Training in Nonviolent Resistance for Foster Parents

Abstract

Objective:

The aim of this study is to evaluate a training in non-violent resistance (NVR) for foster parents who take care of a foster child (ages 6-18) with externalizing problem behavior.

Methods:

A randomized controlled trial was used to compare an intervention group (NVR, n = 31) with a treatment as usual control group (TAU, n = 31). The NVR-intervention consists of ten weekly home sessions. Measures regarding behavioral problems in foster children, parenting stress and parenting practices in foster mothers, and the size of the supportive network were assessed before, after treatment, and at three months follow-up.

Results:

NVR showed to be an acceptable approach that lead to an increase in experienced support and some promising changes in parenting stress and parenting practices.

Conclusion:

Implementation of this intervention might increase the effectiveness of foster care. More longitudinal research using a Multitrait-multimethod-approach is however needed.

Keywords

foster care intervention RCT

Placement in family foster care is growing in many countries (Fernandez & Barth, 2011). Internationally, a high proportion of the children placed in foster families manifest severe behavior problems (Holtan, Ronning, Handegard, & Sourander, 2005), which poses serious challenges to the foster-care system (e.g., Sawyer, Carbone, Searle, & Robinson, 2007). The proportion of foster children with serious behavior problems is estimated between 33% and 85% (Holtan et al., 2005). Although some studies have found minor improvement in the child’s behavior during foster placement (e.g., Wilson, 2006), most reported no improvement or even deterioration (e.g., Fernandez, 2009; Nilsen, 2007). This creates severe difficulties for the foster parents, as the severity of behavior problems is one of the main causes of foster parent stress (Fisher & Stoolmiller, 2008; Whenan, Oxlad, & Lushington, 2009) and the breakdown of placements in foster families (Chamberlain et al., 2006; Oosterman, Schuengel, Slot, Bullens, & Dorelijers, 2007). Severe behavior problems have also been linked to an increase in noneffective parenting practices (Linares, Montalto, Rosbruch, & Li, 2006; Vanderfaeillie, Van Holen, Trogh, & Andries, 2012) probably by making the foster parents feel more helpless and consequently more punitive (Fisher, Gunnar, Chamberlain, & Reid, 2000). The availability of a supportive social network is considered to be an important protective factor (Orme et al., 2004) being associated with lower stress levels in foster parents and more resilience in the face of difficulties (Fisher & Stoolmiller, 2008). Lack of support on the other hand has been implicated in foster parent distress and in reduced effectiveness of foster placements (Fisher & Stoolmiller, 2008). A survey of more than 1,600 Flemish foster carers showed that for a considerable group of foster parents, the supportive network was very limited (Bronselaer, Vandezande, & Verreth, 2011), and for many, foster-care workers were the main if not the only source of support (Vanderfaeillie, 2011).

Consequently, interventions aiming at improving the well-being of foster children and foster families are badly needed (Havlicek, Garcia, & Smith, 2013; Leve et al., 2012). For the present project, there were four reasons for choosing nonviolent resistance (NVR) as a training method for foster parents: (1) Most evidence-based programs for foster families focus on younger children (Leve et al., 2012), whereas programs involving older children tend to be less effective (Maughan, Christiansen, Jenson, Olympia, & Clark, 2005). NVR has shown itself similarly applicable and helpful with younger children and adolescents (Ollefs, von Schlippe, Omer, & Kriz, 2009; Weinblatt & Omer, 2008); (2) NVR focuses on parent distress (Lavi-Levavi, Shachar, & Omer, 2013; Oleffs et al., 2009). Addressing foster parent distress has been shown to be a critical target for interventions aiming at improving foster children’s well-being (Fisher & Stoolmiller, 2008); (3) Dropout rates in NVR are very low, contrasting favorably with the typically higher dropout rates of many other parent training programs (Friars & Mellor, 2009); and (4) NVR places strong emphasis on the creation of a supportive network for the parents.

This article presents the results of a randomized controlled trial (RCT) in foster care comparing an NVR intervention group with a treatment as usual (TAU) condition. The main hypothesis was that NVR would lead to improvements in child behavior problems, foster parent stress, foster parenting educational strategies, and experienced support relative to TAU.

Method

Enrollment of Participants

In order to maximize the external validity, the intervention was carried out and examined in a “real-life setting.” The study took place in three of the five Flemish provinces (Dutch speaking part of Belgium) from July 2010 to September 2012. Participants were recruited in two ways until the study reached its scheduled date of closure. First, all new foster-care placements with a long-term perspective (>1 year) of children aged between 6 and 18 (n = 304) were screened approximately 4 months after the start of the placement. Secondly, foster-care workers could sign up foster parents of ongoing long-term foster-care placements (n = 89). Foster mothers completed a Child Behavior Checklist (CBCL/6-18; Achenbach & Rescorla, 2001) and foster-care workers completed a questionnaire regarding characteristics relating to the foster child, the current foster-care placement, the case history, the parents, and the foster parents. Foster parents were eligible if their foster child had a borderline or clinical score on the externalizing broad band or on one of the externalizing small-band scales of the CBCL. The following exclusion criteria were used: intellectual disability, autism, unstable use of psychotropic medication (psychotropic medication use must have started at least 2 months before the start of the intervention and must be stable for at least 2 weeks before start of the intervention), and behavioral problems stemming from medical problems (e.g., Prader–Willi syndrome) or medication (e.g., anticonvulsive drugs). Foster parents who were currently involved in divorce proceedings or foster parents with a current mental health disorder, measured with the General Health Questionnaire (Koeter & Ormel, 1991) and defined as a score ≥ 2, were excluded. This is because pilot implementation in nine foster families (as part of the training of the therapists) suggested that additional interventions are needed in such cases (Vanschoonlandt, Van Holen, & Vanderfaeillie, 2012). In families with more than one eligible foster child, the foster child with more serious behavioral problems was considered in the study. Refusal to participate did not influence the regular care offered.

Design

The study used a RCT in which participants were assigned to an intervention group (who received the NVR intervention) or a control group (who received TAU). During the study period, a total of 87 foster families were eligible for the study and 62 (71.3%) families eventually participated. Given the relatively small sample size, restricted randomization (minimization) was used to assign eligible foster parents to the intervention condition (n = 31) or the control condition (n = 31). In this procedure, important prognostic factors are identified before the trial starts, and assignment of a new participant is determined so as to minimize the differences between groups (Scott, McPherson, Ramsay, & Campbell, 2002). Minimization is preferable to simple randomization when working with small, heterogeneous groups (Scott et al., 2002), as it ensures groups are balanced with respect to both the number of participants and the characteristics of each group (Taves, 2010). The disadvantage of minimization is that in certain cases, the next allocation can be predicted (Scott et al., 2002). In order to ensure the concealment of the allocation sequence, as suggested by McEntegart (2003), each allocation of participants was done with a probability of .90. The following factors were included in minimization: age of the foster child (less than 12 years vs. at least 12 years), gender of the foster child, previous out-of-home placements of the foster child (none vs. at least one), type of foster placement (kinship vs. nonkinship), educational level of the foster parents (whether or not at least one foster parent had followed higher education), foster family composition (one parent vs. two parents), duration of foster placement (less than 18 months vs. at least 18 months), and severity of the foster child’s externalizing problems (borderline vs. clinical score on the broad-band externalizing CBCL Scale).

