Sage Journals: Discover world-class research

Abstract

Foster children are at risk for developing disorganized attachment relationships (Van den Dries et al., 2009; Vasileva & Petermann, 2018) and behavioural problems (Goemans et al., 2015; Vanschoonlandt et al., 2012). Video-feedback Intervention to promote Positive Parenting and Sensitive Discipline – Foster Care/Adoption (VIPP-FC/A) is an intervention aimed at improving sensitive parenting in order to promote secure attachment relations and reduce behavioural problems. In this study, a Randomized Controlled Trial (RCT) was used in order to examine the effectiveness of VIPP-FC/A in a Flemish foster care sample (n = 100), using questionnaires at three measurement moments. The hypothesis was that VIPP-FC/A would increase foster mothers’ sensitive parenting and decrease the externalizing behavioural problems and insecure attachment behaviour of the foster children and parenting stress of the foster mothers. The intervention effects on these four outcome variables were analysed by constructing four multilevel models. After controlling for the baseline measurement, there was no significant main effect of condition for sensitive parenting, externalizing problems, insecure attachment behaviour and parenting stress. There was however a significant effect of condition and age on insecure attachment behaviour: within the group of youngest foster children (under 30 months), children in the intervention group scored lower insecure attachment scores than children in the control group. Additionally, there was a significant effect of condition and type of foster care on externalizing behaviour problems. In the group of children placed in kinship foster care, children in the intervention group had lower external behavior scores than children in the control group.

Keywords

Foster care VIPP-SD video-feedback intervention VIPP-FC

Introduction

Foster care for young flemish foster children

Out-of-home care is indicated when the safety, development and/or well-being of children are endangered in a profound way, or when parents cannot take the responsibility to raise their children (Pecora et al., 2009; Van den Bruel & Stroobants, 2016). Family foster care is considered the first option of choice when an out-of-home placement is needed in Flanders, the Dutch speaking part of Belgium (Van den Bruel, 2012). This need for family foster care is even more stressed for the youngest group (0–6 year) of children in need of an out-of-home placement (Van den Bruel & Stroobants, 2016), as a family like home environment supports the child to develop optimally (De Baat & De Lange, 2013). Young children make up a significant part of the new foster placements in Flanders, as 36% of the foster children were under six years old when placed in foster care (Pleegzorg Vlaanderen, 2020).

Flanders has five foster care agencies, each for one province (Van den Bruel, 2012), responsible for the recruitment and screening of foster carers, matching and monitoring the placement. There are different types of foster care: supporting, short-term, long-term and treatment foster care. Supporting foster care aims at offering a short-term break or a part-time stay (e.g. weekends or holidays) in order to support the family, or offers short-term emergency care. Short-term foster care has a maximum duration of 1.5 year, during which intensive efforts are made to improve the family’s situation. If reunification is impossible, a long term out-of -home placement needs to be considered. Long-term foster care creates permanency and continuity for the foster child, although a return home can be considered at any given moment. As adoption after a foster care placement is extremely rare in Flanders, children can be placed in foster care until adulthood. During treatment foster care, the foster child, foster parents and/or parents are given additional treatment and support (Van den Bruel, 2012).

Besides these different types of foster care, kinship and non-kinship care can also be distinguished. Kinship care are placements in the broader family or social network, 72% of all Flemish placements are currently kinship families (Pleegzorg Vlaanderen, 2020).

The need for interventions in foster care

According to the most recent numbers, 9,914 children are living in Flemish foster families (Agentschap Opgroeien, 2022). These children often experienced adversities, putting them at risk for problematic outcomes (Juffer, 2010). Research has shown similar rates of secure attachment relationships between foster children and their foster carers when these were compared to results of research in the regular population, but foster children were found to be more likely to have disorganized attachment relationships (van den Dries et al., 2009; Vasileva & Petermann, 2018). Insecure disorganized attachment is considered the least desirable attachment style, as it is associated with difficulties in dealing with stress, externalizing behavioral problems and psychopathology in later life (Van Ijzendoorn et al., 1999). A secure attachment relationship between children and their attachment figures has a positive effect on the overall development (Van IJzendoorn, 2008) and lowers the chances of foster placement breakdown (Strijker et al., 2008).

Additionally, foster children have more behavioural problems than children of the general population (Goemans et al., 2015; Vanschoonlandt, Vanderfaeillie, van Holen, et al., 2012). Longitudinal Flemish research found problem behaviour to have a direct negative impact on parenting behaviour, leading to less supportive parenting and more negative control (Vanderfaeillie et al., 2012). Behavior problems are associated with breakdown (Leathers, 2006).

Foster caregivers experience more parenting stress than parents of the general population (Bastiaensen, 2001). Neece et al. (2012) found evidence for a bidirectional relationship between parenting stress and child behavior problems in the general population. In the foster care population, research also shows an association between behavior problems and parenting stress (Goemans et al., 2018; Perry, 2015). Parenting stress has a negative impact on foster caregivers' parenting behaviors (i.e., decreased support and increased negative control). Consequently, parenting stress is therefore identified as a threat to the foster care placement (Bastiaensen, 2001; van Oijen, 2010; Vanderfaeillie et al., 2012; Vanderfaeillie & Van Holen, 2010; Vanschoonlandt et al., 2013; Wilson et al., 2004).

Foster children are at risk for developing disorganized attachment (van den Dries et al., 2009) and behaviour problems (Vanschoonlandt, Vanderfaeillie, van Holen, et al., 2012), their foster parents are at risk for higher levels of parenting stress (Bastiaensen, 2001). Sensitive parenting and sensitive disciplining are negatively associated with insecure attachment relations and behavior problems (Vanderfaeillie et al., 2013; West et al., 2020), so interventions are needed to promote sensitive parenting and sensitive disciplining. Early interventions are the best and most inexpensive interventions (Allen, 2011).

In response to the aforementioned challenges, various parenting interventions employing video feedback have been developed with the aim of enhancing parental sensitivity and promoting secure attachment in foster care settings. The Foster Carer-Foster Child Intervention (FFI; van Andel, et al., 2012) is a Dutch program that combines psychoeducation and video feedback. It focuses on reducing stress in foster children by bolstering the emotional availability, parenting skills, and confidence of caregivers. An RCT assessment notably positive effects on parenting skills and the responsiveness of foster children. However, there were no significant differences in stress levels for either caregivers or children (van Andel, et al., 2012).

Among the most widely recognized video-feedback interventions for foster caregivers is Attachment and Biobehavioral Catch-up (Dozier, 2013). Research has demonstrated its effectiveness in enhancing foster caregiver sensitivity (Bick & Dozier, 2013), reducing parental stress (Sprang, 2009), reducing children’s behavioral issues (Dozier, et al., 2006; Sprang, 2009), improving their executive functioning capabilities (Lind, et al., 2017), and reducing diurnal cortisol slopes (Dozier, et al., 2006) as well as avoidant attachment behavior (Dozier, et al., 2009).

Promoting First Relationships (Kelly et al., 2008), is an infant mental health training program that incorporates reflective video feedback. The objective is to improve caregiver sensitivity and understanding of children’s difficult behaviour to enable them to become more secure and behaviourally regulated. In a sample involving foster care, it has been demonstrated to be effective in improving caregivers' sensitivity and understanding of toddlers’ social-emotional needs, ultimately resulting in increased child competence. Nevertheless, there were no enduring post-intervention effects on attachment outcomes (Spieker, et al., 2012).

Video-feedback intervention to promote positive parenting for foster children

Video-feedback Intervention to promote Positive Parenting (VIPP) is an intervention based on Bowlby’s (1969) attachment theory and aims at increasing the sensitivity of caregivers and promote positive child–carer interactions (Juffer et al., 2017a). During the sessions, parent–child interactions are recorded, viewed and discussed with the parents (Juffer et al., 2017b).

VIPP is used for families with infants up to the first birthday. For older children (between one and six), Video-feedback Intervention to promote Positive Parenting and Sensitive Discipline (VIPP-SD) was developed. Additions were added to VIPP-SD to work on the discipline strategies of parents in dealing with difficult child behaviour (Juffer et al., 2017a).

A variant of VIPP-SD for foster children was developed, VIPP-FC/A, by adding two elements to make VIPP-SD more relevant to foster care and adoption situation (FC/A) (Schoemaker et al., 2020). The first was a focus on sensitive bodily contact, in order to promote a secure attachment (Duhn, 2010) and support the stress regulation of both foster parents and foster children (Schoemaker et al., 2018). The second element concerned foster care-specific behaviour of foster children, as foster children often learned to hide their feelings and needs. Their avoidance of proximity and pushing foster parents away can make it more difficult for carers to build a relationship with them (Stovall-McClough & Dozier, 2004). In VIPP-FC/A, foster parents’ attention is drawn to the subtle signals that their foster child gives and ways of responding appropriately to these signals.

