Family Risk Factors and Emotional–Behavioral Problems in Children in Protective Care

Abstract

Children in residential care constitute a particularly vulnerable group at high risk of developing emotional and behavioral difficulties as a consequence of adverse experiences and dysfunctional family environments. Identifying risk and protective factors is essential for designing interventions tailored to their needs; however, the available research remains limited and does not always provide the evidence required to guide effective programs within the child protection system. The aim of this study was to examine the prevalence of emotional and behavioral problems among children in residential care and to analyze the role of family factors, sex, and age in these difficulties. The sample consisted of 210 children aged 6 to 18 years institutionalized in residential care centers and supervised apartments. A cross-sectional design was employed, administering the Strengths and Difficulties Questionnaire (SDQ) along with an ad hoc questionnaire to collect socio-family variables. The results reveal a high prevalence of emotional and behavioral difficulties. The multivariable models explained between 8.1% and 29.4% of the variance in emotional and behavioral functioning and showed that age, sex, exposure to gender-based violence, parental substance use, and parental intellectual disability were associated with specific emotional and behavioral dimensions. The study highlights the need to develop and implement educational and therapeutic programs aimed at strengthening children’s emotional regulation, addressing behavioral difficulties, and considering family-related adversity in intervention planning.

1. Introduction

Childhood is a crucial stage in development, during which the cognitive, affective, and behavioral foundations are established and later shaped by interactions with the environment and lived experiences. Studies show that adverse experiences during the early stages of development constitute a significant risk factor for mental health. In particular, child maltreatment, dysfunctional family dynamics, or exposure to vulnerable contexts have been associated with a higher likelihood of presenting emotional and behavioral difficulties in childhood and adolescence (Ball et al. 2024; Gardner et al. 2019; Racine et al. 2020). Moreover, these early adverse conditions increase the risk of developing various psychological disorders throughout the life span (Sasidharan and Talwar 2023).

Within this framework, a particularly vulnerable population to the effects of early adversity is that of minors in residential care due to child protection concerns, as a significant proportion of them have a history of child maltreatment.

Although Spain has made significant progress in recent years toward expanding family-based foster care—consistent with international recommendations and with evidence demonstrating the developmental benefits of growing up in family environments—residential care remains an essential resource for those children whose circumstances require specialized attention or for whom appropriate family alternatives are not available. For this reason, the present study focuses on children and adolescents residing in residential child protection centers.

Residential care plays a crucial role in safeguarding children who are in situations of neglect or inadequate protection. When it becomes necessary to remove a child from their family environment, residential care centers constitute one of the essential resources within the child protection system. These facilities are conceived as temporary settings designed to promote the adaptation and socialization of children whose families are unable to adequately fulfill their parental responsibilities.

Several studies highlight that minors who have been victims of child maltreatment show high levels of emotional and behavioral difficulties (Dubois-Comtois et al. 2021; Maguire et al. 2024; Trubey et al. 2024). Research in this population has documented the high prevalence of emotional symptoms—such as anxiety and depression—as well as behavioral difficulties, including attention problems, aggressive behaviors, disruptive conduct, and social withdrawal (de Vries et al. 2024; Martín et al. 2020). Westlake et al. (2023) report that approximately half of children in residential care present some form of psychopathology, with behavioral problems and trauma-related symptoms being particularly prevalent.

Despite the clinical and social relevance of this phenomenon, research on the characterization of psychopathological symptoms among minors in protective residential care is still limited (Mersky et al. 2017; Sainero et al. 2014; Strathearn et al. 2020), which restricts the generalization of findings. There is therefore a need to deepen our understanding of the profile of emotional and behavioral manifestations and the identification of sociofamily circumstances and risk factors that influence or predict their emergence. This information is crucial for designing evidence-based interventions aimed at enhancing adaptation and promoting resilience among minors. In this regard, several studies highlight the protective role of secure attachment relationships and stable caregiving environments, which can buffer the adverse effects associated with maltreatment (Magalhães et al. 2021; Pinheiro et al. 2024).

The scarcity of studies focused on the mental health of children in residential care is particularly relevant in Spain, where this resource constitutes one of the main child protection measures. According to the Boletín de Datos Estadísticos de Medidas de Protección a la Infancia y Adolescencia (Observatorio de la Infancia, 2024), a total of 17.112 minors were placed in residential care in 2023, reflecting an upward trend compared to the previous two years and highlighting the growing importance of this resource within the protection system. Furthermore, nearly half of these minors remain in residential care until they reach adulthood, which underscores the high proportion of children and adolescents who spend their entire minority within the protection system (Poole et al. 2025). This scenario emphasizes the urgent need to expand knowledge regarding their emotional and behavioral difficulties (Del Valle et al. 2011), as well as their socio-familial characteristics, in order to guide intervention and decision-making in the field of child protection.

Current research shows marked methodological heterogeneity, with a predominance of studies centered on protection modalities other than residential care, in particular family foster care (Maguire et al. 2024). Moreover, most studies conducted in the field of residential care do not systematically incorporate analyses of family-related risk factors that may serve as relevant predictors of the emotional and behavioral symptomatology observed in children and young people (González-García et al. 2023; Moreno-Manso et al. 2020; Soriano-Díaz et al. 2023). This limitation hinders a comprehensive understanding of the mechanisms underlying the psychological adjustment of this population and highlights the need for studies to adopt broader, context-sensitive approaches. Added to this is the absence of a standardized data-recording system that would allow for systematic and comparable collection of information about minors and their families. Nonetheless, recent efforts are beginning to address this fragmentation through studies with larger and more representative samples (Águila-Otero et al. 2024).