Foster mothers filled out a questionnaire at three points: (1) within a period of 2 weeks prior to the start of the intervention (T ₀), (2) immediately after the intervention (T ₁), and (3) follow-up, 3 months after the intervention (T ₂). We chose to involve only foster mothers in data collection because almost a quarter of our sample consisted of single parent families parented by women (see further).

Enrollment of participants and allocation sequence was generated by one of the researchers, completely separated from the team administrating the intervention, the foster-care workers, and the foster parents. Besides adding a probability, the allocation was concealed from the researcher using the free, open source, computerized minimization program “MinimPy” (Saghaei, 2011). Blinding could not be implemented. After assignment, participants and care providers were informed. Informed consent of the participants was required prior to participation. Full details of the trial can be obtained from the first author. The study was approved by the faculty management of the Faculty of Psychology and Educational Sciences of the Vrije Universiteit Brussel. The trial was not registered.

Description of the Sample

Foster families in the intervention group (see Table 1) were mainly two parent families (74.2%) with on average 1.74 biological children (SD = 1.46) and 1.55 foster children (SD = .68). All single parent households (n = 8) were parented by women. Foster mothers had a mean age of 47.0 years (SD = 12.06) and foster fathers were, on average, 47.4 years old (SD = 12.18). In most households, no foster parent (58.1%) had higher education. In all, 48.4% of the target children were boys, and the mean age was 11.6 years (SD = 3.46). Most of the foster children (64.5%) were previously placed in out-of-home care. The current foster placement had a mean duration of 46.7 months (SD = 53.54), and 54.8% were kinship placements.

Table 1.

Foster Family, Foster Child, and Placement Characteristics of the Study Groups.^a

Characteristic	Intervention Group (n = 31)	Control Group (n = 31)
Type of household
Single parent	8 (25.8)	7 (22.6)
Two parent	23 (74.2)	24 (77.4)
Number of biological children	1.74 (1.46)	1.61 (1.26)
Number of foster children	1.55 (0.68)	1.42 (0.62)
Age foster mother, years	47.0 (12.06)	46.8 (13.15)
Age foster father, years	47.4 (12.18)	46.8 (14.08)
Education foster parents
No higher education	18 (58.1)	19 (61.3)
Higher education (at least one foster parent)	13 (41.9)	12 (38.7)
Gender foster child
Boy	15 (48.4)	17 (54.8)
Girl	16 (51.6)	14 (45.2)
Age foster child, years	11.6 (3.46)	12.3 (3.49)
Placement history
No	11 (35.5)	9 (29.0)
Yes	20 (64.5)	22 (71.0)
Duration of placement (in months)	46.7 (53.54)	35.1 (39.91)
Type of placement
Kinship	17 (54.8)	20 (64.5)
Nonkinship	14 (45.2)	11 (35.5)

^aData are means (SD) or numbers (%).

Foster families in the control group (see Table 1) were mainly two parent families (77.4%) with on average 1.61 biological children (SD = 1.26) and 1.42 foster children (SD = 0.62). All single parent households (n = 7) were parented by women. Foster mothers had a mean age of 46.8 years (SD = 13.15) and foster fathers were, on average, 46.8 years old (SD = 14.08). In most households, no foster parent (61.3%) had higher education. In all, 54.8% of the target children were boys, and the mean age was 12.3 years (SD = 3.49). Most of the foster children (71.0%) were previously placed in out-of-home care. The current foster placement had a mean duration of 35.1 months (SD = 39.91), and 64.5% were kinship placements.

Treatment Description (Intervention Group)

The intervention (Van Holen, Vanderfaeillie, & Omer, 2016) was an adaptation for foster families of the NVR treatment program for parents of violent and self-destructive children (Omer, 2004, 2011). NVR places escalation processes at the center of attention (Omer, Schorr-Sapir, & Weinblatt, 2008). The underlying assumption is that parental submission and power struggles are mutually enhancing and that they feed on and are fed by negative feelings. Foster parents, who previously felt helpless and were caught up in escalation with the foster child, are trained to effectively resist the foster child’s negative behavior without lashing out or giving in (Omer & Lebowitz, 2016; Omer, Steinmetz, Carthy, & von Schlippe, 2013). Patterson’s coercion theory (Patterson, Dishion, & Bank, 1984) is the underlying model. To achieve this, NVR focuses on the following four intervention areas.

Prevention of escalation

Emotional regulation of foster parents is trained in order to prevent and halt escalating cycles. Foster parents learn to recognize escalatory patterns and identify their own and their foster child’s typical reactions and the associated thoughts and feelings. Alternative ways of responding in nonescalating manners are taught and rehearsed. For example, foster parents learn to delay their response (“Strike the iron when it’s cold!”) and to abstain from controlling and domineering messages (“You don’t have to win, only to persist!”).

Resisting problem behavior

The foster parents aim at resisting rather than controlling the child’s negative behaviors. Depending on the risks and the foster child’s specific problems, Omer (2004, 2011) developed well-documented techniques to help foster parents to resist problem behavior in a respectful and nonviolent way:

Delivery of a formal announcement in which the foster parents declare their decision to resist the child’s negative behaviors

This announcement is delivered in writing and read aloud by the foster parents. In accordance with the treatment’s emphasis on parental self-control, it is written in the first person plural (“We will no longer accept …”) and not in the second person singular (“You will have to …”). The announcement also stipulates that the foster parents will not keep the problems secret but will seek help from supporters. Foster parents rehearse how to deliver the announcement and how to develop nonescalating responses to the foster child’s reactions.

Performance of “sit-ins”

The foster parents enter the child’s room at a quiet time, sit down, and announce that they will sit and wait for a proposal by the child to stop the problem behavior that triggered the sit-in: “We are here because we are no longer willing to accept the kind of behavior you displayed. We will sit here and wait for a proposal as to how this behavior might end.” The foster parents are trained to remain quiet and strictly avoid arguments or escalation. The therapist helps them to develop ways of coping with typical reactions, such as attempts to expel them, ignore them, or deride them, and instructs them as to how to end the sit-in and resume daily life. The sit-in serves as a manifestation of resistance that does not depend on the child’s compliance for success and that can be performed without escalating into negative cycles of aggression.