The effectiveness of VIPP-FC/A was examined in three studies. The first two examined a sample of Italian mothers and their internationally adopted children, there is a partial database overlap between them. In Barone et al. (2018) (n = 80), VIPP-FC/A was associated with increased emotional availability of the children to their mothers (r = .46) and higher levels of maternal positive parenting (r = .39). In the second study, Barone et al. (2019) (n = 83) found increased sensitivity among mothers in a six-month follow-up of the intervention group (b = 83, p = .001), which contributed to higher levels of emotional availability and decreased externalizing behavioural problems among the children (b = .55, p = .001, b = 2.64, p = .001). The third study involved Dutch foster mothers and their foster children (n = 60). Schoemaker, Juffer, et al. (2020) found no evidence that VIPP-FC/A was effective in improving foster carers’ sensitive parenting or positive attitudes, the researchers suggested that the findings could be explained by a selection bias and subsequent ceiling effect.

The foster parents in their sample generally had high scores of parental sensitivity, sensitive discipline, and parenting attitudes at the pretest, leaving limited room for enhancement. Furthermore, the authors state that there might also be a selection bias at play: the foster families that could have taken advantage the most benefit from a parenting intervention might not have been part of the study sample.

Research question

VIPP-SD has proven to be effective at improving different outcome variables. Juffer, Bakermans-Kranenberg, and van IJzendoorn (2018) conducted a meta-analysis on VIPP(SD), examining 12 randomized controlled trials (RCTs). The overall effect size in the meta-analysis for sensitivity was d = .47, and for attachment, it was d = .36. The combined effect size for improved child outcomes was d = .37, and for reduced child problem behavior, d = .26. Van IJzendoorn et al. (2023) conducted a series of meta-analyses of 25 randomized controlled trials testing the VIPP-SD method. They found combined effect sizes on parenting behavior (d = .37) and attitudes (d = .32), and on child attachment security (d = .47), but not on child externalizing behaviour problems (d = .14).

However, the results on the variant for foster care and adoptive situations are less clear and there-are no studies regarding VIPP-SD, nor VIPP-FC/A in Flanders. The hypothesis was that VIPP-FC/A increases foster mothers’ sensitive parenting and decreases the externalizing behavioural problems and insecure attachment behaviour of foster children and parenting stress of the foster mothers.

Methods

Participants

In Table 1, the descriptive statistics of the control and intervention group are presented. There were no differences between both groups on demographic characteristics.

Table 1.

Sample characteristics.

Variable		Control group (n = 49) n (%)/M (SD)	Intervention group (n = 51) n (%)/M (SD)
Age in months		36.10 (15.21) (range 17–74)	36.65 (13.93) (range 17–72)
Age in months at start current placement		20.54 (17.29) (range 0–64)	19.43 (16.17) (range 0–64)
Gender	Boys	23 (46.9)	24 (47.1)
Gender	Girls	26 (53.1)	27 (52.9)
Reason of inclusion	New placement	41 (83.7)	40 (78.4)
Reason of inclusion	Curative	8 (16.3)	11 (21.6)
Placement type	Kinship	17 (34.7)	18 (35.3)
Placement type	Non-kinship	32 (65.3)	33 (64.7)
Number out-of-home placements		1.04 (1.15) (range 0–6)	1.08 (1.25) (range 0–5)
Number additional children in foster home		1.38 (1.36) (range 0–4)	1.23 (1.00) (range 0–3)
Age foster mother		43.83 (8.09) (range 27–61)	41.60 (58.10) (range 30–61)
Higher education foster mother	Yes	21	17
Higher education foster mother	No	25	31
EA-SR baseline		122.76 (14.04) range (70–143)	116.82 (12.87) range (80–142)
EA-SR postintervention		124.10 (14.92) range (70–143)	118.55 (12.43) range (77–145)
EA-SR follow-up		123.88 (15.12) range (70–144)	120.00 (13.18) range (80–147)
AISI baseline		45.67 (12.82) range (21–87)	51.43 (13.71) range (29–87)
AISI postintervention		44.90 (12.73) range (23–87)	50.22 (14.20) range (26–87)
AISI follow-up		45.12 (12.65) range (26–87)	50.78 (13.98) range (32–87)
CBCL ext. problems baseline		14.45 (9.29) range (2–37)	18.65 (10.80) range (3–41)
CBCL ext. problems postintervention		15.33 (9.78) range (0–36)	17.97 (10.86) range (2–44)
CBCL ext. problems follow-up		15.08 (10.50) range (1–40)	18.41 (10.65) range (2–43)
PSQ baseline		51.49 (10.59) range (39–86)	54.88 (10.39) range (37–83)
PSQ postintervention		51.55 (9.67) range (37–74)	57.10 (11.27) range (39–82)
PSQ follow-up		52.14 (10.67) range (38–80)	56.25 (11.02) range (39–90)

Procedure

All five Flemish foster care agencies agreed to participate to this study. They provided information of the study to the eligible families and returned the informed consents of the participating foster mothers.

Inclusion criteria were:

- A foster child between 17 months and six years old at time of inclusion

- The foster child was placed in the foster family in the past year (new placements)

- It was also possible to be included in the study when the children were already residing in the family for more than one year, when the families or the foster care workers assesses the risk for attachment, parenting or behaviour problems as high (curative)

- Placements had to be long-term, children in supporting and short-term foster care were excluded

- Foster children had no known cognitive disability, severe motor disability or ASS

- Foster mothers had to be able to read Dutch (criterium: ‘able to read and understand a Dutch paper’)

When more than one foster child from the same foster family was eligible, the youngest child was included. Twins were excluded from the study.

After collecting an informed consent, foster care workers filled in a questionnaire regarding characteristics related to the foster child, foster care placement, case history, parents and foster parents (Vanschoonlandt et al., 2012). Foster mothers filled out questionnaires at baseline (T0), postintervention (5 months later; T1) and at follow-up (2 months after T1; T2). All families allocated to the control group were offered the intervention after completing the follow-up questionnaires. The first author sent out the questionnaires by email, families that struggled to answer digitally received the questionnaires by mail. All procedures were approved by the boards of the foster care agencies and by the ethical board of [blinded for review].

The study was designed as a RCT. The control group received treatment as usual. They were offered the care that participants in the control group would typically receive by the foster care workers, including interventions, if they were not enrolled in the trial. To achieve equal distribution between the intervention and control group, randomization by minimization was used. We wanted to ensure that both intervention and control group were comparable on different meaningful characteristics, especially because our sample size was rather small. This approach aims to obtain two comparable groups in terms of gender and age of the foster child (<2.5 years/>2.5 years), reason for inclusion in the study (new placement/curative), type of foster placement (kinship/nonkinship) and previous out-of-home placements (none/at least one), ensuring balanced participant numbers and characteristics across both groups (Taves, 2010). When a child was eligible for this study, the randomization was carried out by the first author. Allocations to the intervention or control groups were done using MinimPy Program .3 (Saghaei & Saghaei, 2011). MinimPy Program .3 is an open-source minimization program that was written in Python programming language. Initially, the first participants are randomly assigned to one of the groups. For subsequent participants, the allocation occurs following a hypothetical assignment to each group, followed by the calculation of an imbalance score. Based on these imbalance scores, the program decides to which group the new participant should be assigned to, with the goal of minimizing the disparity in prognostic factors (Saghaei & Saghaei, 2011).

We included various child and foster care placement characteristics in the randomization with minimization and subsequent analyses because, based on literature, we expected that these characteristics could potentially have an effect on the level externalizing behaviour problems, the quality attachment and/or the effectivity of the intervention. The first characteristic is the gender of the foster children, since in the general population boys exhibit more externalizing behavior problems than girls (Bongers et al., 2003; Inchley et al., 2020). Additionally, some researchers found girls more likely to be securely attached to their foster parents than boys shortly after the placement started (Gabler et al., 2014; Lang et al., 2016). The second factor was the age of the foster children, because in the general population externalizing behaviour problems decrease with age (Bongers et al., 2003).The results according age and externalizing behaviour problems are inconsistent in foster care research (West et al., 2023). Regarding attachment, some research points to younger children having more often a secure attachment shortly after start of the placement (Lang et al., 2016; Ponciano, 2010). Reason for inclusion in the RCT was decided because children who are enrolled in the study because of curative reasons will on average experience more difficulties than regular starting foster care placements. The fourth factor is type of foster care. Kinship foster carers are more often vulnerable because they are older, have more health issues, have a lower educational level, lower income, receive less support and experience more parenting stress (Fuentes-Peláez et al., 2014). In international research, children in kinship foster care have fewer behavioural problems (Holtan et al., 2005; Keller et al., 2001), these differences are not found in Flemish and Dutch research or wereby other variables such as the number of previous placements (Bastiaensen, 2001; Strijker et al., 2003; Strijker & Van Oijen, 2006; Vanschoonlandt, Vanderfaeillie, Van Holen, et al., 2012). The last factor was previous out-of-home placements, because replacements have negative consequences for foster children (Pecora et al., 2009; Van den Bergh & Weterings, 2010; Van Holen et al., 2012). For example, it was found that relocation can increase the behavioural problems of foster children (Herrenkohl et al., 2003; Newton et al., 2000). The maternal educational level was the only factor that was used in the analyses, but was not included in the randomization with minimization. Some researchers showed that parenting interventions can be especially advantageous for mothers with limited education (Mendelsohn et al., 2007). Klein Velderman et al. (2006) point out that parenting information is less available to mothers with a low education, making them interesting candidates for the VIPP-SD intervention

Instruments

Emotional availability–Self report

The Dutch version of the Emotional Availability–Self Report (EA-SR) was used to measure the self-reported parental sensitivity of the foster mothers (Vliegen et al., 2005). The EA-SR is made out of 32 items scored on a 5-point scale (1 = completely disagree, 5 = completely agree). The results of the scores on these 32 items can be grouped into five subscales: affective attunement (mutual factor), parental involvement (child factor), affective quality (mutual factor), intrusiveness (parental factor), and hostility (parental factor), and a total score (Vliegen et al., 2005). The Cronbach alpha coefficient of the total scale was α = .85 in this study.

