How early is early intervention and who should get it? Contested meanings in determining thresholds for intervention

The latest wave of reforms of the child protection system in Australia have been based on attempts to provide support to all families with vulnerable children, rather than increasing surveillance of ‘at risk’ families and forensic responses to incidents of maltreatment. This includes a drive to widen the remit of child protection from the statutory child protection agency and involve other government agencies such as health and education as well as the non-government sector in child protection. This paper reports on the effects of one such reform, the NSW initiative Keep Them Safe. It focuses on the classification of families as needing either early intervention or intensive support, using thematic analysis of qualitative interview data.MethodInterviews and focus groups were conducted with practitioners and managers from human service agencies (total n = 115), and discussed their perceptions of the initiative and the changes it had introduced to service delivery.FindingsPractitioners discussed family needs in ways which contested the policy meanings of ‘early intervention’: whether families are conceptualised in terms of their needs or risk; whether engagement with services should be voluntary or mandated; and whether the agencies to support them should be the statutory agency or an NGO. The implications for these tensions, in terms of policy and practice, are discussed.     

Recovery coaches and the stability of reunification for substance abusing families in child welfare

Substance abuse is a long-standing challenge for child welfare systems. Parental substance abuse disrupts family stability, family cohesion, and jeopardizes the well-being of children. In the current study we test an intervention to improve child welfare outcomes for substance abusing families, specifically the probability of families achieving a stable (at least 12 months) reunification. The intervention was an integrated case management model where recovery coaches were appointed to substance abusing parents associated with an open foster care placement. A diverse group of families (n = 1623) were randomly assigned to either a control group (services as usual) or an experimental group (services as usual plus a recovery coach). Multinomial logistic regression indicated that substance abusing parents associated with a recovery coach were significantly more likely to achieve a stable reunification as compared with similar families in the control group.     

The relationship between parent immigration status and concrete support service use among Latinos in child welfare: Findings using the National Survey of Child and Adolescent Well-being (NSCAWII)

The underutilization of concrete services by immigrants is widely documented across several service sectors, yet evidence is lacking on the use of such services among immigrants reported to child welfare for the purposes of reducing maltreatment. It has been suggested that Latino immigrants involved with the child welfare system may face steep challenges to receiving needed services due to issues surrounding legal status, language and cultural barriers. The purpose of this study was to determine whether referral to and receipt of concrete services by Latino families reported to child welfare agencies, was associated with legal immigration status. The sample included children of Latino parents who participated in the second National Survey of Child and Adolescent Well-being (NSCAWII), who remained in the home following a child welfare investigation (n = 561). Over a third (37%) of Latino families were referred for at least one concrete service, yet only 17% received any. Weighted logistic regression models showed that families in which the primary caregiver was undocumented had significantly lower odds (OR = .24) of receiving services once referred. Families who had trouble paying for basic necessities (OR = 7.52), those with active domestic violence in the home (OR = 4.98), and those receiving ongoing child welfare services (OR = 4.52) had increased odds of referral for services by the caseworker. The odds of receiving services increased when the primary caregiver was unemployed (OR = 5.24), when there was domestic violence in the home (OR = 4.59), and with the receipt of child welfare agency services (OR = 8.83). There appears to be an unmet need for concrete services among Latinos investigated by child welfare, as demonstrated in the gap between overall service referral and receipt. A parent's legal status may be one reason for that unmet need, implying that children of undocumented parents are less likely to have basic needs met to mitigate economic stress and reduce maltreatment risk upon contact with child welfare. Policy recommendations and implications for child welfare practice are discussed.     

An intensive mental health home visiting model for two at-risk early childhood populations

This study examined proximal outcomes of a mental health home visiting model for two populations at risk for child maltreatment: families with young children referred by child protective services (CPS) and at-risk pregnant women (Prenatal) referred by community agencies. Family- and caregiver-level outcomes were measured using the Family Assessment Form (FAF). Families (n = 215) showed significant improvement in all eight family functioning factors over the course of their participation in mental health home visiting services. Initially, CPS-referred families (n = 84) scored higher on the FAF measure of Interactions between Caregivers, indicating greater conflict between caregivers in the family. Prenatal referred families (n = 131) were at greater risk initially on Housing. Prenatal-referred families demonstrated greater risk reduction on measures of Supports to Caregivers, Developmental Stimulation, Caregiver Personal Characteristics and Housing. In addition, all families demonstrated significant improvements in functioning on 11 of 12 items comprising the Caregiver Personal Characteristics factor. Overall, CPS-referred families scored at higher risk on items reflecting externalizing problems, while Prenatal-referred families showed greater improvement on items reflecting internalizing problems. This model was successful in reducing risk factors and promoting protective factors for CPS-referred and Prenatal at-risk families. Implications and future directions are discussed.     