Regarding the factors that may influence the presence of emotional and behavioral symptoms in minors, several studies point to differential patterns based on sex and age. Specifically, a higher prevalence of externalizing symptoms has been observed among boys, whereas girls tend to display a greater prevalence of internalizing symptomatology (Trubey et al. 2024). Moreover, poor adjustment has also been documented among older children and adolescents residing in residential care. This latter finding may be explained, among other factors, by the vulnerability associated with adolescence, a developmental period during which the expression of emotional and behavioral difficulties tends to intensify (Montserrat et al. 2022). In addition, the socio-familial characteristics present prior to the separation from the biological home constitute highly relevant factors, as contextual variables can exert significant influence on the developmental trajectories of minors, acting as risk factors or even as predictors of later adjustment (Konijn et al. 2019; Maguire et al. 2024).

In this context, the objectives of the study were: (a) to examine the presence of emotional and behavioral symptoms among children and young people in residential care; (b) to analyze the presence of family-related problems that may act as risk factors for the symptomatology presented by minors (gender-based violence, mental health problems, substance use, intellectual disability, and economic hardship); (c) to determine whether significant differences exist in emotional and behavioral symptomatology according to age, sex, and family risk factors; and (d) to assess the independent associations of age, sex, family risk factors, length of stay, and maltreatment history with the emotional and behavioral symptoms of minors. Based on the theoretical review conducted, four hypotheses were formulated: (Hypothesis 1) we anticipated that children and young people would exhibit emotional and behavioral symptomatology; (Hypothesis 2) family risk factors (gender-based violence, mental health problems, substance use, intellectual disability, and economic hardship) would be associated with the emotional and behavioral symptoms of minors; (Hypothesis 3) significant differences would be found in emotional and behavioral symptomatology according to age, sex, and the presence of family risk factors—with greater difficulties expected among older minors and those exposed to more adverse family contexts; and (Hypothesis 4) age, sex, family risk factors, length of stay, and maltreatment history would be independently associated with the emotional and behavioral symptomatology of children and young people.

2. Materials and Methods

2.1. Participants

The sample consisted of 210 children and young people in residential care within the Autonomous Community of Extremadura, aged between 6 and 18 years (M = 13.08; SD = 2.67). Of the total sample, 28.1% (n = 59) were between 6 and 11 years old, whereas 71.9% (n = 151) were between 12 and 18 years old. The sample included 102 girls (48.6%) and 108 boys (51.4%).

With respect to length of institutionalization, 92 minors (43.8%) had been in residential care for less than two years, whereas 118 (56.2%) had been in care for more than two years. Length of stay ranged from 3 months to 3 years. Regarding the child protection measure, 101 minors (48.1%) had experienced some form of maltreatment, while 109 (51.9%) were institutionalized under other types of protection measures.

Among cases of child maltreatment, the predominant type was distributed as follows: physical maltreatment (13.3%; n = 28), physical neglect (32.9%; n = 69), and sexual abuse (1.9%; n = 4). As is common in this population, many minors who experienced physical maltreatment also presented associated emotional maltreatment, and likewise, those with physical neglect often showed emotional neglect. However, classification was conducted based on the predominant type of maltreatment.

Given the vulnerability associated with experiences of maltreatment, neglect, and adverse family histories, a purposive sampling strategy was used. Inclusion criteria were: (a) being under an official child protection measure in Extremadura; (b) being between 6 and 18 years old; (c) being placed in residential care as a formal protection measure (excluding emergency or assessment placements); and (d) having been in the residential setting for at least 3 months to ensure that the key worker could reliably provide informant-report information when applicable. Unaccompanied foreign minors and minors with intellectual disabilities were included, with case-by-case evaluation of whether instrument completion was appropriate based on comprehension level.

Data were collected in residential care centers and supervised flats located in different areas of the region. All residential care centers belong to the regional child welfare authority (Dirección General de Servicios Sociales, Infancia y Familia) and share similar organizational characteristics, particularly in terms of the number of minors served and staff-to-child ratios. These facilities function as protective environments that ensure the coverage of basic needs and promote the comprehensive development of minors. Placement in either a residential care center or a supervised flat is determined based on an individualized assessment of each child’s personal and family circumstances.

The characteristics of the children and young people included in the study correspond to the situations of neglect and risk defined in Law 4/1994, of 10 November, on the Protection and Care of Minors of the Autonomous Community of Extremadura. Within this framework, the types of child maltreatment vary according to different classification systems, although there is broad consensus in identifying several core categories.

Physical maltreatment includes actions that cause harm or illness, such as cutaneous, skeletal, or internal injuries. Emotional maltreatment refers to behaviors that may cause harm to physical, mental, spiritual, moral, or social development, such as degradation, rejection, humiliation, or the induction of fear. Physical neglect involves the failure to attend to basic needs such as nutrition, hygiene, protection, education, or health care, while emotional neglect is linked to the absence of a stable attachment figure (Myers 2011).

In this study, maltreatment classification was based on the predominant type experienced by each minor, as some cases involved mixed forms, with co-occurring physical and emotional manifestations or combined forms of neglect.

Additionally, minors placed under protection measures other than child maltreatment were included. These measures encompass: the inability of parents to fulfill their parental responsibilities (due to reasons such as imprisonment or orphanhood), cases of abandonment or voluntary relinquishment (particularly with newborns or situations of complete disengagement), parental incapacity to exert adequate control (when caregivers are unable to establish appropriate boundaries), and child-to-parent violence, as well as other circumstances recognized by the competent child welfare authority.