Attachment insecurity screening inventory 2–5 years

The Attachment Insecurity Screening Inventory 2–5 years (AISI 2–5 years, (Polderman & Kellaert-Knoll, 2008; Wissink et al., 2016) is a questionnaire measuring insecure attachment between caregiver and the child. It contains 20 questions scored on a 6-point scale ranging from never (1) to always (6), in this study scored by the foster mothers. Items can be grouped in three subscales, but only the AISI total insecurity scale was used in this study. Higher scores indicate more attachment insecurity, total scores of 50 and higher are considered clinical. Convergent validity was supported by negative associations with observed attachment security measured with the Attachment Q-Sort (AQS, Waters & Deane, 1985) and total scores on the AISI. The AISI total score discriminates between children with a secure and insecure attachment with their caregivers measured with the AQS, with an AUC-value for mothers of .91 (Wissink et al., 2016). The Cronbach alpha coefficient for the AISI total insecurity scale in current study was α = .72.

We only used the total insecurity scale, as only that scale was used to examine the convergent validity (Wissink et al., 2016). The decision to use the AISI instead of the Strange Situation Procedure (SSP) was based on the reasons for using questionnaires mentioned before. Additionally, the RCT involved repeated measures, with only a two-month gap between the second and third measurements. Using the SSP with such a short time gap could sensitize the children to the procedure.

Parenting Stress Questionnaire

The Dutch version of the Parenting Stress Questionnaire (PSQ, Vermulst et al., 2012) was used to measure parenting stress of foster mothers. The PSQ contains 34 questions scored on a 4-point scale, ranging from (1) not true to (4) very true. Five subscales can be calculated, but only the total scores were used in this study. Higher scores on the PSQ represent higher perceived levels of parenting stress. T-scores under 60 are considered non-clinical, scores between 60 and 63 subclinical and scores of 64 and higher clinical. In this study, the Cronbach alpha coefficient of the total parenting stress scale (over 34 questions) was α = .92. The validity of the PSQ is good (Vermulst et al., 2012).

Child behavior checklist 1,5–5

Behaviour problems of the foster children were quantified using the Dutch translation of the Child Behaviour Checklist (CBCL; (Achenbach & Rescorla, 2000; Verhulst & van der Ende, 2000). The CBCL contains 99 statements regarding behaviour of the child scored by the foster parent on a 3-point scale. The questionnaire provides scores for three broad-band scales: internalizing, externalizing and total problems (Achenbach & Rescorla, 2000). Subclinical scores are scores between 60 and 63, clinical scores are 64 and above. In this study, the internalizing subscale has a Cronbach’s Alpha of α = .93, the Cronbach’s Alpha for the externalizing subscale is α = .87 and for the total score it is α = .96. The validity of the CBCL is considered good (Achenbach & Rescorla, 2000). Only the externalizing problems subscale was used in this study.

VIPP-FC/A intervention

VIPP-FC/A is implemented with one foster (or adoptive) parent and a child in the family’s own home. At the start of each of the seven VIPP-FC/A sessions, short video-recordings are made in which care taker and child are asked to interact as usual during daily situations. These recordings are then viewed by the foster parents and a VIPP-FC/A practitioner, giving micro-feedback to the foster carer. In the first two sessions, the feedback is formulated from the foster child’s perspective. In sessions three and four, focus is on parental sensitivity. The last two sessions are used to review what has been learned. General information about child development is given in all the sessions.

VIPP-FC/A practitioners are foster care workers that had a four days training and did a supervised test VIPP-FC/A intervention. In this study, 16 trained foster care workers delivered the VIPP-FC/A interventions. Each had a degree (bachelor’s or master’s) in social work, psychology or pedagogy. The families receives the whole intervention including seven sessions.

Data analysis

A priori power analysis showed that a sample of 98 foster families resulted in an adequate statistical power (.81; Repeated measures ANOVA within-between interaction, G*Power 3.1.9.7) (Faul et al., 2009), based on an effect size f of .13, two groups, a repeated measures design with three measurement points, and an α = .05. The effect size f = .13 was based on the meta-analytic results of Juffer et al. (2018), who found the smallest effect size of VIPP-SD was the combined effect size for reduced behaviour problems (d = .26).

Data were analysed using Intention-to-treat analysis, which involves analyzing participants based on their initial group assignments. By employing this analysis method, the prognostic balance achieved through randomization is maintained (McCoy, 2017). Figure 1 gives an overview of the occurrence of missing data. The pattern of missingness was explored using a missing data dummy code (missing value = 0, non-missing value = 1) (Cohen & Cohen, 1983). This dummy code was used in t-tests and χ²-tests with the other variables. Results show that gender of the foster childer was significantly associated with the dummy code. That is data of boys was more often missing (χ² (1) = 8.47, p = .004). In addition to the efforts done to minimize missing data, ‘Last Observation Carried Forward’ method (LOCF) was used. LOCF is a common statistical approach for analysing longitudinal repeated measures data with missing observations (Whittaker et al., 2006). In a LOCF-analysis, a missing value is replaced by that subject’s previously observed value, meaning the last observation is carried forward. The combined dataset, consisting of observed and imputed data, is then analysed as if there were no missing data present. There has been some criticism on the use of LOCF and in literature it is stated that it should be used with caution (Whittaker et al., 2006). Using LOCF is however a more conservative method than running a multiple imputation algorithm to deal with these missing data (O’Neill & Temple, 2012). It was preferred over running a multiple imputation algorithm in this study because of this more conservative approach, we were cautious not to overestimate a possible treatment effect. LOCF was applied to 16 families at T1 and 29 families at T2. 100 families were included in the analyses after LOCF.

Figure 1.

Overview of participants.

Demographic background variables and baseline outcomes of the intervention and control group were compared using t-tests and χ²-tests.

The intervention effects of VIPP-FC/A on sensitive parenting (EA-SR), insecure attachment behavior (AISI), behavior problems (CBCL) and parenting stress (PSQ) were analyzed by constructing 4 multilevel models, one for each outcome variable. Because of the baseline differences for EA-SR, AISI and CBCL between intervention- and control group, all models were constructed as longitudinal covariance analysis models (Lydersen, 2022; Twisk, 2021). The baseline measurements were added as independent variables instead of as parts of the outcome variables (Lydersen, 2022; Twisk, 2021), thus only the post-intervention (T1) and follow-up (T2) data were included as dependent variables.

Note that using multilevel modelling, a distinction is made between level 1 and level 2 predictor variables. In current study, the level 1 variables are Time (T1 = post-intervention, T2 = follow-up) and the baseline measurements of the outcome variable. The level 2 explanatory predictor variables are condition (intervention, control group), reason for inclusion (new foster care placement or foster parents/social workers requesting VIPP-FC/A for curative reasons), age (younger or older than 30 months), duration of current placement (continuous variable), placement type (kinship, non-kinship), gender of the foster children (male of female), previous placements of the foster children (yes, no) and higher education of the foster mother (yes: holding a college or university degree or no).

A stepwise, bottom-up model building approach was adopted (Hox et al., 2018). The first model was the intercept-only model (M₀), which gives a benchmark value of the deviance. In the second model (M₁), the level 1 explanatory variables were added (i.e., Time and the baseline values). In the third model (M₂), the level 2 explanatory variables and cross-level interactions were added. Fourthly, we assessed if the slopes of the level 1 variables have a significant variance component between groups in the random coefficient model (M₃). No evidence was found for the presence random slopes.

At each step, variables were added to the model on a variable-by-variable basis with an alpha-to-enter of .05. Variables that became unsignificant were removed from the model (alpha-to-remove = .10). Because of its theoretical importance, the main effects of Time and Condition and the two-way interaction of Time and Condition were always kept in the model. Model improvement at the end of each step was assessed using AIC values. For all models, the intraclass correlation (ICC) is computed.

