to a regional government‐issued child abuse prevention manual. Thus, the decision to initiate ED‐to‐CPT reporting was based on clinician judgment, whereas the reporting process was standardized through use of the notification form and subsequent CPT review. For visits requiring urgent attention, ED staff additionally contact the 24‐h on‐call CPT physician. For nonurgent reports, the completed notification form is reviewed by the multidisciplinary CPT on the next weekday, and the team determines whether CPT involvement is required. We screened all consecutive entries in the CPT registry during the study period. We then reviewed the electronic medical records, including clinical progress notes, to confirm patient characteristics, ED presentation, relationship to the child, reasons for CPT reporting, and subsequent CPT activities.

There are two pathways for CPT notification in the ED: (1) suspicion of maltreatment identified through a child's ED visit, and (2) suspicion of maltreatment identified through a caregiver's ED visit, based on concerns regarding children in the patient's care. The present study focused exclusively on the latter pathway.

2.2. Study Period and Participants

We reviewed the electronic medical records of adults who visited the ED between January 1, 2022, and December 31, 2024, to confirm eligibility. The index caregiver was defined as the presenting adult ED patient whose visit was reported from the ED to the CPT and subsequently judged by the CPT to require involvement regarding suspected maltreatment or safety concerns for a child in that patient's care.

ED visits were included if they met all of the following criteria: (1) the presenting adult patient was a caregiver of at least one child younger than 18 years, (2) the visit was reported from the ED to the CPT regarding suspected maltreatment or safety concerns for a child in that patient's care, and (3) the CPT subsequently judged that involvement was required. We excluded cases with insufficient documentation, cases not evaluated as a standard ED visit, cases in which the adult patient rather than a child in their care was the target of intervention, consultation‐only cases that did not proceed to CPT involvement, cases involving caregiving concerns related to parental illness without suspected maltreatment, and fetal consultation cases. Children younger than 18 years under the care of the index caregivers were referred to as “children in their care.”

2.3. Data Collection

Using electronic medical records, we collected data on the index caregivers' age, sex, receipt of public assistance, number of children in their care, caregiver relationship to the child or children involved, history of mental health conditions, ED‐recorded diagnoses, ED disposition, reasons for CPT notification, prior involvement with the CPT or the Council for Children Requiring Care and Support (CCRCS), ages of the children in their care, and responses implemented after CPT notification.

Two investigators independently abstracted data from electronic medical records and CPT documentation using predefined data items. Mental health history was defined as self‐reported or physician‐documented diagnoses recorded before or at the index ED visit and was grouped into predefined categories based on DSM‐5 terminology. ED‐recorded diagnoses were based on routine ED documentation at the time of the visit and were presented as ED diagnostic categories rather than formal psychiatric diagnoses. Because more than one ED diagnosis could be recorded during a single visit, a single visit could be classified into multiple diagnostic categories. Reasons for CPT notification were classified according to the World Health Organization categories: physical abuse, sexual abuse, emotional abuse, neglect, and multiple types. Variables requiring judgment, including WHO maltreatment categories, mental health history categories, and ED diagnostic categories, were independently classified according to predefined definitions. Any discrepancies were resolved through discussion and consensus. Post‐CPT responses, such as new registration with the CCRCS or temporary child protection, were summarized on a per‐index caregiver basis.

2.4. Statistical Analysis

Descriptive statistics were used to summarize demographic and clinical characteristics of index caregivers and children in their care. Continuous variables were reported as medians with interquartile ranges, and categorical variables were presented as numbers and percentages. All analyses were descriptive.

2.5. Ethics Approval

The study protocol and data collection procedures were approved by the institutional ethics review board (approval number: 7‐293).

3. Results

During the study period, 83,333 ED visits were recorded, of which 52,329 (62.7%) involved patients aged 18 years or older.

A total of 287 entries were identified from the institutional CPT registry, and all consecutive entries during the study period were screened for eligibility. Of these, 141 involved ED patients aged 18 years or older. The final study population consisted of 101 ED visits involving 79 unique index caregivers (Figure 1). Among the 79 index caregivers, 61 had one included ED visit, 14 had two visits, and 4 had three visits during the study period (Table 2).

FIGURE 1.

Study flow diagram. The figure shows the number of consecutive Child Protection Team registry entries identified during the study period, adult emergency department cases screened for eligibility, exclusions with reasons, and the final analytic sample. CPT, Child Protection Team; ED, emergency department.

TABLE 2.

Characteristics of ED visits associated with Child Protection Team notification (n = 101 visits).