2.2. Instruments

Standardized instruments with adequate psychometric properties were used in the study. The instruments employed were the following:

• Basic Information Questionnaire. First, an ad hoc questionnaire specifically designed for this study was administered. This instrument collects relevant case information, including sociodemographic variables of the minor (sex and age), as well as socio-familial characteristics of parents or caregivers that may influence the child’s development: history of gender-based violence, presence of mental health problems, substance use, intellectual disability, and family economic situation. The questionnaire also gathers information on the protection measure in force for each case, the characteristics of the situation of neglect or maltreatment that led to the opening of the protection file, and the length of stay of the minor in the corresponding residential resource. Information on family-risk variables was obtained from official child protection case records and the reports available in each case file, and was coded according to the presence or absence of each risk factor.
• Strengths and Difficulties Questionnaire (SDQ) (Goodman 1997; Ortuño-Sierra et al. 2015). The SDQ is a widely used screening instrument for assessing emotional and behavioral difficulties, as well as prosocial behavior, in children and adolescents, and it allows for multi-informant assessment. In this study, the Spanish version of the SDQ was administered according to participants’ age. For children aged 6–10 years, the informant-report version was completed by the key worker or residential care professional with direct knowledge of the child. For participants aged 11–18 years, the self-report version appropriate to their age was administered directly. In both cases, administration was conducted under the supervision of the research team, and evaluator support was provided when necessary to ensure comprehension of the instructions and item wording. This support was limited to reading or clarifying items and did not involve suggesting or modifying responses. The questionnaire consists of 25 items organized into a dimensional model composed of five scales, each containing five items: Emotional Problems, Conduct Problems, Hyperactivity, Peer Problems, and Prosocial Behavior. The first four scales make up the Total Difficulties score. Each dimension is scored from 0 to 10, allowing classification into normal, borderline, or abnormal ranges. The SDQ has extensive evidence of validity and reliability in clinical and educational contexts. The original version reports an overall internal consistency of α = 0.68, whereas in the present study, an internal consistency of α = 0.77 was obtained.

2.3. Procedure

First, authorization was requested from the child welfare authority responsible for the guardianship of minors in the Autonomous Community of Extremadura, acting as their legal representative, to carry out the study. Once institutional approval and consent were obtained, the assessment instruments were administered. Data collection took place in the residential care centers and supervised flats where the participants resided.

The assessments were conducted individually, and all children and adolescents participated voluntarily. The corresponding instruments were administered according to the age-specific administration procedure. Throughout the entire process, evaluators remained present to clarify doubts, supervise proper administration, and ensure the quality of the data collected. No relevant difficulties in comprehension or handling of the instruments were observed among participants.

The study was reviewed and approved by the Ethics Committee of the University of Extremadura. All procedures adhered to the ethical standards of this institution (Ref.: 181/2020), as well as to the principles established in the 1964 Declaration of Helsinki and its subsequent amendments or comparable ethical guidelines.

2.4. Data Analysis

First, the sociodemographic characteristics of the participants were described. Subsequently, a descriptive analysis of the main variables was conducted, including the emotional and behavioral problems of minors in residential care, as well as the socio-familial difficulties identified in parents and/or caregivers (gender-based violence, mental health problems, substance use, intellectual disability, and severe economic hardship), which were considered relevant risk factors in this context.

Inferential analyses were then performed. Given the nature of the variables and the sample size, the use of parametric tests was deemed appropriate. Prior to conducting the mean comparisons, assumptions of normality and homogeneity of variance were examined. Independent-samples t tests were used to determine whether statistically significant differences existed in the emotional and behavioral problems of minors according to sex, age, and the socio-familial variables. Effect sizes for mean comparisons were estimated using Cohen’s d. To reduce the risk of Type I error due to multiple comparisons, p values were adjusted using the Benjamini–Hochberg procedure (Benjamini and Hochberg 1995).

In addition, multivariable linear regression analyses were conducted to assess the extent to which sex, age, socio-familial risk factors, length of stay, and maltreatment history were independently associated with the levels of emotional and behavioral difficulties among the minors. All predictors were entered simultaneously in each model. For the regression analyses, age was coded as 0 = 6–11 years and 1 = 12–18 years; sex was coded as 0 = female and 1 = male; family-risk variables were coded as 0 = absence and 1 = presence; length of stay was entered as a continuous variable in months; and maltreatment history was coded as 0 = other protection measures and 1 = child maltreatment history. Regression assumptions were examined through residual plots and diagnostic statistics, including checks for multicollinearity.

Statistical processing of the data was carried out using SPSS for Windows, version 29.

3. Results

Below, we present the results obtained after the application of the instruments.

3.1. Descriptive Analysis

Table 1. Descriptive Statistics of Emotional and Behavioral Difficulties in Minors.

	Normal Range		Borderline Range		Abnormal Range		M	SD
	n	%	n	%	n	%
Total Difficulties Scale	95	45.2	75	35.7	40	19.0	15.45	5.83
Emotional Problems	136	64.8	37	17.6	37	17.6	3.75	2.53
Conduct Problems	84	40.0	43	20.5	83	39.5	3.70	2.11
Hyperactivity	114	54.3	53	25.2	43	20.5	4.64	2.35
Peer Problems	101	48.1	72	34.3	37	17.6	3.36	1.88
Prosocial Behavior	136	64.8	35	16.7	39	18.6	6.68	2.49

presents the descriptive analysis of the emotional and behavioral difficulties assessed with the SDQ among children and young people in residential care, corresponding to Hypothesis 1. These results provide an overview of the distribution of scores across the different dimensions of the instrument, as well as the general profile of difficulties in the sample.

The results obtained using the SDQ show that the mean score on the Total Difficulties scale falls within the normal range, although close to the borderline threshold (M = 15.45; SD = 5.83). Regarding the distribution of the sample, 45.2% (n = 95) of minors in residential care fall within the normal range for this scale, whereas 35.7% (n = 75) fall within the borderline range and 19% (n = 40) within the abnormal range. Overall, the proportion of minors in the borderline and abnormal ranges exceeds that of those within the normal range, indicating that a considerable percentage of children and young people in residential care present emotional and behavioral problems of varying intensity.

With respect to Emotional Problems, the mean score lies within the normal range (M = 3.75; SD = 2.53). The distribution shows that 64.8% (n = 136) of minors present emotional levels considered normal, while 17.6% (n = 37) fall within the borderline range and another 17.6% (n = 37) within the abnormal range. Although most participants score within normal limits on this dimension, it is noteworthy that 74 minors present emotional difficulties of different degrees of severity.