All model assumptions were met. A p-value below .05 was considered significant. Analyses were performed in R version 4.3.0, using the lme4 (Bates et al., 2009), lmerTest (Kuznetsova et al., 2017) and car (Fox et al., 2023) packages. The figures were generated using sjPlot (Lüdecke et al., 2023).

Results

Baseline differences

Despite the design using minimization to randomly assign and balance participants’ characteristics across control and intervention groups, there were significant differences for three outcome variables at the baseline measurement. The scores of the EA-SR were higher in the control group (M = 122.76, SD = 14.04) than in the intervention group (M = 116.82, SD = 12.87), t (98) = 2.20, p = .030. For the externalizing problem scale of the CBCL, the scores were higher in the intervention group (M = 18.65, SD = 10.80) than in te control group (M = 14.45, SD = 9.29), t (98) = 2.08, p = .040. The scores on the AISI were higher in the intervention group (M = 51.43, SD = 13.71) than in the control group (M = 54.67, SD = 12.82), t (98) = −2.17, p = .033. This indicates more external problems and insecure attachment behavior in the intervention group than in the control group, and higher sensitive parenting in the control group. There were no significant differences between PSQ-scores of the control (M = 51.49, SD = 10.59) and intervention group (M = 54.88, SD = 10.39) at baseline, t (98) = −1.60, p = .113. Because of the baseline differences for EA-SR, AISI and CBCL between intervention- and control group, baseline imbalances were adjusted for by constructing the models as longitudinal covariance analysis models (Lydersen, 2022; Twisk, 2021). Although there were no significant differences for the baseline PSQ-scores, the model was also constructed as longitudinal covariance analysis model to promote consistency, and because the PSQ was scored by foster mothers of the same sample where there are clearly significant differences between both condition on several variables.

Intervention effects of VIPP-FC/A

Multilevel mixed models were constructed to assess the intervention effects for each dependent variable. More details on the model building process for each multilevel model can be found in Appendix 1.

Sensitive parenting behavior

The results of the multilevel model for EA-SR can be found in Table 2. There was no significant effect for time (

\hat{β}

(time) = .78, p = .437), nor was there for condition (

\hat{β}

(condition) = −.52, p = .734) or for the two-way interaction of time and condition (

\hat{β}

(time*condition) = −1.11, p = .441). The only significant predictor for sensitive parenting was its baseline measurement (

\hat{β}

(baseline) = .88, p < .001).

Table 2.

Final multilevel models for EA-SR, AISI, CBCL external problems and PSQ as dependent variables.

Model	EA-SR	AISI	CBCL	PSQ
Fixed part	B (SE)
Intercept	16.82 (6.16)**	4.22 (3.59)	3.56 (1.70)*	10.90 (3.27)**
Baseline measurement	.88 (.05)***	.81 (.06)***	.79 (.06)***	.84 (.06)***
Time [T2 = follow-up]	.78 (1.01)	.55 (1.40)	1.07 (.74)	−.21 (.76)
Condition [0 = control group]	−.52 (1.53)	2.39 (2.88)	−1.11 (1.61)	−1.43 (1.33)
Age [0 = older than 30 months]	—	6.9 (2.14)**	—	—
Educational level foster mother [0 = no higher education]	—	.38 (2.49)	—	—
Placement type [0 = kinship]	—	—	−1.08 (1.84)	—
Time*condition	−1.11 (1.44)	2.65 (2.10)	−.79 (.93)	.55 (1.98)
Condition*age	—	−7.73 (3.06)*	—	—
Time*educational level foster mother	—	.28 (2.35)	—	—
Condition* educational level fostermother	—	2.40 (3.51)	—	—
Time*placement type	—	—	−1.97 (.97)*	—
Condition*placement type	—	—	5.35 (2.56)*	—
TimeCondition educational level foster mother	—	−7.00 (3.30)*	—	—
Random part
$σ_{ϵ}^{2}$	25.75	30.37	10.80	14.54
$σ_{u 0}^{2}$	30.41	35.17	29.11	27.95
ICC	.54	.54	.73	.66
BIC	1367.67	1320.44	1265.11	1278.45
AIC	1344.58	1278.36	1232.13	1225.43

Note: * p < .05 ** p < .01 *** p < .001. $σ_{ϵ}^{2}$ = variance of individual level residuals, $σ_{u 0}^{2}$ = variance of group level residuals.Time (2 = T1, 3 = T2), condition (0 = control group, 1 = intervention group), age (0 = older than 30 months, 1 = younger than 30 months), educational level (0 = no higher education, 1 = higher education), placement type (0 = kinship, 1 = non-kinship), reason for inclusion (0 = new placement, 1 = curative), previous out of home placements (0 = no previous placements, 1 = previous placements).

Insecure attachment behaviour

Based on the information presented in Table 2, there was no significant effect of time ( $\hat{β}$ (time) = .55, p = .696) for insecure attachment. The two-way interaction of time and condition was not significant ( $\hat{β}$ (time*condition) = 2.65, p = .209). Baseline score was a significant predictor for later insecure attachment scores ( $\hat{β}$ (baseline) = .81, p < .001). There was a significant main effect of age, children younger than 30 months had higher scores in the control condition than in the intervention condition ( $\hat{β}$ (age) = 6.95, p = .002).

There was no significant effect of condition ( $\hat{β}$ (condition) = 2.39, p = .408) for insecure attachment. There was a significant two-way interaction effect of condition and age. To allow easy interpretation, Figure 2 was added. Within the oldest age category (older than 30 months), there was no significant difference between both conditions ( $\hat{β}$ (condition) = 2.39, p = .408). Within the youngest age category (younger than 30 months), there was a significant difference between control and intervention group ( $\hat{β}$ (condition*age) = −7.73, p = .013). Within this age category, children in the intervention group had on average lower scores than children in de control group.

Figure 2.

Predicted values of AISI for age and condition.

Additionally, Figure 3 shows a significant three-way interaction of condition, time and educational level ( $\hat{β}$ (condition*time*higher education) = −7.00, p = .036). The effect of higher education of foster mothers was negative in the intervention group at T2, resulting in lower insecure attachment scores at T2 in the intervention group when foster mothers had a college of university degree. There was no significant difference for education in the control condition at T2, or for both conditions at T1.

Figure 3.

Predicted levels of AISI for time, condition and educational level.

Externalizing behaviour problems

The baseline score was predictive for externalizing behavior scores ( $\hat{β}$ (baseline) = .79, p < .001). There was no significant effect for time ( $\hat{β}$ (time) = .1.07, p = .150). Because of the significance of the two-way interaction for time and type of foser care, Figure 4 was generated. The two-way interaction for time and type of foster care (Figure 4) shows that within the group of kinship families, there were no significant differences for externalizing behavior problems for time ( $\hat{β}$ (time) = .1.07, p = .150). Within the group of kinship families, there were higher scores at T2 than at T1 ( $\hat{β}$ (time*type of foster care) = −1.97, p = .046).

Figure 4.

Predicted values for CBCL with time and type of foster care.

There was no significant effect of condition ( $\hat{β}$ (condition) = −1.11, p = .493) for externalizing behavior problems. The two-way interaction of time and condition was not significant ( $\hat{β}$ (time*condition) = −.79, p = .396). There was a significant two-way interaction effect of condition and age (Figure 5). Within the scores of children placed in non-kinship foster care, there was no significant difference between both conditions ( $\hat{β}$ (condition) = −1.11, p = .493). Within the kinship group, there was a significant difference between control and intervention group ( $\hat{β}$ (condition*type of foster care) = 5.35, p = .039). In the group of children placed in kinship care, children in the intervention group had on average lower scores than children in de control group.

Figure 5.

Predicted values for CBCL with condition and type of foster care.

Parenting stress

As indicated in Table 2, there was no significant effect of time ( $\hat{β}$ (time) = −.21, p = .779) and condition ( $\hat{β}$ (condition) = −1.43, p = .283) for parenting stress. The two-way interaction of time and condition was not significant ( $\hat{β}$ (time*condition) = .55, p = .616). The baseline score was a significant predictor for parenting stress ( $\hat{β}$ (baseline) = .83, p < .001).

Discussion

In this study, a RCT of VIPP-FC/A was conducted. It was hypothesised that VIPP-FC/A would increase foster mothers’ sensitive parenting and a decrease of foster mother’s parenting stress. Moreover, it was expected to result in a decrease of externalizing problems behaviour and insecure attachment behaviour of the foster children. Contrary to the hypothesis, there were no main intervention effects for sensitive parenting, externalizing problems, insecure attachment behaviour and parenting stress. There was a significant effect of condition and age on insecure attachment behaviour: within the group of foster children under 30 months, children in the intervention group had lower insecure attachment scores. There was also a significant effect of condition and type of foster care on externalizing behaviour problems. In the group of children placed in kinship foster care, children in the intervention group had lower external behavior scores than children in the control group.