Documentation of negative behaviors

The foster child’s unacceptable acts are documented by the foster parents, shown to the foster child, and distributed to the supporters. Foster parents tell their foster child that they are no longer keeping the events secret and that they will send their report to whomever they feel is appropriate. Supporters are specifically asked to address the foster child in a positive way, to make clear that they know what happened, and to offer help in finding solutions for stopping those behaviors.

Increasing supervision by telephone rounds or parental visitation

In the telephone rounds, foster parents react to the foster child’s failing to come home in time. Foster parents call a previously prepared list of friends, acquaintances, and relevant contacts, telling them that their foster child has not come home, asking for help, and requesting them to tell the foster child that they are looking for him or her. Foster parents are rigorously instructed as to how to prevent escalation, once the foster child returns home. In the parental visitation, foster parents actually go to the place where the foster child spends his or her time without parental permission. They are instructed in detail on how to behave so as to prevent escalation.

Creating a network of support

Foster parents are encouraged to activate potential sources of support in their social network such as family, friends, acquaintances, and professionals (e.g., school staff). Involving other people in what is happening at home and seeking their help is a major factor in coping with the child’s negative behavior. Whenever possible, a meeting with the supporters is organized by the therapist to explain the purpose and principles of the treatment and to discuss how and when the supporters can help. When a supporters’ meeting is not feasible, supporters are recruited on an individual base. Some typical roles of supporters are: to back the foster parents’ acts of resistance, to offer emotional and/or practical help for foster parents and/or the foster child, to help in breaking the seal of secrecy that often surrounds negative behaviors, to mediate in situations of polarization, to help defuse situations of acute escalation, and to offer help in finding acceptable solutions.

Relational gestures

Foster parents are encouraged to initiate positive interactions by systematic relational gestures such as signs of appreciation, suggestions of shared activities, and symbolic gifts. Frequently used is the album or box of positive memories, which documents good times, and positive opinions about the child such as short stories, a ticket from a nice vacation, photos, and reminders of events such as a family trip, parties, and so on. Foster parents invite friends and members of the birth family to participate. These gestures are unilateral initiatives by the foster parents. They are independent of the foster child’s behavior and are aimed at promoting positive aspects of the parent–child relationship. They are acts of caring that show the foster parents’ love independently of their ongoing resistance to the foster child’s negative behaviors.

The foster parent intervention consisted of 10, usually weekly, home sessions of 75 min and 1 telephone support session between every 2 home sessions. A detailed training manual was developed, describing the treatment’s rationale, providing guidelines for each intervention area, and outlining the sequence and contents of the treatment sessions (Van Holen, Vanderfaeillie, & Vanschoonlandt, 2013). The training manual, including training materials, can be obtained from the first author. A summarized overview of the intervention can be found in Table 2. The main modifications of the original program include (1) use of a home-visit format in order to lower barriers to service access; (2) development of practical aids, such as hand-outs, worksheets, a workbook for foster parents, and a DVD illustrating NVR techniques; (3) development of special components for foster families and foster children (e.g., guidelines describing when and how to involve members of the biological family in the support network, for instance to engage them in relational gestures); and (4) treatment administration by experienced foster-care workers who are best acquainted with the needs of foster families (Maaskant, 2010).

Table 2.

Overview Content of the Intervention.

Session 1—intake session (in the presence of the regular foster-care worker)	An overview of the content of the intervention is given Escalation patterns are identified Treatment goals are formulated Homework assignment: Foster parents write an announcement
Session 2	Discussion/modification of the announcement Rehearsal of the execution of the announcement Homework assignment: Foster parents read the announcement to the foster child
Session 3	Escalatory patterns, typical reactions, and associated thoughts and feelings are identified Alternative ways of responding in nonescalating manners are rehearsed Homework assignment: Foster parents bring into practice deescalation steps
Session 4	A plan for the systematical use of relation gestures is developed Homework assignment: Continuing bringing into practice deescalation Start-up offering relation gestures
Session 5	The supportive network is explored in detail Homework assignment: systematic and cumulative offering of relation gestures
Session 6–9	Activating social support. If feasible, a supporters’ meeting is organized A plan to deal with negative behaviors is developed. Foster parents rehearse and bring into practice-specific nonviolent resistance techniques: Sit-in Documentation of negative behaviors Telephone rounds Parental visitation
Session 10—evaluation (in the presence of the regular foster-care worker)	Treatment is evaluated Developments in foster parents and foster child are discussed Relapse prevention is done

Treatment in the experimental group was administered by three experienced foster-care workers who received special training in NVR consisting of 12 4-hr sessions. As part of the training, each therapist treated three foster families under close supervision. Treatment integrity and quality was ensured by fortnightly group supervision sessions. Fidelity measurements were made by comprehensive treatment checklists.

TAU (Control Group)

The control group was given TAU. In Flanders, foster-care workers organize support for the foster child, optimize contacts with birth parents and family, and coach and train foster parents (Verreth, 2009). More specifically, the support for foster-care situations comprises of at least seven face-to-face contacts a year (Sprangers, 2009). However, it is not defined with whom these contacts should take place. They can be with foster parents, foster children, birth parents, the wider context of the foster child (e.g., grandparents), and combinations of the parties involved (e.g., foster parents and foster child together). Furthermore, certain aspects of good practice (e.g., the use of care plans) are obligatory (www.pleegzorgvlaanderen.be). Although foster-care workers have great autonomy within these guidelines, a caseload of 25 foster-care placements for a full-time foster-care worker (Vanschoonlandt, Vanderfaeillie, Van Holen, De Maeyer, & Robberechts, 2013) hinders them from providing intensive support to foster parents. Bronselaer, Vandezande, and Verreth (2011), for example, found that the majority of foster parents (56%) only had a few contacts (face-to-face or by telephone) a year with their foster-care worker. Herewith, nothing is said about the content of these contacts nor about the practices used by the foster-care worker. In addition to the regular foster-care support described above, foster parents have access to external mental health-care services for themselves or for their foster child. A Flemish study showed that for 17.5% of the foster children and for 9.1% of the foster parents, respectively, additional external help was offered or planned (Vanschoonlandt et al., 2013). In short, the help offered during a foster-care placement is very diverse and heterogeneous and the support for foster families varies enormously (Van Holen, Vanderfaeillie, De Maeyer, & Gypen, 2015). As a consequence, it is not unthinkable that the TAU received by foster families in a control group differs considerably between participants. To control this factor, we asked foster-care workers to register not only their own contacts with the foster family but also referrals to external mental health services (Herdewyn, 2011).

