Family Empowerment (FAME): A feasibility trial of preventive multifamily groups for asylum seeker families in the Netherlands.

This study evaluated the feasibility of Family Empowerment (FAME), a preventive multifamily program for asylum seeker families in the Netherlands. FAME aims to reinforce the parent-child relationship, family functioning, and social support. We used an uncontrolled pre-test-post-test design, embedded in a mixed-methods approach. FAME was offered to 46 asylum seeker families, mostly originating from Eritrea, Armenia, or Syria. Twenty-seven parents gave consent to participate in this study. Program integrity and evaluations of participating parents and trainers were assessed. Family functioning and parental symptoms of depression and anxiety were measured pre- and post-FAME. Six participants completed all assessments. Most participants valued gathering with multiple families. Although FAME might coincide with decreases in anxiety and depression, the program had a limited impact on family functioning. Possibly, the aims of FAME did not align with some families' current needs. Lessons learned and recommendations to further improve interventions for refugee families are discussed.

INTRODUCTION

In January 2020, over 27,000 refugees and asylum seekers resided in reception centers, asylum centers, and family locations in the Netherlands. Approximately a quarter of this group were minors under the age of 18, the majority of whom arrived in the Netherlands with at least one parent (Centraal Orgaan opvang asielzoekers, 2020). The lives of these families can be put under considerable pressure as a result of adverse past events in combination with continuous stressors linked to the post‐migration environment, such as insecurities concerning status, family reunification, social isolation, finances, and finding housing (Fazel et al., 2012; Utržan & Wieling, 2020).

A refugee is defined by the United Nations High Commissioner for Refugees (UNHCR) as “any person who owing to well‐founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his/her nationality and is unable or, owing to such fear, is unwilling to avail him/herself of the protection of that country; or who, not having a nationality and being outside the country of his/her former habitual residence as a result of such events, is unable or, owing to such fear, is unwilling to return to it” (p. 121), whereas an asylum seeker is defined as “an individual who is seeking international protection. In countries with individualized procedures, an asylum‐seeker is someone whose claim has not yet been finally decided on by the country in which the claim is submitted” (p. 118) (UNHCR, 2013). In 2019, approximately 30,000 persons applied for asylum in the Netherlands, most of whom originated from Syria (18%), Eritrea (8%), and Nigeria (7%) (Immigration & Naturalisation Service, 2019). In the Netherlands, asylum seeker families are located in asylum centers until their asylum application is granted or rejected. If the application is rejected, families are placed in a family location, where they are prepared for deportation.

The adjustment of children after adverse events can be influenced by the parent–child relationship (El‐Khani et al., 2020; Fazel & Betancourt, 2018). Refugee and asylum seeker parents are at risk of developing stress‐related symptoms, and mental health problems can undermine their parenting skills, the parent–child relationship, and family functioning (Sangalang et al., 2017; Van Ee et al., 2016). Moreover, these parents are often forced to leave their social support networks behind, causing feelings of loneliness and isolation (Stewart et al., 2015). A well‐functioning parent–child relationship, family system, and social support system can play an important protective role in families’ adjustment after hardship (Betancourt & Khan, 2008; El‐Khani et al., 2020; Fazel & Betancourt, 2018).

Family interventions can be an important means to prevent the negative impact of resettlement‐related adversity. However, studies in this area have been scarce (Slobodin & de Jong, 2015). There is some evidence that family‐focused interventions for families exposed to trauma and displacement can have a positive impact on mental health utilization, social support, family hardiness, parental involvement, children’s problem behavior, emotion regulation, and family functioning (Asen, 2002; Ballard et al., 2017; El‐Khani et al., 2020; Weine et al., 2003, 2008). However, the scarcity of studies evaluating family‐focused interventions limits the possibility to draw clear conclusions on their effectiveness (Slobodin & de Jong, 2015). The studies that have been conducted emphasized the importance of cultural and contextual adaptations when offering programs to families in the context of trauma and resettlement. For instance, by offering program by facilitators who are refugees themselves (Weine et al., 2008) or by using examples relevant to the context of the participating families (El‐Khani et al., 2020).