Finish what you start: A study of Design Team change initiatives' impact on agency climate

This study employs a multi-site longitudinal design to examine the effect of a Design Team intervention on organizational climate. Thirteen private, not-for-profit child welfare agencies from one state participated in a Design Team intervention to address workforce needs. A total of 407 workers from those agencies responded pre and post intervention to a survey that measures worker perceptions of the psychological climate of their organization using the Parker Psychological Climate Survey. Workers in organizations that completed the Design Team intervention had statistically significant increases in three of the four dimensions of the Parker scale. On the role dimension, significant change was noted on all three subscales on the interaction between Time 1 and Time 2 (ambiguity: p = 0.012; conflict: p = 0.04; overload: p = 0.05). On the organization dimension, the justice and support subscales had significant differences in the desired direction (justice: p = 0.05; support: p = 0.03). On the supervisor dimension, significant change was observed in the desired direction for both the goal emphasis and work facilitation subscales (goal emphasis: p = 0.02; work facilitation: p = 0.00). Statistically significant improvements in the organizational climates of child welfare agencies suggest the benefit of future research to test the effectiveness of Design Team interventions in other service areas. These findings build on intervention research with organizations by linking the ability of an organization to fully implement a change initiative to their capacity to improve the workplace climate for employees.     

Programmatic effects to modify sources of financial support among Hispanic teenage mothers

This study evaluates whether a randomized school-based intervention for adolescent teenage mothers successfully helped participants achieve better financial independence and economic prospects. Project Mothers and Schools (PMAS) is an initiative intended to modify attitudes and beliefs related to self-sufficiency, parenting, educational goals and achievement, career goals, health behaviors, and interpersonal relationships with family members. PMAS participants were surveyed at baseline and 12 months after intervention enrollment to evaluate the receipt of financial support from 11 sources. The 48 “core group” participants received a basic level of services, whereas the 43 “core-plus group” participants received the same basic services and additional enhanced services. Wilcoxon signed-rank tests were used to identify statistically significant changes in participant responses from baseline to 12 months post-intervention. Both groups reported higher levels of employment 12 months post-intervention compared to baseline (z = − 3.162, p = 0.002 and z = − 2.646, p = 0.008, respectively). The core-plus group used more Food Stamps (z = − 2.673, p = 0.008) and reported receiving more child support (z = 2.236, p = 0.025) 12 months post-intervention versus baseline. PMAS benefited all participants by promoting employment 12 months post-intervention versus baseline. The enhanced services offered to the core-plus group appeared to also increase participant access to Food Stamps and child support, sources which assist participants down the path toward eventual financial independence.     

Contact with birth family in intercountry adoptions: Comparing families headed by sexual minority and heterosexual parents

Contact between adoptive families and birth families in the context of intercountry adoption, as well as adoption by sexual minorities (e.g., lesbians and gay men), represent understudied topics. In the current study, we examine the extent and type of contact with birth family in intercountry adoptive families headed by heterosexual and sexual minority parents. Data were drawn from the Modern Adoptive Families project, a nationwide, non-random survey of adoptive parents' beliefs and experiences that was conducted from 2012 to 2013. The current sample consisted of 479 families headed by heterosexual parents (H) and 38 families headed by sexual minority women (SM) whose oldest adopted child was younger than 18 years of age and who had been placed from another country. Although no family type difference was found in contact with birth family prior to or at the time of placement (H = 9.6%; SM = 13.2%), sexual minority respondents reported a higher level of contact with one or more members of their children”s birth families following adoptive placement than did heterosexual respondents (SM = 28.9%; H = 14.4%), as well as currently (SM = 21.1%; H 9.8%). They also reported more contact with their children”s birth mothers than did heterosexual parents, although no family type differences were found for contact with other birth family members. Policy and practice implications are discussed.     