Measures

CBCL/6-18 (Achenbach & Rescorla, 2001). This questionnaire assesses child behavior problems. For 118 concrete behavioral, emotional, and social problems, foster mothers were asked to indicate how often they had occurred on a 3-point scale. The results of the questionnaire form a total problem score, an internalizing and externalizing score, and eight problem scale scores. We used the internalizing, externalizing, and total problem scores as (general) indices for internalizing, externalizing, and overall behavioral problems. The reliability and validity of the CBCL are viewed as established (Evers, van Vliet-Mulder, & Groot, 2000).

Nijmegen Parenting Situation Scale (Nijmeegse Vragenlijst voor de Opvoedingssituatie—NVOS; Wels & Robbroeckx, 1996). This questionnaire measures parenting stress. Foster mothers indicate on a 5-point scale how closely concrete statements relate to them. Four scales from the first part of the questionnaire were used in this study. These scales are viewed as the core components of parenting stress by the authors of the NVOS:

Coping ability (8 items— $α_{T_{0}}$ = .80; $α_{T_{1}}$ = .87; $α_{T_{2}}$ = .79) refers to the feeling of being able to cope with the parenting situation. For example, “Raising … requires a lot of my strength.”

Problem severity (7 items— $α_{T_{0}}$ = .72; $α_{T_{1}}$ = .78; $α_{T_{2}}$ = .75) refers to the severity of the problems as experienced by foster mothers. For example, “I’m glad when … is out for some time (e.g., at school, with friends, playing outside).”

Viewing parenting as a burden (7 items— $α_{T_{0}}$ = .81; $α_{T_{1}}$ = .88; $α_{T_{2}}$ = .88) refers to the extent to which parenting this specific child is experienced as a burden. For example, “Raising … is a real burden for me.”

Wishing for changes in the parenting situation (6 items— $α_{T_{0}}$ = .79; $α_{T_{1}}$ α_T1 = .86; $α_{T_{2}}$ = .79) refers to the extent to which foster mothers desire the parenting situation to change. For example, “Things should go really differently between me and ….…”

The reliability and validity of the NVOS are rated as satisfactory (Evers et al., 2000).

Ghent Parental Behavior Scale (Van Leeuwen & Vermulst, 2004). This questionnaire measures specific parenting practices. Foster mothers indicate on a 5-point scale how often they exhibit concrete parenting behaviors. Only the following subscales with sufficient reliability (on average α ≥ .65 at three measurement points) were used for further analyses:

Positive parenting (11 items— $α_{T_{0}}$ = .91; $α_{T_{1}}$ = .86; $α_{T_{2}}$ = .90): a higher score refers to more positive parent–child interactions. For example, “I take time to listen to my child, when he/she wants to tell me something.”

Monitoring (5 items - $α_{T_{0}}$ = .58; $α_{T_{1}}$ = .72; $α_{T_{2}}$ = .75): supervision of the child’s behavior. For example, “I check on the friends my child spends time with.”

Stimulating autonomy (3 items - $α_{T_{0}}$ = .55; $α_{T_{1}}$ = .65; $α_{T_{2}}$ = .75): a higher score means more encouragement of autonomous behavior. For example, “I teach my child to take decisions on his/her own.”

Rules (6 items - $α_{T_{0}}$ = .74; $α_{T_{1}}$ = .71; $α_{T_{2}}$ = .71): a higher score means more rules are taught. For example, “I teach my child to adapt to our family habits.”

Discipline (6 items - $α_{T_{0}}$ = .79; $α_{T_{1}}$ = .78; $α_{T_{2}}$ = .83): a high score might indicate that parents deliver punishment so often that it limits the possibilities of positive interactions. For example, “When my child misbehaves, I give him/her a chore as punishment.”

Inconsistent punishment (3 items - $α_{T_{0}}$ = .62; $α_{T_{1}}$ = .69; $α_{T_{2}}$ = .72): the scale refers to threatening to punish without executing or repeatedly announcing a punishment before carrying it out. For example, “I threaten punishment, but in the end I don’t carry it out.”

Harsh punishment (4 items - $α_{T_{0}}$ = .71; $α_{T_{1}}$ = .75; $α_{T_{2}}$ = .82): this refers to corporal punishment or verbal reproach. For example, “I slap my child when he/she has done something wrong.”

The size of the supportive network was measured by asking the foster mother to write down the people on whom she could rely for emotional support, practical support, and information/advice.

Statistical Methods

Intention-to-treat (ITT) was used to analyze the outcomes. ITT means that all randomized participants were included in the analyses, regardless of whether they received the intervention they were allocated (Altman, 2009; Nich & Carroll, 2002). Despite efforts to minimize the occurrence of missing data, some participants had missing data at one point in time, either due to breakdown of the placement or due to other reasons (see Figure 1). In the intervention group, two T₂ questionnaires were missing due to breakdown of the placement, in the control group, two T₀ questionnaires, five T₁ questionnaires (four due to breakdown) and six T₂ questionnaires (five due to breakdown) were missing. In cases of breakdown of the foster-care placement, one can assume that data are not missing at random. In those cases, Last Observation Carried Forward was used to impute missing data. This assumes that no improvements had been made, which, according to O’Neill and Temple (2012), is a more conservative method than running a multiple imputation algorithm. In case of ongoing placements, the missing data were considered as missing completely at random (Little’s missing completely at random test: χ²(680) = 605.88, p = .98). Missing data were then imputed using the multiple imputation algorithm of the Statistical Package for the Social Sciences (SPSS; IBM Corporation, 2012), which is preferred over traditional single imputation methods (Acock, 2005). Missing baseline data were imputed by including the minimization factors as well as all baseline and postintervention data in the algorithm. Missing postintervention data and follow-up data were imputed by including the minimization factors as well as all baseline, postintervention and follow-up data in the algorithm.

Figure 1.

Flow diagram of participants through each stage of the randomized clinical trial.

Analysis of covariance (ANCOVA) was used to analyze short-term effects (i.e., group differences between the experimental and control group in the development from T₀ to T₁) and follow-up effects (i.e., development from T₀ to T₂) for the outcome measures. In every ANCOVA, the baseline data of the outcome variable (T₀) were included as covariate. Besides statistical significance, Cohen’s d effect sizes were calculated based on the most commonly used pooled standard deviation (Rosnow & Rosenthal, 1996). Cohen (1988) suggested that effect sizes should be interpreted as small when above .20, medium when above .50, and large when above .80.