Regarding Conduct Problems, the data indicate that the mean score is close to the borderline range (M = 3.70; SD = 2.11). The distribution shows that 40% (n = 84) of minors in residential care fall within normal limits, whereas 20.5% (n = 43) fall within the borderline range and 39.5% (n = 83) within the abnormal range. Consequently, the proportion of minors scoring in the borderline and abnormal ranges clearly exceeds that of those in the normal range. Specifically, of the 210 minors evaluated, 83 present significant conduct problems and 43 show moderate difficulties.

Regarding Hyperactivity, the mean score is also close to the borderline range (M = 4.64; SD = 2.35). A total of 54.3% (n = 114) of minors show levels considered normal, whereas 25.2% (n = 53) fall within the borderline range and 20.5% (n = 43) within the abnormal range. Although most participants fall within the normal range, it is noteworthy that 96 children and young people under protective measures present hyperactivity-related manifestations of varying intensity.

With respect to Peer Problems, results show that the mean score is close to the borderline range (M = 3.36; SD = 1.88). A total of 48.1% (n = 101) of minors in residential care fall within normal limits, whereas 34.3% (n = 72) fall within the borderline range and 17.6% (n = 37) within the abnormal range. Thus, the proportion of minors scoring in the borderline and abnormal ranges is very similar to that of those within the normal range. In total, 109 children and young people present difficulties in peer relationships with varying degrees of severity.

Regarding Prosocial Behavior, results indicate that the mean score falls within the normal range (M = 6.68; SD = 2.49). A total of 64.8% (n = 136) of minors show adequate prosocial levels, while 16.7% (n = 35) fall within the borderline range and 18.6% (n = 39) within the abnormal range. Although most participants display appropriate prosocial behavior, it is important to highlight that 74 minors under protective measures present reduced levels in this dimension.

Overall, the results indicate that a considerable proportion of children and young people in residential care present notable difficulties in several areas assessed by the SDQ: conduct problems (39.5%), hyperactivity manifestations (20.5%), emotional difficulties (17.6%), and peer problems (17.6%).

Next,

Table 2. Distribution of the Sample Across Family-Related Variables.

Family Variable	Yes	No
Gender-based violence	48 (22.9%)	162 (77.1%)
Mental health problems	66 (31.4%)	144 (68.6%)
Substance use	92 (43.8%)	118 (56.2%)
Intellectual disability (suspected and/or diagnosed)	39 (18.6%)	171 (81.4%)
Severe family economic hardship	78 (37.1%)	132 (62.9%)

presents the incidence of the various family-related difficulties (gender-based violence, mental health problems, substance use, intellectual disability, and severe economic hardship) that may act as risk factors for the emotional and behavioral symptomatology exhibited by children and young people, corresponding to Hypothesis 2.

The data reveal a significant presence of family-related risk factors among minors under protective measures, which may contribute to the emotional and behavioral symptomatology observed.

Regarding gender-based violence within the family environment, the results indicate that 48 minors (22.9%) have been witnesses and/or victims of this form of violence. Of these, 81.25% (n = 39) witnessed the aggressions directed toward their mothers—observing both physical and emotional consequences—whereas 18.75% (n = 9) experienced direct victimization, having intervened during the assaults in an attempt to protect their mothers.

With respect to mental health problems in parents or caregivers, a notable prevalence was observed: 66 cases (31.4%) presented some type of psychological disorder. The distribution shows a higher presence of difficulties among mothers (71.21%, n = 47) than among fathers (28.79%, n = 19). Among mothers, the most prevalent conditions were anxiety disorders, depression, trauma- and stressor-related disorders, bipolar disorders, and personality disorders (primarily borderline and avoidant). In contrast, among fathers, the most prevalent difficulties included psychotic disorders, personality disorders (paranoid, schizoid, and antisocial), and bipolar disorders.

Additionally, in 19.7% of families (n = 13), both parents or caregivers presented mental health problems, which increases the complexity and risk level of the family environment.

Regarding the consumption of toxic substances, the results indicate a high prevalence in the families of minors in residential care. A total of 92 parents and/or caregivers (43.8%) presented some degree of substance use. The distribution shows a higher presence of this risk factor among fathers (59.78%, n = 55) compared to mothers (40.22%, n = 37). Furthermore, in 42.39% of families (n = 39), both caregivers were substance users, further increasing the vulnerability of the family context.

Cases were also identified in which at least one caregiver presented some degree of intellectual disability or where there was a professional suspicion of below-average cognitive functioning, according to the assessment of Social Services. This factor was present in 18.6% of cases (n = 39), with a higher prevalence among mothers (71.79%, n = 28) than fathers (28.21%, n = 11). The distribution of cognitive impairment levels showed that: in 38.4% of families (n = 15) there was suspected intellectual disability or cognitive deterioration; 28.2% (n = 11) had a diagnosis of borderline intellectual functioning; 26.6% (n = 10) presented mild intellectual disability; and 7.6% (n = 3) showed moderate intellectual disability.

Finally, economic situation also constituted a relevant risk factor. In 78 families (37.1%), the economic situation was described as very precarious. The data reflect income instability, difficulties in meeting basic needs, and a significant risk of housing loss. In addition, poor housing conditions were frequently observed, including insufficient space, overcrowding, deficiencies in safety and hygiene, and lack of basic equipment or furniture.

3.2. Inferential Analysis

To test Hypothesis 3, a mean comparison analysis was conducted with the aim of determining whether significant differences existed in the emotional and behavioral symptomatology of minors in residential care according to age (6–11 years vs. 12–18 years), sex, and the presence of family risk factors (gender-based violence, mental health problems, substance use, intellectual disability, and economic hardship).

Table 3. Mean Comparison of Emotional and Behavioral Difficulties by Age and Sex.