To prevent the occurrence or deterioration of mental health problems and to reinforce the parent–child relationship, family functioning, and social support for families, the secondary preventive multifamily program Family Empowerment (FAME) was developed (Mooren & Bala, 2016). FAME was developed specifically for families residing in asylum centers and living locations in the Netherlands.

The development of FAME is based on 10 years of experience with offering multifamily groups for refugee families at a psychotrauma center. Experiences with offering FAME at an asylum center have been used to strengthen the manual (Mooren & Bala, 2016). The development of the FAME manual has been in line with the cycle for developing and evaluating preventive programs for refugee families as proposed by Weine (Van Es et al., 2019; Weine, 2011).

During FAME, multiple families facing similar difficulties are brought together, allowing families to share experiences, to offer and receive mutual support and feedback, and to learn about different perspectives. FAME aims to strengthen parental competence and resilience. A key concept is mentalization, the ability to reflect upon and understand the state of mind of yourself and the other (Allen et al., 2008; Bateman & Fonagy, 2013). Although opinions about the parent–child relationship, parenting styles, and family functioning are often culturally bound, several universal ideas concerning the importance of sensitivity and responsiveness exist. For example, being attentive and responsive to the needs of a child is considered as beneficial to children worldwide (Mooren & Bala, 2016; Rohner, 2004; Van IJzendoorn & Sagi‐Schwartz, 2008). FAME aims to address such universal concepts in a culturally sensitive way.

During FAME, families gather for approximately eight weekly sessions. Each session has a similar structure: an energizing activity, activities focused on the main theme of that session, and closure, for example by reflecting on the session and/or practicing with relaxation exercises. The program protocol is described in detail elsewhere (Mooren & Bala, 2016; Van Es et al., 2019).

Deviations from the protocol often occur when a intervention program is applied in a naturalistic setting (Kösters et al., 2017). Program integrity, the extent to which the program is implemented as intended, can affect the treatment outcome (Kösters et al., 2017; Carroll et al., 2007). The current study aims to evaluate the feasibility of FAME when delivered in a naturalistic setting by studying (a) four dimensions of program integrity (adherence to the intervention, exposure to the intervention, quality of delivery, and participant responsiveness); (b) evaluations of FAME by participants and trainers; and (c) whether FAME coincided with decreases in symptoms of anxiety and depression, and improvements in family functioning. This pilot intervention study is an important step in the development of FAME before conducting trials at other locations and in other contexts (Weine, 2011).

METHODS

Design

The current study is a mixed‐methods pre‐test–post‐test feasibility trial. A self‐constructed program integrity list, semi‐structured individual and group interviews, a focus group, and standardized questionnaires were used. The study protocol has been published in a peer‐reviewed journal (Van Es et al., 2019). Ethical approval was obtained from the Medical Research Ethics Committee of of the Leiden UniversityMedical Centre (P17.268).

Participants

Trainers offered FAME at one Dutch asylum center and two family locations between November 2018 and July 2019. Per location, two treatment groups were started (six groups in total). Families were excluded from participation in FAME if a family member was not able to function in a group or was unlikely to benefit from participation, for example, as a result of severe psychiatric illness (e.g., psychosis). No one met this exclusion criterion. All parents who took part in at least one session of FAME were invited to participate in the study.

Procedure

Asylum centers and family locations were selected based on convenience sampling. Families with a similar cultural background and language were invited for an introduction session of FAME through flyers. In addition, researchers and/or local professionals visited families to offer information on FAME and to invite them to an introduction session. During this session, trainers explained several aspects of FAME, such as the frequency and the structure of sessions and mutual expectations of trainers and families. An overview of themes and aims for each session of FAME can be found in the study protocol (Van Es et al., 2019). Finally, researchers offered verbal and written information on the study.

Parents who were interested in taking part in the study were invited for an appointment at their home or in a quiet room for the completion of pre‐test measures (t 1, see Table 1). Written informed consent was obtained. After the final session of FAME, the final assessment took place (t 3). If it was not possible to make individual appointments, researchers handed out questionnaires to the group of families and conducted semi‐structured group interviews. Interviews were audio‐taped and transcribed verbatim or minutes were taken. During each assessment, a researcher and a (telephone) interpreter were present.