Results

TAU

In another publication we have compared the TAU that was offered in the control condition of the present study with TAU provided to a similar group of foster parents that did not participate in the study (Van Holen et al., 2015). The families in the TAU group of the present study were carefully matched with a pool of other foster families that did not take part in the present study based on the following variables: externalizing problem behavior, type of foster family (kinship vs. nonkinship), duration of placement, age and gender of the foster child. The results showed that being part of the control condition of the present study was associated with a higher counseling frequency from the foster-care worker, as well as with more referrals and utilization of external mental health services for the foster parents (15.8% vs. 5.3%) and the foster children (35.1% vs. 14%). The considerably higher frequency of referrals and utilization of external services shows that the TAU control condition in the present study was actually a “highly enhanced TAU.”

Short-Term Outcomes

As shown in Table 3, differences regarding foster child’s behavior and parenting stress did not reach statistical significance. The effect sizes indicated small effects on the total problem scale (d = .36) and on three out of four parenting stress scales, namely “experienced coping ability” (d = .22), “experiencing problem severity” (d = .35), and “desiring changes in the parenting situation” (d = .23).

Table 3.

Short-Term and Follow-Up ANCOVA Results and Effect Sizes.

Measure	Experimental Group			Control Group			ANCOVA T ₁–T ₀		ANCOVA T ₂–T ₀
	M ₀ (SD ₀)	M ₁ (SD ₁)	M ₂ (SD ₂)	M ₀ (SD ₀)	M ₁ (SD ₁)	M ₂ (SD ₂)	p	Effect Size (95% CI)	p	Effect Size (95% CI)
Foster child
Internalizing problems	15.71 (8.51)	11.29 (5.87)	9.94 (6.36)	12.81 (8.02)	9.75 (6.24)	8.83 (6.53)	.78	0.17 [−0.34, 0.663]	.84	0.21 [−0.28, 0.72]
Externalizing problems	23.06 (7.39)	18.23 (9.79)	16.94 (7.64)	24.32 (8.20)	20.48 (9.51)	20.08 (10.09)	.51	0.13 [−0.37, 0.63)	.21	0.24 [−0.26, 0.74]
Total problems	66.77 (20.03)	50.9 (23.27)	48.13 (22.28)	62.61 (20.36)	53.90 (19.50)	50.81 (21.20)	.19	0.36 [−0.15, 0.86]	.22	0.33 [−0.16, 0.84]
Foster mother
Coping ability	20.69 (3.87)	18.72 (5.72)	17.78 (5.61)	18.67 (5.39)	17.75 (5.14)	18.28 (4.89)	.66	0.22 [−0.28, 0.72]	.20	0.53 [0.03, 1.04]
Problem severity	19.65 (4.60)	17.93 (4.71)	16.76 (4.73)	17.91 (4.42)	17.78 (4.84)	18.23 (4.70)	.23	0.35 [−0.15, 0.85]	.01	0.70 [0.20, 1.22]
Desiring change in parenting situation	16.93 (3.81)	14.55 (4.90)	14.55 (4.86)	15.32 (3.77)	13.81 (3.98)	14.10 (3.84)	.67	0.23 [−0.27, 0.73]	.69	0.30 [−0.20, 0.81]
Parenting is burden	20.91 (3.95)	20.56 (5.43)	19.73 (5.89)	21.32 (4.84)	20.43 (5.40)	20.60 (4.68)	.58	−0.12 [−0.38, 0.62]	.60	0.10 [−0.40, 0.60]
Positive parenting	40.78 (7.82)	43.92 (6.64)	41.97 (7.49)	44.20 (5.88)	44.98 (5.21)	45.06 (5.18)	.11	0.34 [−0.16, 0.84]	.40	0.05 [−0.45, 0.55]
Monitoring	18.03 (4.32)	19.36 (3.59)	18.33 (4.05)	15.88 (4.89)	15.89 (4.44)	15.59 (4.81)	.005	0.29 [−0.22, 0.79]	.12	0.13 [−0.37, 0.63]
Stimulating autonomy	10.53 (1.96)	10.84 (2.10)	11.00 (2.21)	11.45 (1.70)	11.99 (.36)	12.01 (1.65)	.13	−0.12 [−0.37, 0.62]	.22	−0.05 [−0.45, 0.55]
Rules	26.10 (3.28)	27.19 (3.28)	26.35 (2.74)	27.27 (2.13)	27.54 (2.20)	27.05 (2.86)	.99	0.30 [−0.21, 0.80]	.92	0.17 [−0.33, 0.67]
Discipline	17.94 (4.04)	17.90 (4.19)	17.10 (4.03)	18.46 (4.80)	18.52 (4.14)	18.95 (4.36)	.69	0.02 [−0.48, 0.52]	.07	0.29 [−0.20, 0.80]
Inconsistent punishment	9.35 (2.69)	8.98 (2.35)	8.35 (2.51)	8.70 (2.16)	9.08 (2.93)	8.95 (2.73)	.33	0.30 [−0.20, 0.81]	.03	0.51 [0.01, 1.02]
Harsh punishment	4.16 (.37)	4.13 (.43)	4.19 (.60)	4.47 (1.25)	4.51 (1.50)	4.49 (1.63)	.70	0.07 [−0.42, 0.57]	.80	−0.01 [−0.45, 0.46]
Number of persons for emotional support	4.68 (3.74)	5.82 (4.59)	6.16 (5.27)	4.81 (2.93)	4.86 (3.04)	3.81 (2.29)	.30	0.33 [−0.18, 0.82]	.03	0.72 [0.22, 1.25]
Number of persons for practical support	3.50 (3.05)	4.15 (4.00)	3.83 (2.66)	4.00 (3.60)	3.45 (2.31)	3.15 (2.35)	.20	0.36 [−0.14, 0.86]	.09	0.35 [−0.15, 0.85]
Number of persons for information/advice	2.58 (1.94)	3.81 (4.66)	3.04 (2.42)	3.95 (3.28)	2.40 (1.47)	2.50 (1.34)	.01	1.05 [0.50, 1.56]	.30	0.72 [0.19, 1.22]

Note. ANCOVA = analysis of covariance; CI = confidence interval.

With regard to parenting practices, one significant difference was found: experimental foster mothers scored significantly higher on monitoring (M = 19.36, F(1, 59) = 9.08, p = .005, d = .29) than control foster mothers (M = 15.89). The effect size was small. Further, Cohen’s effect size indicated small effects on three out of six examined parenting practices, namely on “positive parenting” (d = .34), “‘rules” (d = .30) and “inconsistent punishment” (d = .30).

Concerning the social support measures, one significant difference was found: foster mothers from the experimental group could rely on more people for information and advice (M = 3.81, t(44) = 2.54, p = .01, d = 1.05) than foster mothers from the control group (M = 2.4). The effect size indicated a large effect. With regard to the number of people the foster mother could rely on for emotional and practical support no significant differences were found. The effect sizes, however, indicated small effects in favor of the intervention group (respectively d = .33 and d = .36).