Services use by children and parents in multiproblem families

BackgroundMultiproblem families are multi-users of psychosocial and health care services, but little is known about factors associated with their care utilization in the general population. The aim of this study was to assess which factors were associated with the overall and psychosocial care use of two members—i.e., child and parent—of each multiproblem family.MethodsDuring well-child visits or psychosocial care, we identified 354 children and their parents who had problems in several life domains (response 69.1%). We used multivariate stepwise backward logistic regression analyses to identify the factors related to their use of overall and psychosocial care.ResultsA child's overall care use was associated with greater social support from family and friends (odds ratio, OR, 95% confidence interval, CI; OR = 1.05, CI = 1.01–1.08) compared to less perceived social support; and with more psychosocial problems in the child (OR = 1.84, CI = 1.04–3.24). Child's psychosocial care use was more likely among older children (OR = 1.94, CI = 1.20–3.15); greater social support by family and friend (OR = 1.03, CI = 1.00–1.06); more psychosocial problems (OR = 1.75, CI = 1.04–2.97); and when there were more parenting concerns (OR = 1.19, CI = 1.06–1.33). Parental overall and psychosocial care use was more likely when the family experienced a higher number of life events (OR = 1.27, CI = 1.17–1.38, and OR = 1.39, CI = 1.25–1.55).ConclusionsCare use in multiproblem families is related to family factors as well as psychosocial problems. It may be possible to use these family risk factors to identify such families early, whose intensive care use is possibly explained by the relationship with inadequate use of social support.     

Is child protective services effective?

A number of studies have concluded that there is little observable connection between CPS involvement and improved outcomes for children and families. Evidence of CPS effectiveness is complicated by the presence of selection bias and difficulty controlling for confounding. To understand outcomes by group and intervention effects, comparable groups are necessary and difficult to ascertain using CPS administrative case record data. This study examines the causal effect of CPS involvement on the likelihood of future maltreatment using administrative case management records from July 1, 2010 and June 30, 2011. The current study accounts for differences in pre-existing condition between groups to establish sound estimates of CPS involvement effects. Logistic regression models were used to examine the difference in subsequent substantiated investigation between families with comparable risk and differing service recommendation (p = 0.83), recurrence among families with comparable risk, the same service recommendation that did or did not receive services (p = 0.83). Hazard models were used to explore risk of substantiated investigation among families with comparable risk and differing service recommendation (p = 0.77). Results indicate receipt of CPS services had no observable effect on recurrence of maltreatment overall and among families with similar levels of risk of recurrence. Further inquiry into worker attributes, decision-making, types of and quality of services offered to families could help explain the effective, or ineffectiveness, or services.     

Mental health services for children and caregivers remaining at home after suspected maltreatment

ObjectiveBehavioral problems are common among children remaining at home after suspected maltreatment, but the effectiveness of current mental health services to improve these behavioral problems is unknown. The objective was to determine whether receipt of child and caregiver mental health services was associated with improvements in behavioral problems in maltreated children remaining at home.MethodsWe retrospectively analyzed Second National Survey of Child and Adolescent Well-being data. We included 1117 children ages 2–17 remaining at home after a maltreatment investigation, excluding children with missing outcome, covariate, or survey weight data. We compared mean Child Behavioral Checklist (CBCL) change scores from baseline to 18 months between children who did and did not receive mental health services, before and after adjusting for child, caregiver, and child welfare agency factors using survey-weighted linear regression.ResultsNearly one-quarter (22.6%) of children and 16.0% of caregivers received mental health services. Children receiving services had worse unadjusted baseline and 18-month CBCL scores than children not receiving services (all P < 0.001). Adjusted CBCL change scores revealed behavioral worsening among children receiving services but improvement among children not receiving services (all P < 0.001). However, children had improved behavior, regardless of their own service receipt, if their caregivers received services and reported an absence of depression at 18 months.ConclusionsChildren receiving mental health services had worse behavioral changes than children not receiving services. Caregiver receipt of services was associated with improved child behavior, suggesting that a family-centered approach may be most influential in improving behavioral outcomes among this population.     