	Female		Male				6–11 Years		12–18 Years
	M	SD	M	SD	t	d	M	SD	M	SD	t	d
Total Difficulties	16.48	5.62	14.48	5.88	2.52 *	0.35	14.12	4.26	15.97	6.27	−2.09 *	−0.32
Emotional Problems	4.54	2.52	3.01	2.32	4.57 **	0.63	1.97	2.02	4.45	2.37	−7.11 **	−1.09
Conduct Problems	3.57	1.97	3.81	2.24	−0.846	−0.11	4.05	1.78	3.56	2.22	1.53	0.23
Hyperactivity	4.63	2.27	4.66	2.44	−0.092	−0.01	3.98	1.78	4.90	2.50	−2.57 *	−0.40
Peer Problems	3.75	1.85	3.00	1.85	2.92 *	0.41	4.12	1.58	3.07	1.91	3.75 **	0.58
Prosocial Behavior	7.39	2.00	6.00	2.71	4.25 **	0.58	5.63	2.75	7.09	2.25	−3.95 **	−0.61

Note: d = Cohen’s d. p values were adjusted using the Benjamini–Hochberg procedure. * p < 0.05; ** p < 0.01.

shows that statistically significant sex differences were found in Emotional Problems, t(207) = 4.57, p = 0.006, d = 0.63; Peer Problems, t(207) = 2.92, p = 0.015, d = 0.41; Prosocial Behavior, t(207) = 4.25, p = 0.006, d = 0.58; and the Total Difficulties score, t(207) = 2.52, p = 0.036, d = 0.35. The results indicate that girls present higher levels of emotional problems, peer problems, and overall difficulties than boys; however, they also score higher in prosocial behavior.

Significant differences were also observed as a function of age, specifically in Emotional Problems, t(208) = −7.11, p = 0.006, d = −1.09; Hyperactivity, t(208) = −2.57, p = 0.015, d = −0.40; Peer Problems, t(208) = 3.75, p = 0.006, d = 0.58; Prosocial Behavior, t(208) = −3.95, p = 0.006, d = −0.61; and in the Total Difficulties scale, t(208) = −2.09, p = 0.039, d = −0.32. As shown in Table 3, adolescents aged 12–18 years present more emotional problems, higher levels of hyperactivity, and greater overall difficulties than children aged 6–11 years. However, they also show higher prosocial behavior scores. In contrast, younger children (6–11 years) present more peer-related difficulties.

A mean comparison analysis was also conducted to examine whether significant differences existed in the emotional and behavioral problems of minors as a function of the different socio-familial variables studied (see

Table 4. Mean Comparison of Emotional and Behavioral Difficulties According to Exposure to Gender-Based Violence and Parental Mental Health Problems.

	Gender-Based Violence						Parental Mental Health Problems
	Yes				No		Yes				No
	M	SD	M	SD	t	d	M	SD	M	SD	t	d
Total Difficulties	17.81	5.33	14.76	5.81	−3.22 **	0.53	17.29	6.49	14.61	5.30	−3.15 *	0.47
Emotional Problems	4.15	2.36	3.64	2.58	−1.22	0.20	4.42	2.63	3.44	2.43	−2.64 *	0.39
Conduct Problems	4.47	2.34	3.48	1.99	−2.86 *	0.48	4.21	2.31	3.46	1.98	−2.42 *	0.36
Hyperactivity	5.28	2.23	4.45	2.36	−2.13 *	0.36	4.85	2.49	4.55	2.29	−0.856	0.13
Peer Problems	3.91	1.96	3.19	1.83	−2.34 *	0.39	3.80	1.91	3.16	1.83	−2.32 *	0.34
Prosocial Behavior	6.72	1.85	6.65	2.65	−0.167	0.03	6.71	2.30	6.66	2.57	−0.141	0.02

Note: d = Cohen’s d. p values were adjusted using the Benjamini–Hochberg procedure. * p < 0.05; ** p < 0.01.

,

Table 5. Mean Comparison of Emotional and Behavioral Difficulties According to Parental Substance Use and Intellectual Disability.

	Parental Substance Use						Parental Intellectual Disability
	Yes				No		Yes				No
	M	SD	M	SD	t	d	M	SD	M	SD	t	d
Total Difficulties	16.98	6.07	14.26	5.35	−3.43 **	0.48	16.13	5.93	15.30	5.80	−0.802	0.14
Emotional Problems	4.26	2.63	3.36	2.38	−2.60 *	0.36	3.77	2.82	3.75	2.46	−0.046	0.01
Conduct Problems	4.16	2.16	3.33	2.01	−2.88 *	0.40	4.18	2.13	3.58	2.10	−1.59	0.28
Hyperactivity	5.07	2.40	4.31	2.27	−2.32 *	0.33	4.72	2.51	4.63	2.32	−0.220	0.04
Peer Problems	3.49	1.99	3.26	1.79	−0.864	0.12	3.46	1.86	3.34	1.89	−0.366	0.06
Prosocial Behavior	6.58	2.37	6.75	2.57	0.514	−0.07	5.87	2.59	6.86	2.43	2.26 *	−0.40

Note: d = Cohen’s d. p values were adjusted using the Benjamini–Hochberg procedure. * p < 0.05; ** p < 0.01.

and

Table 6. Mean Comparison of Emotional and Behavioral Difficulties According to Family Economic Situation.

	Family Economic Hardship
	Yes				No
	M	SD	M	SD	t	d
Total Difficulties	14.87	6.49	15.80	5.39	1.11	−0.16
Emotional Problems	3.67	2.61	3.80	2.48	0.376	−0.05
Conduct Problems	3.45	2.03	3.84	2.15	1.30	−0.19
Hyperactivity	4.63	2.49	4.65	2.27	0.069	−0.01
Peer Problems	3.13	1.98	3.50	1.81	1.38	−0.20
Prosocial Behavior	6.79	2.51	6.61	2.47	−0.531	0.07

Note: d = Cohen’s d. p values were adjusted using the Benjamini–Hochberg procedure.

).

As shown in Table 4, the results indicate statistically significant differences among minors depending on exposure to gender-based violence. Significant differences were found in Conduct Problems, t(207) = −2.86, p = 0.017, d = 0.48; Hyperactivity, t(207) = −2.13, p = 0.040, d = 0.36; Peer Problems, t(207) = −2.34, p = 0.044, d = 0.39; and the Total Difficulties score, t(207) = −3.22, p = 0.006, d = 0.53. These results indicate that minors who have been witnesses and/or victims of gender-based violence present more conduct problems, higher levels of hyperactivity, greater difficulties in peer relationships, and more global difficulties than those who have not experienced such situations.