Characteristics of ED visits	ED visits (n = 101)
Arrival by ambulance, n (%)	78 (77)
Caregivers with multiple ED visits, n	18
Maximum number of ED visits per caregiver, n	3

Note: Percentages were calculated at the visit level unless otherwise indicated. Repeated ED visits by the same index caregiver were counted as separate visits. “Index caregivers with more than one included ED visit” represents the number of unique index caregivers, not the number of additional ED visits.

Abbreviation: ED, emergency department.

The median age of the index caregivers was 32 years, and 75 (95%) were female. Among the 79 index caregivers, 74 (94%) were mothers, 4 (5%) were fathers, and 1 (1%) was a grandmother (Table 1).

TABLE 1.

Baseline characteristics of index caregivers (n = 79) and children in their care.

Characteristics of index caregivers	Index caregivers (n = 79)
Age, median (IQR)	32 years (28–39)
Female sex, n (%)	75 (95)
Relationship to child, n (%)
Mother	74 (94)
Father	4 (5)
Grand mother	1 (1)
A history of mental health conditions, n (%)	42 (53)
Receipt of public assistance, n (%)	27 (34)
Caring for two or more children, n (%)	32 (40)
Total number of children in their care, n	144
Age of children in their care, median (IQR)	8 years (3–10)
Age category of children in care, n
< 1 year	8
1–6 year	51
7–12 year	49
13–17 year	22
Unknown	14

Note: “Index caregiver” refers to the caregiver whose emergency department visit triggered notification to the hospital‐based Child Protection Team. Values are presented as median (IQR) or number (%), as appropriate. Percentages were calculated using the total number of children in care as the denominator, including children with unknown age.

Abbreviation: IQR, interquartile range.

Regarding mode of arrival, 78 visits arrived by ambulance and 23 were walk‐in visits (Table 2). A history of mental health conditions was documented in 42 index caregivers (53%). Twenty‐seven (34%) were receiving public assistance, and 32 (40%) were caring for two or more children. The total number of children in their care was 144, with a median age of 8 years (Table 1).

Regarding ED‐recorded diagnoses, drug intoxication was the most common diagnosis (26 visits), followed by alcohol‐related disorders (25 visits), somatic symptom‐related or dissociation‐related diagnoses (18 visits), injuries related to DV (10 visits), and self‐inflicted injuries (7 visits) (Table 3).

TABLE 3.

Distribution of emergency department‐recorded diagnoses (n = 101 visits).

ED‐recorded diagnoses (multiple responses allowed)	ED visits (n = 101)
Drug intoxication, n (%)	26 (26)
Alcohol‐related disorder, n (%)	25 (25)
Somatic symptom‐related or dissociation‐related diagnoses, n (%)	18 (18)
Injuries related to domestic violence, n (%)	10 (10)
Self‐inflicted injury, n (%)	7 (7)
Precipitous delivery (unbooked pregnancy), n (%)	5 (5)
Others, n (%)	19 (19)

Note: Multiple responses were allowed; therefore, percentages do not sum to 100%. Diagnoses were based on routine ED documentation at the time of the visit.

Abbreviation: ED, emergency department.

Reasons for CPT notification included neglect (74 visits), emotional abuse (42 visits), and physical abuse (19 visits); multiple reasons were identified in 31 visits (Table 4).

TABLE 4.

Reasons for Child Protection Team notification classified according to World Health Organization maltreatment categories (n = 101 visits).

Reason for CPT notification (multiple responses allowed)	ED visits (n = 101)
Neglect, n (%)	74 (73)
Emotional maltreatment, n (%)	42 (42)
Physical maltreatment, n (%)	19 (19)
Multiple concerns, n (%)	31 (31)

Note: Multiple reasons could be recorded for a single visit. Classification was based on documentation in the medical records.

Abbreviations: CPT, Child Protection Team; ED, emergency department.

Among the index caregivers, 33 (42%) had prior involvement with the CCRCS, and 18 (23%) had been previously reported to the hospital‐based CPT. Following the CPT notification, 24 (30%) were newly registered with the CCRCS or had children placed under temporary protection (Table 5).

TABLE 5.

Previous support history and post‐Child Protection Team responses among index caregivers (n = 79).

Support history and post‐CPT responses	Index caregivers (n = 79)
History of support before the index caregiver ED visit
Previous involvement with CCRCS, n (%)	33 (42)
Previous report to the hospital CPT, n (%)	18 (23)
Post‐CPT responses (multiple responses allowed)
Information sharing with municipal offices or child guidance centers, n (%)	73 (92)
Temporary child protection or new registration with the CCRCS	24 (30)

Note: Post‐CPT responses were summarized on a per‐index caregiver basis. Multiple responses were allowed; therefore, percentages do not sum to 100%.