Parent perspectives on adoption preparation: Findings from the Modern Adoptive Families project

Preparing prospective adoptive parents for receiving a child into their family is an important task for child and family professionals. This study uses data from the recent Modern Adoptive Families survey to understand parent perspectives on their preparation for adoption. Logistic regression (n = 917) and qualitative thematic analysis of adoptive parents' comments about their adoption preparation were conducted to understand aspects of the preparatory experience that were satisfactory (n = 623) or dissatisfactory (n = 283). Child emotional and behavioral problems significantly predicted parent dissatisfaction with adoption preparation. Major themes for satisfactory aspects of preadoption services included receiving information about adoption in general and, more specifically, about clinical and special populations, and parenting tools and strategies. In addition, parents identified opportunities to connect with others touched by adoption and access to specialized post-adoption services as helpful. Conversely, parents expressed dissatisfaction when information was lacking or withheld, when they had quality concerns with the worker or agency, and when there was a dearth of services and supports following adoption. Implications of these findings include the need for adoption mental health competent training for adoption professionals to better support families preparing for or experiencing adoption.     

Practitioner views on the core functions of coaching in the implementation of an evidence-based intervention in child welfare

Despite coaching being identified as an important implementation strategy, scant information is available on the core functions of coaching, and few empirical studies are specific to coaching in the child welfare setting. This study explored practitioners' perceptions of the core functions of coaching by using semi-structured focus groups with coaches (n = 13) and interviews with coachees (n = 11) who were delivering Parent Management Training, Oregon (PMTO) model to families of children in foster care. Four themes were identified as the core functions of coaching: (1) supporting practitioners via strengths-oriented feedback; (2) promoting skill-building via collaboration and active learning strategies; (3) problem-solving for appropriate use and adaptation of the EBI with real-world cases, and (4) providing an accountability mechanism for high fidelity implementation. Collectively, this study's findings build knowledge on the core functions of coaching, which may be a critical strategy for integrating evidence-based interventions (EBI) into usual practice in child welfare settings. The findings suggest that this implementation strategy is more than a simple extension of training. Coaching was viewed as vital for supporting practitioners full adoption of the intervention in their day-to-day practice, fitting an EBI to the complex needs of child welfare families, and ensuring high-quality implementation. Also identified were some unique aspects of PMTO coaching, such as a strict strengths-orientation and observation-based feedback via mandatory video recordings of client sessions. Further research is needed to explore different coaching techniques, protocols, and formats to examine whether certain features promote a more effective path to implementation and, ultimately, client outcomes.     

Psychotherapeutic care for sexually-victimized children – Do service providers meet the need? Multilevel analysis

ObjectivesSurprisingly little is known on the decision to refer sexually-victimized children to psychotherapy. Previous research on service provisions for victims of child maltreatment has analyzed the impact of case characteristics, like child or caregiver functional levels, lack of social support, and socioeconomic status. Findings, however, show that the decision to provide services is not only needs-driven, but also affected by external factors like provincial legislation, institutional policies, and the availability and accessibility of services. By analyzing characteristics behind the decision to refer sexually-victimized children to psychotherapy at the case and institutional level, we aimed to disentangle the complex interplay of factors driving this decision.MethodsThe data for this analysis were drawn from the first nationally-representative agency survey on reported child sexual victimization (CSV) in Switzerland. Over a 6-month data-collection period, 165 child protective services, 87 penal authorities and 98 agencies in the health and social sector documented a total of 911 incidents of CSV. Multilevel logistic regression was applied to analyze factors at both the case and contextual level.ResultsThe main finding was that the severity of consequences was strongly associated with the probability of psychotherapeutic service referrals (OR = 10.4; p < 0.001). However, one bias was identified at the individual level: sexually-victimized children born in Switzerland were more likely to be referred to psychotherapy than immigrant children. Institutional disparities in the decision to refer a sexually-victimized child to psychotherapy were large (median OR = 3.83), with penal authorities referring significantly fewer cases to psychotherapy than specialized agencies in the health and social sector. What exactly was driving the difference between psychotherapy referral in different types of agency remains largely unexplained.ConclusionsFuture research should invest in scrutinizing contextual factors of child protective service decisions. As we operationalize the need for psychotherapy as proxy-rated consequences of victimization, routine screening for mental health needs using standardized measures for children in contact with child protection agencies should be implemented, to help frontline workers to identify the psychotherapeutic needs of victimized children.     