No significant intervention effects of VIPP-FC on sensitive parenting were found. This was a rather unexpected result since a meta-analysis of van IJzendoorn et al. (2023), combining the results of 25 RCT’s testing VIPP-SD, found a small combined effect size on parenting behavior (r = .18). Barone et al. (2019) also detected increased sensitivity among mothers of adopted children in the VIPP-FC/A intervention group (b = 83, p = .001). On the other hand, this result was in line with research conducted by Schoemaker, Juffer, et al. (2020), who found no evidence that VIPP-FC/A was effective in improving foster carers’ sensitive parenting. Foster parents go through a pre-placement selection process and are suppported in positive parenting by social workers throughout the placement. This may result in parenting behavior high in sensitive parenting, which may lead to a ceiling effect. Indeed, Schoemans found high scores on parental sensitivity and sensitive discipline in her sample, indicating little room for improvement (Schoemaker, Juffer, et al., 2020). In our study, foster parents had a mean score of 122.76 (SD = 14.04, range 70–143), at baseline in the control group and 116.82 (SD = 12.87, range 80–142) in the intervention group, while the maximum score for the EA-SR is 160. Another explanation could be the use of the EA-SR instrument. The EA-SR was developed for parents of infants, while the children included in our sample group were older. Moreover, it is a self-reported questionnaire, requiring high levels of introspection and self-knowledge (Vliegen et al., 2009), which could complicate the ability of foster parents to assess their own parental sensitivity.

There was also no main intervention effect for insecurity of attachment between foster children and the foster mothers. We did expect a intervention effect, as the most recent meta-analyses of van IJzendoorn et al. (2023) on VIPP-SD found a small combined effect size on attachment security (r = .23). Schoemaker, Juffer, et al. (2020), on the other hand, did not find any intervention effect of VIPP-FC on quality of attachment between foster children and their foster mothers.

There was however a significant effect of condition and age on insecure attachment behaviour. Within the group of foster children under 30 months, children in the intervention group had lower insecure attachment scores. There was no significant difference between both conditions for the group of foster children of 30 months old. A possible explanation is that the attachment behavior of older children may be less susceptible to change through the VIPP-FC/A intervention. This suggestion is not supported by Schoemaker, Wentholt, et al. (2020), who found no effectiveness of intervention programs for adoptive and foster parents in improving attachment security but also showed that the intervention effects were not moderated by age of the children. Another possibility is that the intervention is less suited for older foster children, as they are more aware of the presence of a VIPP-intervener and the filming, and might behave more social desirable. By not showing their typical behaviour, foster parents might have more trouble implementing the feedback of the VIPP-intervener in their day-to-day life. Contrary to our results, Van Ijzendoorn et al. (2023) in their meta-analysis even found that age of the child did moderate the effectiveness of VIPP-SD, showing stronger effect on attachment in studies with older children (F(14) = 7.48, p = .016).

There was no main intervention effect for externalising problem behaviour of the foster children. Although VIPP-FC aims at diminishing externalising behaviour problems, the results of research of VIPP on problem behaviour is not clear. In the meta-analysis on VIPP-SD (van IJzendoorn et al., 2023), there was no combined effect for externalizing behaviour problems. Schoemaker, Juffer, et al. (2020) were also not able to find an effect of VIPP-FC/A on behaviour problems in their foster care sample. In contrast, Barone et al. (2019) did find decreased externalizing behavioural problems among the children in a study of VIPP-FC/A in a sample of adopted children and their mothers (b = 2.64, p = .01). Van Ijzendoorn et al. (2023) suggested that the effects of the intervention might need more time, referring to Van Zeijl et al. (2006), who found a sleeper effect on externalizing problem behaviour.

There was a significant effect of condition and type of foster care on externalizing behaviour problems. In the group of children placed in kinship foster care, children in the intervention group had lower externalizing behavior scores than children in the control group. This could be due to the pre-existing bond with the child, making foster parents more emotionally investing in their foster child. In addition, non-kinship families receive pre-placement training (Shlonsky & Berrick, 2001), which is likely to be non-existent or limited for kinship families. Therefore, it would be possible that the intervention has a greater effect on the sensitive disciplining of kinship foster caregivers, resulting in a reduction in behavioral problems.

Parenting stress of the foster parents was included because foster parents are known to experience higher levels of parenting stress (Bastiaensen, 2001), which is associated with behaviour problems (Goemans et al., 2018) and difficulties in the attachment relationship between foster children and their foster carers (Gabler et al., 2014; West et al., 2022). There was no main intervention effect of VIPP-FC on parenting stress.

No main intervention effects on the various outcome variables were found. While VIPP-SD has been found to be effective in various at-risk populations, it may not be as effective in foster care situations. This might be due to the population of foster children. Foster children often experience complex social adversity that triggers various developmental mechanisms, which are not limited to the development of maladaptive attachment (Tarren-Sweeney, 2014). Flemish research showed that 85% of the foster children had to deal with at least one traumatic experience besides separation from their primary caregivers (Vanderfaeillie et al., 2014). It might be necessaire to add a specific more trauma-informed part to VIPP-FC/A, in order to fit this foster care sample more.

Another Flemish foster care specific aspect is the temporary nature of the placement. Adoption after a foster care placement is extremely rare in Flanders and although long-term foster care aims at creating permanency, a return home can be considered at any given moment (Van den Bruel, 2012). This can complicate the development of an attachment relationship between foster carers and foster children, cause anxiety for both foster children and foster parents and unconsciously might implicate the engagement of foster parents.

Limitations

Gathering information about families who declined to participate in the study proved to be challenging. Due to the lack of reliable data on non-response rates, the representativeness of the sample could not be assessed. All foster families who participated in the study did so voluntarily, which could introduce a positive selection bias, particularly since the intervention (VIPP-FC) involves filming families, which could be intimidating for those who are struggling with their foster children.

The use of questionnaires instead of observations can be considered a limitation. Nevertheless, questionnaires were used because the researchers had reservations about the feasibility of having foster parents bring the children to a laboratory setting, as it could be a time-consuming and burdensome task, potentially deterring their involvement in the study. The families lived across five provinces, which made asking them to come to a laboratory setting in one central location an important barrier. One of the strengths of VIPP-FC/A is precisely that it is delivered in the home setting. The use of questionnaires in this research did limit the diversity in responses and the kind of data collected. Additionally, all the information was obtained solely from foster mothers, which could potentially lead to concerns regarding common method variance (Podsakoff et al., 2003). Some of the questionnaires required self-reflection, which may have produced different results than direct observations. Furthermore, there is a risk of response bias in studies reliant on questionnaires (Edwards, 1990). As mentioned earlier, the children in our sample were older than the intended age range of the EA-SR and some children in the study were slightly younger than the intended age range for the AISI questionnaire (24 months), which could also affect the results obtained using these questionnaires.

The use of the AISI to assess attachment behavior could also be considered as a limitation. Foster children are more likely to have disorganized attachment relationships with their foster parent (van den Dries et al., 2009; Vasileva & Petermann, 2018). Since this is the least desirable attachment style, an intervention effect of VIPP-FC/A could be a huge benefit. We only used the total insecurity scale, as only that scale was used to examine the convergent validity (Wissink et al., 2016).

The significant dropout rate can be seen as a limitation, as can the fact that families were not blinded to assignment to control or intervention group. Despite all efforts to maximize chances of randomization, the outcome measures at the baseline measurement were not equal for both groups. The scores in the control group were higher for the EA-SR and lower for both the externalizing problem scale of the CBCL and on the AISI were higher in the intervention group, indication more external problems and insecure attachment behaviour.

Suggestions for further research

The findings regarding the effect of VIPP-FC/A for the youngest group of foster children are promising, so it is necessary to do more research in a larger group of young foster children (under 2.5 years). Using the SSP would also provide insight into possible effects of VIPP-FC/A on disorganized attachment behaviour.

The sample consisted of more non-kinship families than kinship families, whereas this is currently reversed in the foster care population. Research with a similar proportion of kinship families as in the population is recommended so that the effect seen on behavioural problems can be revisited. In further research it should be considered to measure sensitive disciplining, as we would expect more sensitive disciplining of foster parents to result in a decrease of behavior problems.

The theoretical basis of the intervention suggests that sensitivity in parenting of foster carers can influence the quality of attachment between foster children and their carers. However, the study did not find any effect of VIPP-FC/A on sensitivity. Nonetheless, an effect of the intervention on insecure attachment behaviour was observed in the youngest group of foster children. More research into the effects of VIPP-FC/A on sensitive parenting is therefore indicated, preferably by adding observations of foster mothers and their foster children.