Outcomes at Follow-Up

Regarding the foster child behavior, no statistically significant differences between both groups were found. Cohen’s effect sizes indicated small effects in favor of the intervention group with regard to internalizing, externalizing, and total problem behavior (d ranging from .21 to .33).

Concerning parenting stress, one significant effect was found: experimental foster mothers experienced significant less problem severity (M = 16.76, F(1, 59) = 7.18, p = .01, d = .70) than control foster mothers (M = 18.23). The effect size was medium. Furthermore, effect sizes indicated medium effects for “coping ability” (d = .53), and small effects for “desiring changes in the parenting situation” (d = .30). For one parenting stress scale (“experiencing parenting as a burden”), no differences between the experimental and control group were found.

Concerning parental practices one significant difference was found: foster mothers of the control group carried out punishments in a more inconsistent way (M = 8.95, F(1, 59) = 4.93, p = .03, d = .51) than experimental foster mothers (M = 8.35). The effect size was medium. For the discipline subscale, a marginally significant effect was found: foster mothers from the experimental group used less discipline practices (M = 17.10, F(1, 59) = 3.43, p = .07, d = .29) than control foster mothers (M = 18.95). The effect size was small. For five parenting practice scales (“positive parenting,” “monitoring,” “stimulating autonomy,” “rules,” and “harsh punishment”), no differences between the experimental and control group were found.

Regarding social support, the number of persons the foster mother could rely on for emotional support differed significantly between both groups, and the number of persons foster mothers could turn to for practical support differed in a marginally significant way. Experimental foster mothers could rely on more people for emotional support, M = 6.16, t(41.46) = −2.20, p = .03, d = .72, than control foster mothers (M = 3.81), and experimental foster mothers could turn to more people for practical support, M = 3.83, F(1, 40) = 3.00, p = .09, d = .35, than control foster mothers (M = 3.15). Effect sizes indicated respectively medium and small effects. Concerning the number of people, foster mothers could turn to for information effects size indicated a medium effect (d = .72).

Discussion and Applications to Practice

We have described an NVR intervention adapted to the foster-care context. The manualized support program including 10 individual sessions aimed at helping foster parents to prevent escalation, engage social support, make use of relation gestures, and implement active NVR to problem behaviors.

No foster parents in the experimental group dropped out prematurely. This finding is in line with the low dropout rates in other studies on NVR (Lavi-Levavi et al., 2013; Ollefs et al., 2009; Weinblatt & Omer, 2008) and contrasts with the high dropout rates that affect many other parent training programs (Friars & Mellor, 2009).

Regarding the foster child’s behavior problems, although comparison did not reach significance, small effects favoring the NVR group were reported on the total problems scale at treatment completion and on internalizing, externalizing, and total problems at the follow-up. These small effects on behavioral problems are comparable with the effects achieved by other training programs for parents (Lundahl, Risser, & Lovejoy, 2006) and are in line with the results of the evidence-based programs for foster parents (Leve et al., 2012). Those relatively modest gains might be explained by three factors: (1) The TAU condition was actually an “enhanced TAU”, as foster parents and children in this group received up to 3 times more mental health services than they normally would. It is notorious in psychotherapy research that large effects are hard to get when the comparison is made with an active treatment group (Freedland, Mohr, Davidson, & Schwartz, 2011); (2) The children in the study were older than in the majority of studies regarding foster placements (M > 12 years). In a recent review, Leve et al. (2012) identified eight effective programs for foster parents, and only two programs include children over the age of 12. It is known that improvements in difficult behaviors are harder to achieve with older children than with younger children (Kazdin, 1997; Lundahl et al., 2006); and (3) The population in this study consisted mainly of kinship foster parents (60%) and was relatively uneducated (60% without higher education). A considerable proportion were single (24%) elderly grandparents (40.3%), who are usually economically disadvantaged and in poorer health (Bronselaer et al., 2011). Disadvantaged or underprivileged families (cf. single, low income, poor health, low socioeconomic status, etc.) usually have more difficulty in gaining from training (Lundahl et al., 2006; Reyno & McGrath, 2006).

Regarding parenting stress, small positive effects were found at treatment conclusion in 3 out of 4 parenting stress scales, increasing to medium-sized effects at follow-up. Foster mothers in the NVR group felt more able to cope, reported less severe problems, and were more satisfied with the parenting situation. No differences were found regarding “parenting is a burden.” This might be logical. Although the intervention may have contributed to a problem reduction, only small effects on problem behavior were found. The upbringing of a foster child might therefore remain burdensome for foster parents.

Concerning parenting practices, in short term, foster mothers significantly monitored the foster child more, although the effect size was small. Furthermore, small positive effects were found on positive parenting, teaching rules, and inconsistent discipline. In the long term, experimental foster mothers significantly disciplined less inconsistent. The effect size was medium. Furthermore, a marginally positive small effect was found regarding discipline. These results suggest that foster mothers did efforts in order to change their parenting behavior. Nevertheless, some encouraging short-term results disappeared in the long term. This is surprising especially with regard to positive parenting and monitoring. In fact, NVR strongly focuses on monitoring and on encouraging positive interactions between foster parents and foster children through the use of relation gestures. We see two explanations: (1) the parenting situation improved, with the result that foster mothers paid less attention to these issues and (2) NVR focuses on parental attitudes and less on concrete behaviors, whereas the questionnaire measures concrete behaviors. More concrete translation of parental attitudes in well-defined behaviors and the teaching of everyday practical parenting skills might need more attention and could contribute to a higher effectiveness.

Regarding experienced support, there were improvements both at treatment completion and at follow-up. Foster mothers received more emotional support and could count on more people to provide them with practical support and with information and advice. The increase of support is in line with Weinblatt and Omer (2008) and Ollefs, von Schlippe, Omer, and Kriz (2009). The enduring improvement in an experienced support in the NVR group might be especially meaningful, as a broad social network is considered as an important protective factor (Orme et al., 2004; Redding, Fried, & Britner, 2000) and is associated with lower stress levels and more resilience in foster parents (Fisher & Stoolmiller, 2008).