Unmet legal and social advocacy needs of children with sickle cell disease: Implications for health care payer costs

Previous studies of advocacy needs faced by children and families have not differentiated needs that require attorney involvement (“legal advocacy needs”) from needs best addressed by social workers or lay advocates (“social advocacy needs”). Studies have also not examined the relationship between either type of need and health care costs. We developed a novel, replicable process to differentiate between legal advocacy needs and social advocacy needs. We then collected cross-sectional data from a sample of 52 children with sickle cell disease who were at least 1 year of age, a population with high advocacy needs and high health care costs. Mean annual health care costs to payers for children whose families had a least one legal advocacy need were $16,314, compared to $5552 for children in families with no legal advocacy needs (P = 0.007). After adjusting for covariates, the presence of a legal advocacy need was associated with $12,040 more in health care costs to payers (P = 0.02). Whether interventions to prevent and resolve legal advocacy needs can reduce health care costs by addressing the social determinants of health warrants future study.     

Factors that influence parent recruitment into prevention programs in early childhood: A concept map of parents', practitioners', and administrators' points of view

Given that parents are the main carriers of change in their child, their participation in programs for preventing behavior problems in early childhood represents a key element in ensuring the success of the intervention. That said, although at-risk families benefit the most from this type of program, they are identified as being hard to reach. This study aims at identifying the factors that influence parent recruitment based on the points of view of the actors concerned. To this end, a concept mapping operation was performed, which involved parents (n = 19), practitioners (n = 19), and administrators (n = 13) from community and public organizations. The participants generated 131 statements to complete the following sentence: In your opinion, parents would enroll more in prevention activities for their child if…. These 131 statements were classified into 12 groups representing the factors that influence parent recruitment, all of which were considered relatively important (between 3.51 and 4.42 on 5). The participants also agree on an understanding of recruitment that tends toward an ecological model. This study shows that parent recruitment into prevention programs targeting behavior problems in early childhood is complex and multifactorial (including factors that concern parents, practitioners, services, organizations, and policies). Because the map shows that the influencing factors in recruitment appear to be interrelated, they should all be considered to increase the chances of reaching at-risk families.     

Educational risk,recidivism, and service access among youth involved in both the child welfare and juvenile justice systems

Dually-involved youth represent a population of youth who receive some level of supervision from both the child welfare and juvenile justice systems concurrently. The current study examined education-related risk factors, recidivism, referrals for services, and service access among dually-involved youth in Los Angeles County. Specifically, whether increased educational risk was associated with referrals to, and access of, educational services and supports and whether higher receipt of educational services reduced recidivism approximately six months post-disposition. Data for this study consisted of a sample of dually-involved youth (N = 131) who were adjudicated delinquent and also had a child welfare case open. An Educational Risk Index (ERI) was developed and included school attendance, credit deficiency, problem school behavior, and current grades. Results indicate that educational risk was negatively associated with mental health services accessed, demonstrating that those with higher educational risk accessed less mental health services. Educational risk, however, was not associated with increased educational service referrals or access, suggesting a possible mismatch in educational need and service referrals. Lastly, there were no significant differences between those that recidivated and those that did not recidivate in service access and educational risk.     

The impact of para-professional social workers and community health care workers in Côte d'Ivoire: Contributions to the protection and social support of vulnerable children in a resource poor country

BackgroundResearch indicates a disproportionate impact of HIV and AIDS in sub-Saharan African countries, leading to many vulnerable families and children. Many of these communities have limited resources to support these vulnerable families, especially orphans and vulnerable children (OVC).Study aims and objectiveThis study set out to investigate how para-professional social workers and community health workers (PSWCHW) impact the provision of services and the psychosocial wellbeing and protection of vulnerable children in the community.MethodsThis quasi-experimental research study used data from an independent Save the Children program evaluation study in Côte d'Ivoire. We compared the health and psychosocial wellbeing of identified vulnerable children supported by para-professionals (n = 334) and children not receiving para-professional support (n = 213).FindingsSupport services and activities provided by PSWCHW included encouraging the children to be part of psychosocial support groups. Many of the children reported legal issues that ranged from getting a birth certificate issued to fighting or quarrelling with adults, disputes, public insults, beatings, and refusing to go to school. We found that the engagement of PSWCHW helped three out of four children go to school (compared to only one in four of the children without PSWCHW). PSWCHW also helped the children improve access to health care services.ConclusionsCommunities in sub-Saharan Africa should continue to consider the integrated utilization of para-professional social workers and community health care workers to support and improve psychosocial wellbeing of orphaned and vulnerable children which, in turn, enhances child protection services and access to healthcare.     