TABLE 1

Assessment schedule of Family Empowerment (FAME) feasibility trial

		Time point
	Respondent	t 1	t 2	t 3
Demographics	Parent	X		X
Feasibility
Program integrity list	Researcher		X
Semi‐structured interview	Parent	X		X
Therapeutic alliance (CSRS)	Parent		X
Focus group	Trainer			X
Family functioning & distress
Family functioning (SCORE‐15)	Parent	X		X
Depression and anxiety (PHQ‐4)	Parent	X		X

t 1: Pre‐FAME, t 2: During FAME, t 3: Post‐FAME.

Abbreviations: CSRS, child session rating scale; PHQ‐4, patient health questionnaire for depression and anxiety; SCORE‐15, systemic clinical outcome and routine evaluation.

Before each session of FAME, trainers visited the families to remind them of the session. The groups were open, meaning that when appropriate, a new family could join the group. Professional interpreters were present at each session. Program integrity lists were filled in by researchers or students during each session of FAME (t 2). All assessors were trained in administering the program integrity list.

After all FAME groups were completed, (co‐)trainers took part in a focus group. They gave verbal consent for the audio‐recording and use of the focus group data for the study. The audio‐recording was transcribed verbatim.

Several adaptations were made to the study protocol published earlier (Van Es et al., 2019). Firstly, the Emotional Availability Scale (EAS), an observational instrument assessing the quality of the parent–child relationship at baseline (Biringen, 2008), was not used. As parents were hesitant to give permission to record a videotape, it was decided not to conduct this measurement in order to reduce attrition. Secondly, the proposed measurements with children proved unworkable as children often did not attend the sessions. Thirdly, we noticed that, due to the language barrier, the adult version of the Session Rating Scale (SRS) and the Outcome Rating Scale took a lot of time and effort to fill in. Therefore, we decided to use the child version of the SRS (CSRS), which is a more visual measure for therapeutic alliance.

Team and team resources

Our team of trainers consisted of three lead trainers and two co‐trainers. At one location, local professionals followed a 1‐day FAME‐course provided by the leading trainers to become co‐trainers. One local professional and one researcher functioned as co‐trainers and, as such, delivered parts of FAME together with the lead trainers. Each session was prepared based on the FAME protocol and reflected on afterward. Trainers took part in several supervision sessions led by a trained therapist who had experience with multifamily therapy, during which challenges faced when implementing FAME and possible solutions were discussed.

Measures

The scheduling of assessments is displayed in Table 1.

Demographics

Researchers registered the following demographics of the participants: age, gender, country of origin, time since arrival in the Netherlands, and number of family members. Demographics were collected prior to FAME or post‐FAME if a participant did not participate in pre‐FAME measurements.

Program feasibility

Program feasibility was studied by using a program integrity list, the CSRS, interviews, a focus group, and standardized questionnaires. To evaluate program integrity, a list was constructed and completed by a researcher during each FAME session (t 2). It was based on four dimensions of program integrity (Carroll et al., 2007): (a) Adherence: assessors rated components of the protocol as executed (yes) or not executed (no), and the percentage of activities that was executed according to protocol was calculated; (b) Exposure: the presence of families, the total number of sessions, and the duration of each session; (c) Quality of delivery: during each session, assessors rated whether trainers applied therapeutic skills and competences specific to FAME (yes, sometimes, no, or not applicable); and (d) Participant responsiveness: during each session, assessors rated the approximate percentage of time that positive interactions (e.g., laughter) took place and rated active participation for each participant on a scale of 0 (not active) to 100 (very active).

To assess therapeutic alliance, the visual analogue Child version of the SRS was filled in by parents. The four single‐item subscales of the CSRS include: (a) respect and understanding; (b) relevance of goals and topics; (c) client–practitioner fit; and (d) overall alliance. The subscales scores range from 0 to 10, and the total score is calculated by adding up the subscale scores (range 0–40). A higher score indicates a better therapeutic alliance. A total score below 36, or a score below nine on a scale, indicates a perceived problem in the therapeutic alliance. The SRS has adequate validity and high feasibility (Duncan et al., 2003). Duncan et al., (2003) reported a Cronbach’s coefficient alpha of .88 for six administrations to 70 participants (N = 420) randomly selected from an outpatient mental health counseling agency.