This study has some limitations. Firstly, only self-report measures were used which increases the vulnerability of the study to over- and underestimation of problems and gains (Rosenfeld et al., 1997; Van Oijen, 2010). A second limitation is that, for the reasons we have described, assessment was restricted to foster mothers. A third limitation is that blinding was not possible. Knowledge of treatment allocation may have influenced the subjects’ assessment of improvement (Hotopf, 2002). A fourth limitation concerns the short follow-up period. A longer follow-up period was not possible on ethical grounds, as the control group should become entitled to NVR training after 3 months. A fifth limitation is the relatively low statistical power. A sensitivity test using G*Power 3.1 (Faul, Erdfelder, Lang, & Buchner, 2007) showed that with a total sample of 62 cases, an α of .05, and a power of .80, only effect sizes above .72 could be detected. This reduced the study’s ability to show significance not only with small but also with medium-sized effects. The fact that the intervention was carried out in a real-life setting, and that a home visit format enabled to offer the intervention to difficult to reach people, is in favor of the applicability to other foster-care situations with children showing externalizing problem behavior. Indeed, demographic characteristics of the participants suggest great diversity. On the other hand, a significant group of foster parents fell out because the screening questionnaires were not filled in or because they refused to participate. As a consequence, a selection bias cannot be excluded, which in turn requires caution regarding the generalizability. Replications of this foster parent intervention using a multitrait-multimethod matrix would be appropriate. An examination of the specific ingredients of NVR that might be especially important and effective should be undertaken (Van Holen, Lampo, & Vanderfaeillie, 2011). In addition, the relationship between results and population characteristics (e.g., age of foster child, kinship vs. nonkinship foster parents) should be examined.

In spite of the limitations, we feel the study shows that NVR is a highly acceptable approach for foster parents, which leads to an increase in experienced support and to some positive changes in parenting stress and parenting practices. Future studies are needed to show how far those changes can also lead to improvements in the behavior problems of foster children.

Footnotes

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Achenbach

T. M.

Rescorla

L. A.

(2001). Manual for the Aseba school-age forms & profiles: Child behaviour checklist for ages 6-18. Burlington, VT: ASEBA.

Acock

A. C.

(2005). Working with missing values. Journal of Marriage and Family, 67, 1012–1028.

Altman

D. G.

(2009). Missing outcomes in randomized trials: Addressing the dilemma. Open Medicine, 3, 51–53.

Bronselaer

Vandezande

Verreth

(2011). Vlaamse pleegzorgers in beeld. Brussel, Belgium: Koning Bouwdewijnstichting.

Chamberlain

Price

J. M.

Reid

J. B.

Landsverk

Fisher

P. A.

Stoolmiller

(2006). Who disrupts from placement in foster and kinship care? Child Abuse and Neglect, 30, 409–424.

Cohen

(1988). Statistical power of the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum Associates.

Evers

van Vliet-Mulder

J. C.

Groot

C. J.

(2000). Documentatie van tests en testresearch in Nederland. Deel 1 Testbeschrijvingen. Assen, the Netherlands: Van Gorcum.

Faul

Erdfelder

Lang

A.-G.

Buchner

(2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175–191.

Fernandez

(2009). Children’s wellbeing in care: Evidence from a longitudinal study of outcomes. Children and Youth Services Review, 31, 1092–1100.

10.

Fernandez

Barth

(Eds.). (2011). How does foster care work? International evidence and outcomes. London, England: Jessica Kingsley.

11.

Fisher

P. A.

Gunnar

Chamberlain

Reid

(2000). Preventive intervention for maltreated preschool children: Impact on children’s behaviour, neuroendocrine activity, and foster parent functioning. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 1356–1364.

12.

Fisher

P. A.

Stoolmiller

(2008). Intervention effects on foster parent stress: Associations with child cortisol levels. Development and Psychopathology, 20, 1003–1021.

13.

Freedland

K. E.

Mohr

D. C.

Davidson

K. W.

Schwartz

J. E.

(2011). Usual and unusual care: Existing practice control groups in randomized controlled trials of behavioural interventions. Psychosomatic Medicine, 73, 323–335.

14.

Friars

Mellor

(2009). Drop-out from parenting training programmes: A retrospective study. Journal of Child and Adolescent Mental Health, 21, 29–38.

15.

Havlicek

J. R.

Garcia

A. R.

Smith

D. C.

(2013). Mental health and substance use disorders among foster youth transitioning to adulthood: Past research and future directions. Children and Youth Services Review, 35, 194–203.

16.

Herdewyn

(2011). Pleegzorgbegeleiding op de cover: Ontwikkeling van een registratieformulier. Niet gepubliceerde masterthesis, Vrije Universiteit Brussel, Brussel.

17.

Holtan

Ronning

Handegard

Sourander

(2005). A comparison of mental health problems in kinship and nonkinship foster care. European Child and Adolescent Psychiatry, 14, 200–207.

18.

Hotopf

(2002). The pragmatic randomised controlled trial. Advances in Psychiatric Treatment, 8, 326–333.

19.

IBM Corporation. (2012). IBM SPSS Statistics for Windows, Version 21.0. Armonk, NY: IBM Corporation.

20.

Kazdin

A. E.

(1997). Practitioner review: Psychosocial treatments for conduct disorder in children. Journal of Child Psychology and Psychiatry, 2, 161–182.

21.

Koeter

M. W. J.

Ormel

(1991). General health questionnaire, Nederlandse bewerking: Handleiding. Lisse, the Netherlands: Swets Test Services.

22.

Lavi-Levavi

Shachar

Omer

(2013). Training in nonviolent resistance for parents of violent children: differences between fathers and mothers. Journal of Systemic Therapies, 32, 79–93.

23.

Leve

L. D.

Harold

G. T.

Chamberlain

Landsverk

J. A.

Fisher

P. A.

Vostanis

(2012). Practitioner review: Children in foster care—Vulnerabilities and evidence-based interventions that promote resilience processes. Journal of Child Psychology and Psychiatry, 53, 1197–1211.

24.

Linares

L. O.

Montalto

Rosbruch

(2006). Discipline practices among biological and foster parents. Child Maltreatment, 11, 157–167.

25.

Lundahl

Risser

H. J.

Lovejoy

M. C.

(2006). A meta-analysis of parent training: Moderators and follow-up effects. Clinical Psychology Review, 1, 86–104.

26.

Maaskant

(2010). Pleegkinderen in de CGGZ. In van den Bergh

Weterings

(Eds.), Pleegzorg in perspectief. Ontwikkelingen in theorie en praktijk (pp. 271–284). Assen, the Netherlands: Van Gorcum.

27.

Maughan

D. R.

Christiansen

Jenson

W. R.

Olympia

Clark

(2005). Behavioral parent training as a treatment for externalizing behaviors and disruptive behavior disorders: A meta-analysis. School Psychology Review, 34, 267–286.

28.

McEntegart

D. J.

(2003). The pursuit of balance using stratified and dynamic randomization techniques: An overview. Drug Information Journal, 37, 293–308.

29.

Nich

Carroll

K. M.

(2002). Intention-to-treat meets missing data: Implications of alternate strategies for analyzing clinical trials data. Drug and Alcohol Dependence, 68, 121–130.

30.