Abbreviations: CCRCS, Council for Children Requiring Care and Support; CPT, Child Protection Team; ED, emergency department.

4. Discussion

This retrospective observational study describes the clinical and epidemiological characteristics of cases in which suspected child maltreatment was reported to a hospital‐based CPT following caregiver ED visits in Japan. To our knowledge, these findings provide novel data in a field where domestic evidence remains limited.

The Hague Protocol aims to identify children at risk of child maltreatment based on psychosocial characteristics of caregivers presenting to the ED with their own medical problems, such as substance use, DV, and suicide attempts [7].

This caregiver‐focused approach is relevant to emergency medicine because adult ED presentations can reveal psychosocial risks that extend beyond the presenting patient and may affect children in the same household. When the presenting adult is a caregiver, the ED encounter may therefore serve as a clinical contact point at which potential child safety concerns become apparent, even when the child is not the presenting patient.

In the present study, ED visits related to drug intoxication, alcohol‐related disorders, injuries related to DV, and self‐inflicted injuries were common among index caregivers. These findings indicate that the included caregiver ED visits often involved presentations similar to those targeted by the Hague Protocol. However, because this study was a retrospective descriptive study of selected cases reported from the ED to the CPT and subsequently judged to require CPT involvement, these findings should not be interpreted as evidence of the effectiveness of caregiver‐focused screening or assessment in the ED. Furthermore, studies of adult ED visits related to intimate partner violence have reported that social work involvement is associated with the completion of child safety assessments [8], suggesting that evaluations initiated during caregiver ED visits may facilitate linkage to family support services.

More than half of the index caregivers in this study had a documented history of mental health conditions. Prior research has highlighted parental mental health problems as an important contextual factor associated with child maltreatment and suboptimal caregiving, and has emphasized the importance of family‐focused practice, including family‐level assessment, interprofessional coordination, and linkage to support services [9, 10]. Accordingly, once suspected child maltreatment is reported to the CPT, the clinical focus may appropriately shift toward the child's safety; however, these cases also highlight the importance of concurrently assessing caregiver needs, including mental health when relevant, and facilitating coordinated linkage to social work and other support services through existing referral pathways.

In this study, 57% of index caregivers had either prior involvement with the CCRCS or a previous CPT report. This proportion was similar to reports from studies of the Hague Protocol, which found that approximately two‐thirds of families identified through caregiver ED screening were already known to support agencies before formal notification [11]. Furthermore, systematic reviews have identified several predictors of maltreatment recurrence, including prior maltreatment, neglect, caregiver mental health conditions, and family conflict [12].

In the present study, prior involvement with the CPT or CCRCS, neglect, and caregiver mental health conditions were frequently documented among the included index caregivers, although this study did not evaluate predictors of recurrence.

Several limitations should be acknowledged. This was a single‐center, retrospective study with a limited sample size, which may restrict the generalizability of the findings. Long‐term outcomes related to child safety and family functioning were not assessed. Because the denominator of all adult ED visits by caregivers of children younger than 18 years was unavailable, this study could not estimate the frequency, detection rate, or screening yield of suspected child maltreatment among caregiver ED visits. The absence of a comparison group also precludes interpretation of the findings as evidence of the effectiveness of caregiver‐focused screening or assessment pathways in the ED.

Repeated ED visits by the same index caregiver may have influenced visit‐level distributions; therefore, visit‐level percentages should be interpreted as describing included ED visits rather than unique caregivers. Accordingly, the findings should be interpreted as a descriptive summary of selected CPT‐involved cases rather than evidence that caregiver ED visits are generally an effective screening pathway.

5. Conclusion

This study described the clinical characteristics of selected cases reported to a hospital‐based CPT following caregiver ED visits for suspected child maltreatment or child safety concerns. ED‐recorded diagnoses commonly included drug intoxication, alcohol‐related disorders, somatic symptom‐related or dissociation‐related diagnoses, injuries related to DV, and self‐inflicted injuries. More than half of the index caregivers had prior involvement with the CPT or the CCRCS, and additional interventions were implemented for some families after CPT notification. Because the study lacked a defined caregiver denominator and comparison group, further prospective studies are needed to evaluate the feasibility, yield, and effectiveness of caregiver‐focused assessment pathways in the ED.

Ethics Statement

All procedures used in this research were approved by the Ethics Committee of Amagasaki General Medical Center (approval no. 7‐293).

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data are available upon reasonable request.