Furthermore, the data also show significant differences as a function of the presence of mental health problems in parents or caregivers. Significant differences were found in Emotional Problems, t(208) = −2.64, p = 0.029, d = 0.39; Conduct Problems, t(208) = −2.42, p = 0.040, d = 0.36; and Peer Problems, t(208) = −2.32, p = 0.044, d = 0.34; as well as in the Total Difficulties scale, t(208) = −3.15, p = 0.015, d = 0.47. Taken together, these results indicate that minors living with at least one parent or caregiver with psychological disorders show higher emotional, behavioral, and relational problems, as well as greater overall difficulties, compared to those whose families do not present mental health issues.

Table 5 presents the results regarding the influence of toxic substance use (alcohol and/or drugs) by parents or caregivers. The analyses show statistically significant differences in Emotional Problems, t(208) = −2.60, p = 0.033, d = 0.36; Conduct Problems, t(208) = −2.88, p = 0.015, d = 0.40; Hyperactivity, t(208) = −2.32, p = 0.044, d = 0.33; and the Total Difficulties score, t(208) = −3.43, p = 0.006, d = 0.48. These results indicate that minors living with at least one parent or caregiver who uses substances present higher levels of emotional difficulties, behavioral problems, and hyperactivity, as well as greater overall difficulties, compared to those whose caregivers do not use toxic substances.

In contrast, no significant differences were found in emotional or behavioral difficulties as a function of the presence of intellectual disability in parents or caregivers, with the exception of Prosocial Behavior, where significant differences were observed, t(208) = 2.26, p = 0.044, d = −0.40. Specifically, minors whose parents or caregivers do not present intellectual disability show higher levels of prosocial behavior than those whose caregivers do present some degree of intellectual disability.

Finally, Table 6 shows that no statistically significant differences were found in the emotional and behavioral difficulties of minors as a function of family economic situation. These results indicate that, although economic hardship constitutes a relevant contextual risk factor, no significant differences were observed in this study in the levels of emotional and behavioral symptomatology associated with this variable. The effect sizes were small across all dimensions.

To test Hypothesis 4, multivariable linear regression analyses were conducted to determine the extent to which age, sex, family risk factors, length of stay, and maltreatment history were independently associated with the emotional and behavioral symptomatology of children and young people in residential care (see

Table 7. Multivariable Linear Regression Models for Emotional and Behavioral Difficulties of Minors.

Emotional and Behavioral Difficulties	Model R2	Age β (t)	Sex β (t)	GBV β (t)	Mental Health β (t)	Substance Use β (t)	ID β (t)	Economic Hardship β (t)	Length of Stay β (t)	Maltreatment History β (t)
Total Difficulties	0.122	0.191 (2.748) *	−0.044 (−0.630)	0.173 (2.255) *	0.070 (0.834)	0.075 (0.913)	0.057 (0.786)	−0.122 (−1.723)	0.031 (0.431)	0.013 (0.192)
Emotional Problems	0.294	0.422 (6.788) **	−0.241 (−3.894) **	−0.057 (−0.830)	0.052 (0.689)	0.140 (1.904)	0.036 (0.563)	−0.060 (−0.952)	−0.067 (−1.044)	0.020 (0.329)
Conduct Problems	0.121	−0.081 (−1.164)	0.101 (1.468)	0.165 (2.152) *	0.043 (0.507)	0.120 (1.467)	0.144 (2.295) *	−0.162 (−1.991)	−0.088 (−1.229)	0.046 (0.672)
Hyperactivity	0.081	0.189 (2.667) *	0.065 (0.914)	0.144 (1.834)	−0.101 (−1.177)	0.169 (2.017) *	0.050 (0.673)	−0.063 (−0.865)	−0.015 (−0.208)	−0.054 (−0.776)
Peer Problems	0.154	−0.290 (−4.258) **	−0.193 (−2.852) *	0.090 (1.198)	0.160 (1.946)	−0.032 (−0.397)	0.010 (0.146)	−0.071 (−1.021)	0.022 (0.320)	0.015 (0.224)
Prosocial Behavior	0.170	0.207 (3.066) *	−0.268 (−3.991) **	−0.050 (−0.672)	0.013 (0.159)	−0.028 (−0.354)	−0.163 (−2.322) *	0.097 (1.408)	0.047 (0.671)	−0.068 (−1.028)

Note: Values are standardized regression coefficients, with t values in parentheses. Model R² refers to the full multivariable model for each SDQ outcome. All predictors were entered simultaneously. GBV = gender-based violence; ID = parental intellectual disability. Statistical significance was adjusted using the Benjamini–Hochberg procedure. * p < 0.05; ** p < 0.01.

). All predictors were entered simultaneously in each model.

Table 7 shows that the full multivariable models explained between 8.1% and 29.4% of the variance in the different SDQ dimensions. Specifically, the model explained 12.2% of the variance in the Total Difficulties scale, 29.4% in Emotional Problems, 12.1% in Conduct Problems, 8.1% in Hyperactivity, 15.4% in Peer Problems, and 17.0% in Prosocial Behavior.

Age was independently associated with Total Difficulties, β = 0.191, t = 2.748; Emotional Problems, β = 0.422, t = 6.788; Hyperactivity, β = 0.189, t = 2.667; Peer Problems, β = −0.290, t = −4.258; and Prosocial Behavior, β = 0.207, t = 3.066. Specifically, older age was associated with higher levels of total difficulties, emotional problems, hyperactivity, and prosocial behavior, whereas younger age was associated with greater peer problems. Sex was also independently associated with Emotional Problems, β = −0.241, t = −3.894; Peer Problems, β = −0.193, t = −2.852; and Prosocial Behavior, β = −0.268, t = −3.991. These results indicate that girls presented higher levels of emotional problems and peer problems, as well as higher prosocial behavior scores.