Nilsen

(2007). Fostering futures: A preventive intervention program for school-age children in foster care. Clinical Child Psychology and Psychiatry, 12, 45–63.

31.

Ollefs

von Schlippe

Omer

Kriz

(2009). Jugendliche mit externalem problemverhalten. Effekte von elterncoaching. Familiendynamik, 34, 256–265.

32.

Omer

(2004). Nonviolent resistance. A new approach to violent and self-destructive children. New York, NY: Cambridge University Press.

33.

Omer

(2011). The new authority: Family, school and community. New York: Cambridge University Press.

34.

Omer

Lebowitz

(2016). Nonviolent resistance: Helping caregivers reduce problematic behaviors in children and adolescents. Journal of Marital and Family Therapy, doi:10.111/jmft.12168

35.

Omer

Schorr-Sapir

Weinblatt

(2008). Non-violent resistance and violence against siblings. Journal of Family Therapy, 30/4, 450–464.

36.

Omer

Steinmetz

Carthy

von Schlippe

(2013). The anchoring function: Parental authority and the parent-child bond. Family Process, 52, 193–206.

37.

O’Neill

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, 550–554.

38.

Oosterman

Schuengel

Slot

N. W.

Bullens

R. A. R.

Doreleijers

T. A. H.

(2007). Disruptions in foster care: A review and meta-analysis. Children and Youth Services Review, 29, 53–76.

39.

Orme

Buehler

McSurdy

Rhodes

Cox

Patterson

(2004). Parental and familial characteristics of family foster care applicants. Children and Youth Services Review, 26, 307–329.

40.

Patterson

G. R.

Dishion

T. J.

Bank

(1984). Family interaction: A process model of deviancy training. Aggressive Behaviour, 10, 253–267.

41.

Redding

Fried

S. C.

Britner

(2000). Predictors of placement outcomes in treatment foster care: Implications for foster parent selection and service delivery. Journal of Child and Family Studies, 9, 425–447.

42.

Reyno

S. M.

McGrath

P. J.

(2006). Predictors of parent training efficacy for child externalizing behaviour problems—A meta-analytic review. Journal of Child Psychology and Psychiatry, 47, 99–111.

43.

Rosenfeld

A. A.

Pilowsky

D. J.

Fine

Thorpe

Fein

Simms

M. D.

… Nickman

(1997). Foster care: An update. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 448–457.

44.

Rosnow

R. L.

Rosenthal

(1996). Computing contrasts, effect sizes, and counternulls on other people’s published data: General procedures for research consumers. Psychological Methods, 1, 331–340.

45.

Saghaei

(2011). An overview of randomization and minimization programs for randomized clinical trials. Journal of Medical Signals and Sensors, 1, 55–61.

46.

Sawyer

M. G.

Carbone

J. A.

Searle

A. K.

Robinson

(2007). The mental health and wellbeing of children and adolescents in home-based foster care. Medical Journal of Australia, 186, 181–184.

47.

Scott

N. W.

McPherson

G. C.

Ramsay

C. R.

Campbell

M. K.

(2002). The method of minimization for allocation to clinical trials. A Review. Controlled Clinical Trials, 23, 662–674.

48.

Sprangers

(2009). Eindrapport labofunctie “Project pleegzorg-palet”: Een intersectoraal, gedifferentieerd en flexibel pleegzorgaanbod voor minderjarigen en hun gezin. Brussel, Belgium: Agentschap Jongerenwelzijn.

49.

Taves

D. R.

(2010). The use of minimization in clinical trials. Contemporary Clinical Trials, 31, 180–184.

50.

Vanderfaeillie

(2011, 31 maart). Ondersteuningsbehoeften en tevredenheid en tevredenheid van Limburgse pleegouders die een kind uit de BJB opvangen. Presentatie van onderzoeksresultaten voor Pleegzorg Limburg, Hasselt.

51.

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, 34–42.

52.

Van Holen

Lampo

Vanderfaeillie

(Eds.). (2011). Applications and illustrations of non-violent resistance and the new authority in family therapy, education and psychiatric care. Brussels, Belgium: University Press.

53.

Van Holen

Vanderfaeillie

De Maeyer

Gypen

(2015). Does allocation to a control condition in a randomized controlled trial affect the routine care foster parents receive? Children and Youth Services Review, 49, 48–53.

54.

Van Holen

Vanderfaeillie

Omer

(2016). Adaptation and evaluation of a nonviolent resistance intervention for foster parents: A progress report. Journal of Marital and Family Therapy, 42, 256–271.

55.

Van Holen

Vanderfaeillie

Vanschoonlandt

(2013). Pleegouders versterken in opvoeden. Behandelprotocol voor trainers. Module Geweldloos Verzet. Brussel, Belgium: Vrije Universiteit Brussel.

56.

Van Leeuwen

K. G.

Vermulst

A. A.

(2004). Some psychometric properties of the Ghent parental behavior scale. European Journal of Psychological Assessment, 20, 283–298.

57.

Van Oijen

(2010). Resultaat van pleegzorgplaatsingen. Een onderzoek naar breakdown en de ontwikkeling van adolescente pleegkinderen bij langdurige pleegzorgplaatsingen. Groningen, the Netherlands: Rijksuniversiteit Groningen.

58.

Vanschoonlandt

Vanderfaeillie

Van Holen

De Maeyer

Robberechts

(2013). Externalizing problems in young foster children: Prevalence rates, predictors and service use. Children and Youth Services Review, 35, 716–724.

59.

Vanschoonlandt

Van Holen

Vanderfaeillie

(2012). Ontwikkeling en proefimplementatie van twee ondersteuningsprogramma’s voor pleegouders. Orthopedagodiek: Onderzoek en Praktijk, 51, 69–83.

60.

Verreth

(2009). Pleegzorg: wanneer? Deel 1: (rechts)vergelijkend onderzoek naar beleid en wetgeving in Vlaanderen en enkele Europese landen. Brussel, Belgium: Kenniscentrum Welzijn, Volksgezondheid en Gezin.

61.

Weinblatt

Omer

(2008). Nonviolent resistance: A treatment for parents of children with acute behaviour problems. Journal of Marital and Family Therapy, 34, 75–92.

62.

Wels

P. M. A.

Robbroeckx

L. M. H.

(1996). Nijmeegse vragenlijst voor de opvoedingssituatie: Handleiding. Lisse, the Netherlands: Swets & Zeitlinger.

63.

Whenan

Oxlad

Lushington

(2009). Factors associated with foster carer well-being, satisfaction and intention to continue providing out-of-home care. Children and Youth Services Review, 31, 752–760.

64.

Wilson

(2006). Can foster carers help children resolve their emotional and behavioural difficulties? Clinical Child Psychology and Psychiatry, 11, 495–511.