Semi‐structured interviews were used to explore parents’ evaluations of FAME. Prior to FAME (t 1), expectations concerning FAME were explored. After FAME (t 3), participants’ evaluations of the usefulness and acceptability of the program were evaluated. Furthermore, we explored whether participants felt that FAME had impacted the parent–child relationship, social support, and coping strategies. Finally, a focus group interview with FAME (co‐)trainers took place after all groups were completed (t 3), during which program integrity, usefulness, and acceptability of FAME were discussed.

Family functioning and distress

To further understand the acceptability of FAME, we investigated whether FAME coincides with an improvement in family functioning and decreases in symptoms of anxiety and depression in participating parents. The Dutch version of the Systemic Clinical Outcome and Routine Evaluation (SCORE‐15) (Stratton et al., 2010) was used to measure family functioning. This self‐report questionnaire includes 15 items (e.g., “When one of us is upset they get looked after within the family”). Scores can be calculated on three dimensions (strength and adaptability, overwhelmed by difficulties, and disrupted communication); the total score offers an index of overall problems concerning family functioning. The questionnaire has good construct validity and is sensitive to change (Hamilton et al., 2015; Stratton et al., 2014).

To measure the severity of parental symptoms of depression and anxiety, the four‐item Patient Health Questionnaire for Depression and Anxiety (PHQ‐4) was used. Each item (e.g., “Feeling down, depressed, or hopeless”) is rated on a scale from 0 (not at all) to 3 (nearly every day). The total score ranges from 0 to 12, with categories of psychological distress being none (0–3), mild (3–5), moderate (6–8), and severe (9–12). In addition, anxiety and depression subscale scores can be calculated. On each subscale, a score of ≥3 is considered positive for screening purposes. The PHQ‐4 is a very brief and well‐validated questionnaire. It is considered suitable for a diverse group of refugees (Kliem et al., 2016). As the SCORE‐15 and PHQ‐4 are not translated to all languages, professional interpreters translated the questionnaire and answered any questions that arose.

Data analysis

Qualitative data

Minutes and transcriptions of the semi‐structured interviews and focus group were analyzed using MAXQDA 10 (VERBI). The data were analyzed using the General Inductive Approach (Thomas, 2003). Specific text fragments that were linked to the study aims were identified and labeled to create categories. Subsequently, overlap between and redundancy of the categories were reduced, and the remaining categories were described. This procedure was conducted independently by two researchers (CvE and TM) and subsequently discussed to increase the reliability of the analysis. The analysis resulted in outcome categories that represent the most important themes.

Quantitative data

Descriptive statistics of the four program integrity dimensions were calculated using SPSS 23 (IBM Statistics). To investigate whether participants showed significant improvements on the PHQ‐4 and SCORE‐15, the Reliable Change Index (RCI), as proposed by Jacobson and Truax (1992), was calculated. The following formula was used: . S diff is calculated using the intraclass correlation coefficient of the questionnaires and the standard deviation of the pre‐FAME scores. The intraclass correlation coefficient of the SCORE‐15 is .91 (Hamilton et al., 2015), and the intraclass correlation coefficient of the PHQ‐4 is .83 (Kim et al., 2021). A calculated RCI larger than |1.96| indicates a clinically reliable change with 95% certainty.

In line with the manual, we allowed one missing item on the SCORE‐15. If one item was missing, the subscales and total scores were calculated by multiplying the average score of the other items with the number of items on the subscale or total score. No missing items were allowed on the PHQ‐4. As not all participants filled in all questionnaires and questions, the number of respondents varies throughout the study. When percentages are reported, they apply to the number of participants answering that specific question or questionnaire.