Regarding family risk factors, gender-based violence was independently associated with Total Difficulties, β = 0.173, t = 2.255, and Conduct Problems, β = 0.165, t = 2.152, indicating that minors who had witnessed and/or experienced gender-based violence presented higher overall difficulties and conduct problems. Parental substance use was independently associated with Hyperactivity, β = 0.169, t = 2.017, indicating higher hyperactivity scores among minors living with at least one parent or caregiver who used substances.

Parental intellectual disability was independently associated with Conduct Problems, β = 0.144, t = 2.295, and Prosocial Behavior, β = −0.163, t = −2.322. Specifically, minors whose parents or caregivers presented some degree of intellectual disability showed higher conduct problems and lower prosocial behavior scores. In contrast, parental mental health problems, family economic hardship, length of stay, and maltreatment history were not significantly associated with any of the emotional or behavioral difficulties evaluated after controlling for the remaining variables.

4. Discussion

The results obtained support that minors in residential care present a high prevalence of emotional and behavioral problems. These difficulties are associated with several socio-familial risk factors which, together with age and sex, show specific associations with different dimensions of emotional and behavioral symptomatology.

The data show that more than half of the minors assessed present emotional and behavioral problems, a prevalence clearly higher than that observed in the general population. Although rates vary across countries, emotional and behavioral problems in the general population typically range between 10% and 22% (Cui et al. 2021; Doyle et al. 2023; Al-Mamun et al. 2024; Labib et al. 2024). When examining the results in greater detail, a higher incidence of behavioral symptoms is observed among minors in residential care, including hyperactivity, attention deficits, impulsivity, oppositional behaviors, irritability, and difficulties with self-regulation. These findings are consistent with those reported by Engel de Abreu et al. (2023) and Rohanachandra et al. (2022). Additionally, although less prevalent, a significant proportion of minors also exhibit emotional problems—anxiety, sadness, discouragement, nervousness, and excessive worry—results that align with findings by González-García et al. (2023), Moreno-Manso et al. (2020), and Muris et al. (2003). This emotional vulnerability may be related to early exposure to adverse experiences. Consequently, Hypothesis 1 is supported, highlighting the need for comprehensive interventions that include strengthening emotional regulation and enhancing resilience to trauma.

Regarding family risk factors, the results show a significant presence of adverse socio-familial circumstances among minors in residential care. A notable finding is the high prevalence of gender-based violence in the family environment, as approximately one quarter of minors had been witnesses and/or direct victims, most having witnessed aggression toward their mothers and, in some cases, becoming victims themselves when attempting to intervene. This pattern is in line with studies showing that exposure to family violence is associated with behavioral problems, difficulties in emotional regulation, and symptoms of anxiety and depression (Ortiz-Jiménez et al. 2025).

Another particularly relevant factor is the presence of mental health problems in caregivers, affecting 31.4% of the families of origin, with higher prevalence among mothers and predominance of affective, anxiety, or personality disorders. This figure even exceeds the prevalence observed in the general population of caregivers, estimated at around 22% (Grant et al. 2024). Likewise, caregiver substance use showed a high prevalence, particularly among fathers, which is consistent with findings from other studies (Castello et al. 2022; McGovern et al. 2023; Singh-Charan et al. 2023). These studies suggest that minors exposed to parental substance use may present greater emotional and behavioral vulnerability, associated with unstable parenting patterns and dysfunctional family environments.

Unlike findings reported in other research, such as the study by Slayter and Jensen (2019), the presence of disability in caregivers was not associated in our study with a generalized increase in emotional and behavioral difficulties in minors, although it was associated with specific difficulties in conduct problems and lower prosocial behavior. A similar pattern appears with regard to family economic situation, as we cannot confirm that it is related to emotional and behavioral symptomatology of the minors. However, in studies such as that by Badini et al. (2023), economic hardship does appear to be linked to increased risk of emotional and behavioral problems.

The results also show the association of sex and age with emotional and behavioral symptomatology. Girls display higher scores in emotional problems, consistent with research indicating a greater tendency toward internalizing difficulties among females (Fernández-Daza 2020; Gutterswijk et al. 2022). Regarding age, higher symptomatology was observed among adolescents, as minors aged 12 to 18 years present higher levels of emotional problems, hyperactivity, and overall difficulties compared to children aged 6 to 11. These findings are aligned with previous evidence (Andreopoulou et al. 2020; Sanchis-Sanchis et al. 2020). In contrast, younger children show more difficulties in peer relationships, possibly due to their limited capacity to adapt to the social and behavioral demands of residential settings.

The effect-size estimates suggest that most statistically significant differences were small to moderate in magnitude. The strongest effect was observed for age differences in Emotional Problems, indicating particularly higher emotional symptoms among adolescents. Sex differences were also notable for Emotional Problems and Prosocial Behavior. Family-risk variables such as gender-based violence, parental mental health problems, and parental substance use showed small-to-moderate associations with several dimensions, whereas economic hardship showed negligible-to-small effects across dimensions. These findings suggest that the observed differences should be interpreted not only in terms of statistical significance, but also according to their magnitude.

The multivariable models further indicate that both sex and age were independently associated with several dimensions of emotional and behavioral functioning. Age was independently associated with total difficulties, emotional problems, hyperactivity, peer problems, and prosocial behavior, whereas sex was independently associated with emotional problems, peer problems, and prosocial behavior. These findings converge with the results of Ahmed et al. (2021) and Sekaran et al. (2024), although they differ from those of Aboobaker et al. (2019), who reported that sex is more strongly associated with behavioral difficulties and that age is more closely linked to emotional problems.