Differential response in child protection: How much is too much?

Recurring maltreatment can have devastating, lifelong consequences for children who are victims of abuse and neglect. The primary mission of child protection services (CPS) agencies is to prevent the recurrence of child maltreatment by offering families services designed to prevent future maltreatment. In order for these services to be effective, however, families must actually engage in them. This study focuses on how differential response programs impact families' successful engagement in services as indicated by long-term child safety outcomes.Differential response refers to a dual track system that allows public CPS agencies to respond to accepted reports of child abuse or neglect with either a traditional investigative response (TR) or an alternative response (AR). The AR is designed to be a less authoritative and less adversarial approach to families presenting a low- to moderate-risk of maltreatment. The AR involves collaboration with the family in a comprehensive assessment of the family's needs, risks and strengths and a referral to services on a voluntary basis. States implementing DR systems statewide vary greatly in the percentage of accepted reports that are assigned to the AR track. The goal of this study is to determine what rate of AR utilization is most effective in reducing subsequent reports of abuse and neglect for cases assigned to the AR.This study compares child safety outcomes for children assigned to the AR and those assigned to the TR within each of the fourteen states which implemented DR statewide between 2000–2012. The analysis uses Cox proportional hazards regression. Data for this observational study are drawn from the 2000–2012 data in the National Child Abuse and Neglect Data System (NCANDS) child files. Child safety is measured by re-reporting of the child to CPS.The theoretical model underpinning the study is McCurdy and Daro's (2001) Integrated Theory of Participant Involvement (ITPI) which outlines a conceptual model of parental engagement in services. This study hypothesizes that AR utilization rates affect the risk level of cases assigned to the AR. The ITPI model offers a possible explanation as to why higher risk cases may not be as responsive to the AR approach which depends upon the family's voluntary participation in services.The key finding of this study is that, in states assigning > 33% of reports to the AR, children on the AR track are being re-reported at equal or higher rates than those on the TR track. Only in those states utilizing the AR track for < 33% of the cases are AR cases being consistently re-reported at lower rates than those on the TR track. This article explores ways to improve the accuracy of track assignment decisions to prevent the assignment of higher risk families to the alternative response.     

Family group conferencing in Dutch child welfare: Which families are most likely to organize a family group conference?

Aim of the present study was to identify which families involved in child welfare are willing to organize a Family Group conference (FGc; phase 1) and which are most likely to complete a conference (phase 2). Data were used of a Dutch randomized controlled trial (N = 229). First, the proportion of families willing to organize an FGc and actually completing a conference was determined. Then, for each of the phases, reasons for dropout according to parents, child welfare workers and FGC-coordinators were assessed and categorized and family characteristics were linked to completion rate. Results showed that 60% of the families (137 families) were willing to organize an FGc and 27% (62 families) eventually completed a conference. Reasons for dropout were lack of motivation, high-conflict divorce situations and need for other professional care. Broken and/or newly formed families were less likely to complete a conference, whereas families with indications for child maltreatment were more likely to complete a conference. Future research is needed to examine other possible explanations for the relatively low success rate, such as attitude of child welfare workers towards FGC and the lack of understanding of the aim of FGC by child welfare workers and families.     

Ethnic disproportionality in the child welfare system: A Norwegian national cohort study

This study of Norwegian child welfare clients examined the extent to which ethnic disparities in involvement with the child welfare system can be attributed to ethnic differences in sociodemographic background. Using logistic regression models and a unique dataset constructed by linking child welfare records to national administrative registers for the 1993–1994 birth cohorts, we computed ethnic disparities in the odds of child welfare involvement at age 6–12 (N = 122,894), both before and after adjustments for sociodemographic background. Compared with ethnic Norwegian cohort peers, non-Western children had twice the unadjusted odds of entering the child welfare system (odds ratio = 2.13). However, the data also indicated pronounced ethnic disparities in eight sociodemographic correlates of child welfare involvement. When adjustments for these background factors were modeled, we found no ethnic differences in the odds of child welfare involvement. Our findings suggest that the association between ethnicity and child welfare involvement is confounded by differences in socioeconomic status, maternal social assistance, family size and family structure.