Similar research also parental attitudes towards parenting and self-efficacy of parents into account. In futhur research, this concepts could be included as outcomes. It might be interesting to include internalizing behaviour problems as well.

The abovementioned additional research should clarify whether additional indications are needed, such as offering the intervention to young foster children or families in network foster care.

Conclusion

In this study the effectiveness of VIPP-FC/A in a Flemish foster care sample was assessed. The intervention effects on these four outcome variables were analysed by constructing four multilevel models. After controlling for the baseline measurement, there was no significant main effect of condition for sensitive parenting, externalizing problems, insecure attachment behaviour and parenting stress. There was however a significant effect of condition and age on insecure attachment behaviour: within the group of youngest foster children (under 30 months), children in the intervention group scored lower insecure attachment scores than children in the control group. Additionally, there was a significant effect of condition and type of foster care on externalizing behaviour problems. In the group of children placed in kinship foster care, children in the intervention group had lower external behavior scores than children in the control group.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Delphine West

Laura Gypen

Johan Vanderfaeillie

Appendix

References

Achenbach

T. M.

Rescorla

L. A.

(2000). Manual for the ASEBA preschool forms and profiles. University of Vermont, Research center for children.

Agentschap Opgroeien . (2022). Aantal pleegkinderen en pleegjongeren in stijgende lijn. [Number of foster children and foster youth on the rise] . https://pers.opgroeien.be/aantal-pleegkinderen-en-pleegjongeren-in-stijgende-lijn

Allen

(2011). Early intervention: the next steps, an independent report to Her Majesty’s government by Graham Allen MP. The Stationery Office. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/284086/early-intervention-next-steps2.pdf

Barone

Dellagiulia

Lionetti

(2018). Testing an attachment-based parenting intervention-VIPP-FC/A in adoptive families with post-institutionalized children: Do maternal sensitivity and genetic markers count? Frontiers in Psychology, 9, 156. https://doi.org/10.3389/fpsyg.2018.00156

Barone

Ozturk

Lionetti

(2019). The key role of positive parenting and children’s temperament in post-institutionalized children’s socio-emotional adjustment after adoption placement. A RCT study. Social Development, 28(1), 136–151. https://doi.org/10.1111/sode.12329

Bastiaensen

P. A. C. M.

(2001). Belaste pleegouders en verscheurde pleegkinderen? [Burdened foster parents and torn foster children? Radboud University.

Bates

Maechler

Bolker

Walker

Christensen

R. H. B.

Singmann

Dai

Scheipl

Grothendieck

Green

Fox

Bauer

Krivitsky

P. N.

(2009). Package ‘lme4. https://dk.archive.ubuntu.com/pub/pub/cran/web/packages/lme4/lme4.pdf

Bick

Dozier

(2013). The effectiveness of an attachment-based intervention in promoting foster mothers’ sensitivity toward foster infants. Infant Mental Health Journal, 34(2), 95–103. https://doi.org/10.1002/imhj.21373

Bongers

I. L.

Koot

H. M.

van der Ende

Verhulst

F. C.

(2003). The normative development of child and adolescent problem behavior. Journal of Abnormal Psychology, 112(2), 179–192. https://doi.org/10.1037/0021-843X.112.2.179

10.

Bowlby

(1969). Attachment and loss. In: Attachment (2nd ed., Vol. 1). Basic Books.

11.

Cohen

(1983). Applied multiple regression/correlation analysis for the behavioral sciences. Erlbaum.

12.

De Baat

De Lange

(2013). Pleegzorgbegeleiding [foster care counseling]. Nederlands Jeugdinstituut.

13.

Dozier

(2013). Attachment and biobehavioral catch-up. Unpublished manuscript, University of Delaware, Newark.

14.

Dozier

Manni

Gordon

M. K.

Peloso

Gunnar

M. R.

Stovall-McCloughEldrethLevine

K. C. D. S.

Levine

(2006). Foster children’s diurnal production of cortisol: An exploratory study. Child Maltreatment, 11(2), 189–197. https://doi.org/10.1177/1077559505285779

15.

Duhn

(2010). The importance of touch in the development of attachment. Advances in Neonatal Care: Official Journal of the National Association of Neonatal Nurses, 10(6), 294–300. https://doi.org/10.1097/ANC.0b013e3181fd2263

16.

Edwards

A. L.

(1990). Construct validity and social desirability. American Psychologist, 45(2), 287–289. https://doi.org/10.1037/0003-066X.45.2.287

17.

Faul

Erdfelder

Buchner

Lang

A. G.

(2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149–1160. https://doi.org/10.3758/BRM.41.4.1149

18.

Fox

Weisber

Price

Adler

Bates

Boud-Bovy

Bolker

Ellison

Firth

Friendly

Gorjanc

Graves

Heiberger

Krivitsky

Laboissiere

Maechler

Monette

Murdoch

Nilsson

Core

(2023). Package “car.” https://r-forge.r-project.org/projects/car/

19.

Fuentes-Peláez

Amorós

Pastor

Molina

Mateo

(2014). Assessment in kinship foster care: A new tool to evaluate the strengths and weaknesses. Social Sciences, 4(1), 1–17. https://doi.org/10.3390/socsci4010001

20.

Gabler

Bovenschen

Lang

Zimmermann

Nowacki

Kliewer

Spangler

(2014). Foster children’s attachment security and behavior problems in the first six months of placement: Associations with foster parents’ stress and sensitivity. Attachment and Human Development, 16(5), 479–498. https://doi.org/10.1080/14616734.2014.911757

21.

Goemans

Geel

M. van

Vedder

(2018). Foster children’s behavioral development and foster parent stress: Testing a transactional model. Journal of Child and Family Studies, 27(3), 990–1001. https://doi.org/10.1007/s10826-017-0941-z

22.

Goemans

van Geel

Vedder

(2015). Over three decades of longitudinal research on the development of foster children: A meta-analysis. Child Abuse and Neglect, 42, 121–134. https://doi.org/10.1016/j.chiabu.2015.02.003

23.

Herrenkohl

E. C.

Herrenkohl

R. C.

Egolf

B. P.

(2003). The psychosocial consequences of living environment instability on maltreated children. American Journal of Orthopsychiatry, 73(4), 367–380. https://doi.org/10.1037/0002-9432.73.4.367

24.

Holtan

Rønning

J. A.

Handegård

B. H.

Sourander

(2005). A comparison of mental health problems in kinship and nonkinship foster care. European Child and Adolescent Psychiatry, 14(4), 200–207. https://doi.org/10.1007/s00787-005-0445-z

25.

Hox

J. J.

Moerbeek

Van de Schoot

(2018). Multilevel analysis: Techniques and applications (3rd ed.). Routledge.

26.

Inchley

J. C.

Stevens

G. W. J. M.

Samdal

Currie

D. B.

(2020). Enhancing understanding of adolescent health and well-being: The health behaviour in school-aged children study. The Journal of adolescent health: Official Publication of the Society for Adolescent Medicine, 66(6S), S3–S5. https://doi.org/10.1016/j.jadohealth.2020.03.014

27.

Juffer

(2010). Beslissingen over kinderen in problematische opvoedingssituaties. Inzichten uit gehechtheidsonderzoek. [Decisions about children in problematic parenting situations. Insights from attachment research]. In: Den haag: Raad voor de Rechtspraak (issue 6). Raad voor de Rechtspraak.

28.

Juffer

Bakermans-Kranenburg

M. J.

van Ijzendoorn

M. H.

(2017a). Pairing attachment theory and social learning theory in video-feedback intervention to promote positive parenting. Current Opinion in Psychology, 15, 189–194. https://doi.org/10.1016/j.copsyc.2017.03.012

29.

Juffer

Bakermans-Kranenburg

M. J.

Van Ijzendoorn

M. H.

(2018). Video-feedback intervention to promote positive parenting and sensitive discipline: Development and meta-analytic evidence for its effectiveness. In Steele

Steele

(Eds), Handbook of attachment-based interventions (pp. 1–26). The Guilford Press.

30.

Juffer

Struis

Werner

Bakermans-Kranenburg

M. J.

(2017b). Effective preventive interventions to support parents of young children: Illustrations from the video-feedback intervention to promote positive parenting and sensitive discipline (VIPP-SD). Journal of Prevention & Intervention in the Community, 45(3), 202–214. https://doi.org/10.1080/10852352.2016.1198128

31.

Keller

T. E.

Wetherbee

Le Prohn

N. S.

Payne

Sim

Lamont

E. R.

(2001). Competencies and problem behaviors of children in family foster care: Variations by kinship placement status and race. Children and Youth Services Review, 23(12), 915–940. https://doi.org/10.1016/S0190-7409(01)00175-X

32.

Kelly

J. F.

Zuckerman

Rosenblatt

(2008). Promoting first relationships: A relationship-focused early intervention approach. Infants and Young Children, 21(4), 285–295. https://doi.org/10.1097/01.iyc.0000336541.37379.0e

33.