Regarding family characteristics, the multivariable models indicate that exposure to gender-based violence, parental substance use, and parental intellectual disability show specific independent associations with different dimensions of emotional and behavioral functioning. Being a witness and/or victim of gender-based violence was independently associated with higher total difficulties and conduct problems, findings supported by previous studies (Almış et al. 2020; Clark and Hankin 2024). Similarly, Gutiérrez-Mercado (2023) reports that exposure to gender-based violence is associated with higher rates of emotional disorders, post-traumatic symptoms, and suicidal behavior.

Parental substance use was independently associated with higher hyperactivity. This finding is consistent with evidence showing that children of parents who use toxic substances present heightened psychological vulnerability and an increased risk of emotional and behavioral difficulties (McGovern et al. 2023). Parental intellectual disability was independently associated with higher conduct problems and lower prosocial behavior, suggesting that this family factor may be related to specific dimensions of behavioral and social adjustment rather than with a generalized increase in emotional and behavioral difficulties.

In contrast, parental mental health problems, family economic hardship, length of stay, and maltreatment history were not significantly associated with the outcomes assessed once the remaining variables were controlled. Therefore, although some of these variables showed significant differences in the mean comparison analyses, their independent contribution was attenuated in the multivariable models. These findings reinforce the need to interpret the results as independent statistical associations rather than causal effects.

Taken together, the findings suggest that emotional and behavioral difficulties among minors in residential care are associated with family variables of an emotional and relational nature—especially gender-based violence and parental substance use—as well as with individual characteristics such as age and sex. Overall, these results should be interpreted as statistical associations, not as evidence of direct explanatory relationships.

From an intervention perspective, the findings suggest the relevance of considering the relational and emotional context, consistent with the findings of Offerman et al. (2022), who highlight that the quality of the family environment plays an important role in the psychological adjustment of minors. However, given that attachment quality and emotional security were not directly measured in the present study, these aspects should be interpreted as implications supported by previous literature rather than as direct findings derived from the data (Magalhães et al. 2021; Pinheiro et al. 2024). This underscores the need for therapeutic and preventive interventions oriented toward trauma history, emotional regulation, and the strengthening of safe and stable relational environments.

A strength of the present study is that it examines emotional and behavioral difficulties in residential care while considering several socio-familial risk factors jointly rather than in isolation. This is relevant because previous research has documented the high psychological vulnerability of children and adolescents in residential care (Sainero et al. 2014; Westlake et al. 2023), but the field remains methodologically heterogeneous and has often focused on other protection modalities, particularly family foster care (Maguire et al. 2024). Moreover, although studies in residential care have described emotional and behavioral difficulties in this population (González-García et al. 2023; Moreno-Manso et al. 2020; Soriano-Díaz et al. 2023), fewer studies have integrated several family-related adversities together with individual characteristics such as age and sex. This approach provides a more comprehensive descriptive and associative view of the factors linked to emotional and behavioral functioning among minors in residential care.

Despite the relevance of the findings, this study presents several limitations that should be taken into account when interpreting the results. First, the cross-sectional design limits the ability to establish causal relationships between socio-family risk factors and children’s emotional and behavioral symptoms. Although the analyses identify significant associations and examine the relative contribution of each variable, the directionality of the relationships cannot be determined. In addition, the use of different SDQ informants should be considered when interpreting the results. Although this procedure was adopted to ensure age-appropriate assessment across the broad age range of the sample, some responses were based on key-worker reports, whereas others were based on self-report. Future research should incorporate longitudinal designs and multi-informant assessment, including caregivers, educators, clinical professionals, and the minors themselves, to obtain a more comprehensive evaluation of emotional and behavioral difficulties.

A second limitation concerns the limited consideration of contextual characteristics inherent to residential care settings. Although length of stay and maltreatment history were included in the multivariable models, other elements such as the emotional climate of the centers, caregiver stability, the number of previous placements, the educator–child ratio, and the educational or therapeutic practices implemented may significantly influence children’s psychological adjustment. Including these variables in future studies would provide a more comprehensive understanding of how pre-placement family risks interact with the conditions of the child protection system.

Additionally, the dichotomous coding of family risk variables—such as exposure to gender-based violence, parental mental health problems, substance use, or maltreatment history—restricts the analysis of their severity, chronicity, frequency, and developmental timing. Although the information was obtained from official child protection case records and the reports available in each case file, no formal inter-rater reliability procedure was conducted. Therefore, the interpretation of these variables should take into account that they reflect the documented presence or absence of each risk factor, rather than a detailed assessment of its intensity or developmental course. Since the intensity and persistence of these adverse conditions may differentially affect children’s developmental outcomes, future research should employ more nuanced and sensitive measures capable of capturing the complexity of these experiences. In this regard, future studies could complement case-record information with more detailed and longitudinal measures of parental substance use and apply advanced statistical approaches to identify profiles of vulnerability among minors in residential care. Recent addiction-methods research, including studies using statistical modeling to identify markers associated with nicotine vaping, illustrates the potential value of these approaches for examining substance-use-related risk with greater precision (Zhao et al. 2025).

5. Conclusions

The study shows that children and adolescents in residential care exhibit high rates of emotional and behavioral difficulties, substantially exceeding those reported in the general population. These challenges were associated with socio-family risk factors—particularly exposure to gender-based violence, parental substance use, and parental intellectual disability—as well as individual characteristics such as age and sex. In contrast, parental mental health problems, family economic situation, length of stay, and maltreatment history did not show significant independent associations with the outcomes assessed in the multivariable models.

The findings highlight the importance of affective bonds as a relevant protective factor to be considered in intervention. Although the quality of these relationships was not directly assessed in the present study, previous literature suggests that strengthening stable and secure relationships—both within the family of origin and with caregivers in residential settings—may constitute a key component for promoting psychological well-being.

Overall, the results underscore the need for educational and therapeutic interventions within residential care that target emotional regulation, behavioral difficulties, and trauma-related experiences, particularly those involving gender-based violence and parental substance use, as well as difficulties associated with lower prosocial behavior and conduct problems. Enhancing relational stability and fostering emotional safety constitute essential elements for improving the adjustment and developmental outcomes of children in residential care.