Klein Velderman

Bakermans-Kranenburg

M. J.

Juffer

Van Ijzendoorn

M. H.

Mangelsdorf

S. C.

Zevalkink

(2006). Preventing preschool externalizing behavior problems through video-feedback intervention in infancy. Infant Mental Health Journal, 27(5), 466–493. https://doi.org/10.1002/imhj.20104

34.

Kuznetsova

Brockhoff

P. B.

Christensen

R. H. B.

(2017). lmerTest package: Tests in linear mixed effects models. Journal of Statistical Software, 82(13), 1–26. https://doi.org/10.18637/JSS.V082.I13

35.

Lang

Bovenschen

Gabler

Zimmermann

Nowacki

Kliewer

Spangler

(2016). Foster children’s attachment security in the first year after placement: A longitudinal study of predictors. Early Childhood Research Quarterly, 36(33), 269–280. https://doi.org/10.1016/j.ecresq.2015.12.019

36.

Leathers

S. J.

(2006). Placement disruption and negative placement outcomes among adolescents in long-term foster care: The role of behavior problems. Child Abuse and Neglect, 30(3), 307–324. https://doi.org/10.1016/j.chiabu.2005.09.003

37.

Lind

Lee Raby

Caron

E. B.

Roben

C. K.

Dozier

(2017). Enhancing executive functioning among toddlers in foster care with an attachment-based intervention. Development and Psychopathology, 29(2), 575–586. https://doi.org/10.1017/S0954579417000190

38.

Lüdecke

Bartel

Schwemmer

Powell

Djalovski

Titz

(2023). Package “sjPlot” title data visualization for statistics in social science. R Foundation for Statistical Computing.

39.

Lydersen

(2022). Justering for utgangsverdi i longitudinelle randomiserte studier [Adjustment for baseline value in longitudinal randomised trials]. Tidsskrift for Den Norske Legeforening. https://doi.org/10.4045/tidsskr.21.0824

40.

McCoy

(2017). Understanding the intention-to-treat principle in randomized controlled trials. Western Journal of Emergency Medicine, 18(6), 1075–1078. https://doi.org/10.5811/westjem.2017.8.35985

41.

Mendelsohn

A. L.

Valdez

P. T.

Flynn

Foley

G. M.

Berkule

S. B.

Tomopoulos

Fierman

A. H.

Tineo

Dreyer

B. P.

(2007). Use of videotaped interactions during pediatric well-child care: Impact at 33 Months on parenting and on child development. Journal of developmental and behavioral pediatrics: Journal of Developmental and Behavioral Pediatrics, 28(3), 206–212. https://doi.org/10.1097/DBP.0b013e3180324d87

42.

Neece

C. L.

Green

S. A.

Baker

B. L.

(2012). Parenting stress and child behavior problems: A transactional relationship across time. American Journal on Intellectual and Developmental Disabilities, 117(1), 48–66. https://doi.org/10.1352/1944-7558-117.1.48

43.

Newton

R. R.

Litrownik

A. J.

Landsverk

J. A.

(2000). Children and youth in foster care: Distangling the relationship between problem behaviors and number of placements. Child Abuse and Neglect, 24(10), 1363–1374. https://doi.org/10.1016/S0145-2134(00)00189-7

44.

O’NeillTempleneill

R. T. R. R. T.

Temple

(2012). The prevention and treatment of missing data in clinical trials: An FDA perspective on the importance of dealing with it. Clinical Pharmacology and Therapeutics, 91(3), 550–554. https://doi.org/10.1038/clpt.2011.340

45.

Pecora

P. J.

White

C. R.

Jackson

L. J.

Wiggins

(2009). Mental health of current and former recipients of foster care: A review of recent studies in the USA. Child and Family Social Work, 14(2), 132–146. https://doi.org/10.1111/j.1365-2206.2009.00618.x

46.

Perry

K. J.

(2015). Contextual and child history predictors of internalizing and externalizing behavior problems in foster children. San Diego State University.

47.

Pleegzorg Vlaanderen . (2020). Registratierapport 2019. Een blik op een jaar pleegzorg in cijfers. [Registration report 2019. A look at one year of foster care in figures] .

48.

Podsakoff

P. M.

MacKenzie

S. B.

Lee

J.-Y.

Podsakoff

N. P.

(2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879–903. https://doi.org/10.1037/0021-9010.88.5.879

49.

Polderman

Kellaert-Knoll

M. G.

(2008). Manual of the attachment insecurity screening inventory 2–5 years (AISI 2–5 years). Basic Trust.

50.

Ponciano

(2010). Attachment in foster care: The role of maternal sensitivity, adoption, and foster mother experience. Child and Adolescent Social Work Journal, 27(2), 97–114. https://doi.org/10.1007/s10560-010-0192-y

51.

Saghaei

(2011). Implementation of an open-source customizable minimization program for allocation of patients to parallel groups in clinical trials. Journal of Biomedical Science and Engineering, 04(11), 734–739. https://doi.org/10.4236/jbise.2011.411090

52.

Schoemaker

N. K.

Jagersma

Stoltenborgh

Maras

Vermeer

H. J.

Juffer

Alink

L. R. A.

(2018). The effectiveness of video-feedback intervention to promote positive parenting for foster care (VIPP-FC): Study protocol for a randomized controlled trial. BMC Psychology, 6(1), 38. https://doi.org/10.1186/s40359-018-0246-z

53.

Schoemaker

N. K.

Juffer

Rippe

R. C. A.

Vermeer

H. J.

Stoltenborgh

Jagersma

G. J.

Maras

Alink

L. R. A.

(2020). Positive parenting in foster care: Testing the effectiveness of a video-feedback intervention program on foster parents’ behavior and attitudes. Children and Youth Services Review, 110, 104779. https://doi.org/10.1016/j.childyouth.2020.104779

54.

Schoemaker

N. K.

Wentholt

W. G. M.

Goemans

Vermeer

H. J.

Juffer

Alink

L. R. A.

(2020). A meta-analytic review of parenting interventions in foster care and adoption. Development and Psychopathology, 32(3), 1149–1172. https://doi.org/10.1017/S0954579419000798

55.

Shlonsky

A. R.

Berrick

J. D.

(2001). Assessing and promoting quality in kin and nonkin foster care. Social Service Review, 75(1), 60–83. https://doi.org/10.1086/591882

56.

Spieker

S. J.

Oxford

M. L.

Kelly

J. F.

Nelson

E. M.

Fleming

C. B.

(2012). Promoting first relationships: Randomized trial of a relationship-based intervention for toddlers in child welfare. Child Maltreatment, 17(4), 271–286. https://doi.org/10.1177/1077559512458176

57.

Sprang

(2009). The efficacy of a relational treatment for maltreated children and their families. Child and Adolescent Mental Health, 14(2), 81–88. https://doi.org/10.1111/j.1475-3588.2008.00499.x

58.

Stovall-McClough

K. C.

Dozier

(2004). Forming attachments in foster care: Infant attachment behaviors during the first 2 months of placement. Development and Psychopathology, 16(02), 253–271. https://doi.org/10.1017/s0954579404044505

59.

Strijker

Knorth

E. J.

Knot-Dickscheit

(2008). Placement history of foster children: A study of placement history and outcomes in long-term family foster care. Child Welfare, 87(5), 107–124.

60.

Strijker

Van Oijen

(2006). (On)gelijkenis van pleegouder en pleegkind in hun rapportage van probleemgedrag [(In)similarity of foster parent and foster child in their reporting of problem behavior]. Tijdschrift Voor Orthopedagogiek, Kinderpsychiatrie En Klinische Kinderpsychologie, 31(1), 69–81.

61.

Strijker

Zandberg

van der Meulen

B. F.

(2003). Kinship foster care and foster care in The Netherlands. Children and Youth Services Review, 25(11), 843–862. https://doi.org/10.1016/S0190-7409(03)00089-6

62.

Tarren-Sweeney

(2014). The clinical application of attachment theory and research: Introducing a series of Clinical Child Psychology and Psychiatry special sections. Clinical Child Psychology and Psychiatry, 19(3), 333–335. https://doi.org/10.1177/1359104514538709

63.

Taves

D. R.

(2010). The use of minimization in clinical trials. Contemporary Clinical Trials, 31(2), 180–184. https://doi.org/10.1016/j.cct.2009.12.005

64.

Twisk

J. W.

(2021). Analysis of data from randomized controlled trials. Springer International Publishing.

65.

van Andel

H. W.

Grietens

Knorth

E. J.

(2012). Foster carer-foster child intervention (FFI): An intervention designed to reduce stress in young children placed in a foster family. Adoption and Fostering, 36(2), 19–28. https://doi.org/10.1177/030857591203600203

66.

Van den Bergh

P. M.

Weterings

(2010). Pleegzorg in perspectief: Ontwikkelingen in theorie en praktijk [Foster care in perspective: Developments in theory and practice]. Uitgeverij Van Gorcum.

67.

Van den Bruel

(2012). Naar een nieuw Vlaams intersectoraal decreet Pleegzorg [Towards a new Flemish intersectoral decree Foster care]. In Vanderfaeillie

Van Holen

Vanschoonlandt

(Eds), Op weg met pleegzorg. Kansen en risico’s (pp. 43–63). Acco.

68.

Van den Bruel

Stroobants

(2016). Uitgangspunten en krachtlijnen bij uithuisplaatsing van jonge kinderen i.f.v. Vlaamse beleidskeuzes [Principles and guidelines for out-of-home placement of young children in Flanders].

69.

van den Dries

Juffer

van Ijzendoorn

M. H.

Bakermans-Kranenburg

M. J.

(2009). Fostering security? A meta-analysis of attachment in adopted children. Children and Youth Services Review, 31(3), 410–421. https://doi.org/10.1016/j.childyouth.2008.09.008

70.

Vanderfaeillie

Van Holen

(2010). Het verloop van pleeggezinplaatsingen [The course of foster care placements] In: Pleegzorg in perspectief. Ontwikkelingen in theorie en praktijk (pp. 177–193). Van Gorcum.

71.

Vanderfaeillie

van Holen

Trogh

Andries

(2012). The impact of foster children’s behavioural problems on Flemish foster mothers’ parenting behaviour. Child and Family Social Work, 17(1), 34–42. https://doi.org/10.1111/j.1365-2206.2011.00770.x

72.

Vanderfaeillie

Van Holen

Vanschoonlandt

Robberechts

Stroobants

(2013). Children placed in long-term family foster care: A longitudinal study into the development of problem behavior and associated factors. Children and Youth Services Review, 35(4), 587–593. https://doi.org/10.1016/j.childyouth.2012.12.012

73.

Vanderfaeillie

Vanschoonlandt

Van Holen

De Maeyer

Robberechts

(2014). Traumatische gebeurtenissen en traumatische stresssymptomen bij pleegkinderen: Een verkennende studie in vlaanderen [traumatic events and traumatic stress symptoms in foster children: An exploratory study in flanders]. Kind en Adolescent, 35(3), 135–149. https://doi.org/10.1007/s12453-014-0021-8

74.

Van Holen

Vanderfaeillie

Vanschoonlandt

(2012). Op weg met pleegzorg: Kansen en risico’s [on the way with foster care: Opportunities and risks. Acco.

75.

Van Ijzendoorn

M. H.

(2008). Opvoeding over de grens. Gehechtheid, trauma en veerkracht [Parenting across borders. Attachment, trauma and resilience ]. Boom Academic.

76.

Van Ijzendoorn

M. H.

Schuengel

Bakermans-Kranenburg

M. J.

(1999). Disorganized attachment in early childhood: Meta-analysis of precursors, concomitants, and sequelae. Development and Psychopathology, 11(2), 225–249. https://doi.org/10.1017/s0954579499002035

77.

van Ijzendoorn

M. H.

Schuengel

Wang

Bakermans-Kranenburg

M. J.

(2023). Improving parenting, child attachment, and externalizing behaviors: Meta-analysis of the first 25 randomized controlled trials on the effects of Video-feedback Intervention to promote Positive Parenting and Sensitive Discipline. Development and Psychopathology, 35(1), 241–256. https://doi.org/10.1017/S0954579421001462

78.

van Oijen

(2010). Resultaat van pleegzorgplaatsingen: Een onderzoek naar breakdown en de ontwikkeling van adolescente pleegkinderen bij langdurige pleegzorgplaatsingen [outcomes of foster care placements: An examination of breakdown and adolescent foster child development in long-term foster care placements]. University of Groningen.

79.

Vanschoonlandt

Vanderfaeillie

van Holen

de Maeyer

(2012). Development of an intervention for foster parents of young foster children with externalizing behavior: Theoretical basis and program description. Clinical Child and Family Psychology Review, 15(4), 330–344. https://doi.org/10.1007/s10567-012-0123-x

80.

Vanschoonlandt

Vanderfaeillie

Van Holen

De Maeyer

Andries

(2012). Kinship and non-kinship foster care: Differences in contact with parents and foster child’s mental health problems. Children and Youth Services Review, 34(8), 1533–1539. https://doi.org/10.1016/j.childyouth.2012.04.010

81.

Vanschoonlandt

Vanderfaeillie

Van Holen

De Maeyer

Robberechts

(2013). Parenting stress and parenting behavior among foster mothers of foster children with externalizing problems. Children and Youth Services Review, 35(10), 1742–1750. https://doi.org/10.1016/j.childyouth.2013.07.012

82.

Van Zeijl

Mesman

Van Ijzendoorn

M. H.

Bakermans-Kranenburg

M. J.

Juffer

Stolk

M. N.

Koot

H. M.

Alink

L. R. A.

(2006). Attachment-based intervention for enhancing sensitive discipline in mothers of 1- to 3-year-old children at risk for externalizing behavior problems: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 74(6), 994–1005. https://doi.org/10.1037/0022-006X.74.6.994

83.

Vasileva

Petermann

(2018). Attachment, development, and mental health in abused and neglected preschool children in foster care: A meta-analysis. Trauma, Violence, & Abuse, 19(4), 443–458. https://doi.org/10.1177/1524838016669503

84.

Verhulst

F. C.

van der Ende

(2000). Nederlandstalige bewerking child behavior checklist [Dutch adaptation child behavior checklist]. Aseba.

85.

Vermulst

Kroes

De Meyer

Nguyen

Veerman

J. W.

(2012). Opvoedingsbelastingvragenlijst (OBVL). Handleiding [parenting stress questionnaire (PSQ). Manual]. Praktikon.

86.

Vliegen

Bijttebier

Boulpaep

Luyten

Cluckers

Biringen

(2005). De EA-SR: Een zelfrapporteringsschaal voor het meten van emotionele beschikbaarheid bij ouders van jonge kinderen (0- 1jaar). [The EA-SR: A parental self-report scale measuring emotional availability in interaction with infants (0-1 year)]. Diagnostiek-Wijzer, 8(4), 137–147.

87.

Vliegen

Luyten

Biringen

(2009). A multimethod perspective on emotional availability in the postpartum period. Parenting, 9(3–4), 228–243. https://doi.org/10.1080/15295190902844514

88.

Waters

Deane

K. E.

(1985). Defining and assessing individual differences in attachment relationships: Q-methodology and the organization of behavior in infancy and early childhood. Monographs of the Society for Research in Child Development, 50(1–2), 41. https://doi.org/10.2307/3333826

89.

West

Luys

Gypen

Van Holen

Vanderfaeillie

(2023). Behavior problems in foster care, systematic review of associated factors. Children and Youth Services Review, 155, 107240. https://doi.org/10.1016/j.childyouth.2023.107240

90.

West

Vanderfaeillie

Van Hove

Gypen

Van Holen

(2020). Attachment in family foster care: Literature review of associated characteristics. Developmental Child Welfare, 2(2), 132–150. https://doi.org/10.1177/2516103220915624

91.

West

Van Holen

Verheyden

Gypen

Vanderfaeillie

(2022). Attachment of young foster children. Developmental Child Welfare, 4(4), 270–287. https://doi.org/10.1177/25161032221129287

92.

Whittaker

Sutton

Burton

(2006). Pragmatic randomised controlled trials in parenting research: The issue of intention to treat. Journal of epidemiology and community health, 60(10), 858–864. https://doi.org/10.1136/jech.2005.044214

93.

Wilson

Sinclair

Taylor

Pithouse

Sellick

(2004). Conceptualising and effecting good outcomes in foster care: An exploration of the research literature. Policy Press.

94.

Wissink

I. B.

Colonnesi

Stams

G. J. J. M.

Hoeve

Asscher

J. J.

Noom

M. J.

Polderman

Kellaert-Knol

M. G.

(2016). Validity and reliability of the attachment insecurity screening inventory (AISI) 2–5 years. Child Indicators Research, 9(2), 533–550. https://doi.org/10.1007/s12187-015-9322-6

Effectiveness of a video-feedback intervention to promote positive parenting for foster children

Abstract

Keywords

Introduction

Foster care for young flemish foster children

The need for interventions in foster care

Video-feedback intervention to promote positive parenting for foster children

Research question

Methods

Participants

Procedure

Instruments

Emotional availability–Self report

Attachment insecurity screening inventory 2–5 years

Parenting Stress Questionnaire

Child behavior checklist 1,5–5

VIPP-FC/A intervention

Data analysis

Results

Baseline differences

Intervention effects of VIPP-FC/A

Sensitive parenting behavior

Insecure attachment behaviour

Externalizing behaviour problems

Parenting stress

Discussion

Limitations

Suggestions for further research

Conclusion

Footnotes

Declaration of conflicting interests

Funding

ORCID iDs

Appendix

References