RESULTS

Descriptive statistics

Six groups of FAME were conducted, including one group with participants from Armenia (Armenian speaking), two groups with participants from Eritrea and Ethiopia (Tigrinya/Amharic speaking), one group with participants from Syria and Iran (Arabic), and two groups with participants from North Africa (English, French, and Arabic speaking). The latter group was conducted by the trainer in English and French and translated to Arabic by an interpreter. In total, 46 families (42 mothers, 6 fathers, and 43 children) joined at least one session of FAME. Of the 48 parents, 27 gave consent for participating in the study. This study presents the data of the parents who gave consent. Reasons for not participating were diverse (see Table 2). For example, families consisting of a mother and a baby were not able to answer all questions concerning family functioning. Two groups did not participate in post‐FAME measurements because the groups did not continue after the introduction session. Local professionals did not give permission to interview the participants to explore why they did not want to continue. They explained that the participants experienced too much stress concerning their asylum procedure and living situation, and they did not want to add to their burden. Table 3 reports the demographics of the participants. Six participants completed pre‐ as well as post‐FAME measurements. The qualitative and quantitative data of these participants are described in detail in the “Evaluation by Participants” section.

TABLE 2

Reasons for not partaking in measurements

	N
Took part in ≥1 session(s) of FAME	48
Reasons for not partaking in study (n = 21)
Not being able to schedule an appointment	7
Group ended prematurely	6
Participant relocated	4
Only joined one session	3
Other	1
Total participation (written consent)	27
Reasons for not partaking at t 1 (n = 13)
Group did not partake in t 1 measures	8
Not being able to schedule an appointment	3
Late enrolment	2
Reasons for not partaking at t 3 (n = 8)
Group ended prematurely	6
Not being able to schedule an appointment	1
Participant relocated	1

FAME, Family Empowerment.

TABLE 3

Demographics of study participants (N = 27)

Variable*	n (%)	M (SD)	Range
Gender (n = 27)
Female	22 (81.5)	—	—
Male	5 (18.5)	—	—
Age (years) (n = 26)		34.6 (8.8)	21–56
Country of origin (n = 27)
Eritrea	10 (37.0)	—	–
Armenia	7 (25.9)	—	—
Syria	3 (11.1)	—	—
Other	5 (18.5)	—	—
Missing	2 (7.4)
Number of children in family (n = 19)		2.16 (1.4)	1–6
Time since arrival in Netherlands (months) (n = 18)		47.8 (38.0)	2–108
Living location (n = 27)
Asylum center	20 (74.1)	—	—
Family location	7 (25.9)	—	—
Partner (n = 25)
Yes	13 (52.0)	—	—
No	12 (48.0)	—	—

The number of participants for whom data were available are given in parentheses.

Table 4 reports the group means and standard deviations of the PHQ‐4 and SCORE‐15 at baseline. The anxiety subscale score suggests an increased risk of anxiety. The total score indicates that, on average, participants who completed the baseline assessment experienced moderate levels of psychological distress. The SCORE‐15 scores indicate few problems with “Strengths and adaptability” and “Disrupted communication” and some problems with “Overwhelmed by difficulties.” The total score indicates that families reported few problems concerning family functioning.

TABLE 4

Baseline measurements on depression, anxiety, and family functioning

	n	M (SD)
PHQ‐4 anxiety a	14	4.1 (2.1)
PHQ‐4 depression a	14	2.9 (2.3)
PHQ‐4 total	14	7.0 (4.2)
SCORE‐15 strengths and adaptability	10	10.6 (5.6)
SCORE‐15 overwhelmed by difficulties	9	14.3 (5.4)
SCORE‐15 disrupted communication	9	10.1 (5.0)
SCORE‐15 total	9	35.2 (15.3)

: PHQ‐4, patient health questionnaire for depression and anxiety; SCORE‐15, systemic clinical outcome and routine evaluation.

Cut‐off score (positive for screening purposes) = 3.

Program feasibility

Program integrity

Program integrity aspects are summarized in Tables 5 and 6. On average, trainers adhered to 60% of the protocol. Deviations from the protocol occurred, for example, as a result of not being able to conduct certain activities because no children attended the group or in order to address themes brought up by participants.

TABLE 5

Program integrity: the extent to which the program was implemented as intended

	M (SD)	Range
Adherence (%)	60.0 (22.8)	6.7–93.8
Exposure
Number of sessions	5.3 (3.0)	1–8
Duration of sessions (minutes)	91.6 (16.7)	60–120
Number of families present	5.0 (3.3)	1–14
Responsiveness
Positive interactions (0–100)	59.5 (20.4)	20–90
Active participation (0–100)	78.0 (7.9)	59–90
Child session rating scale (CSRS) a
Respect and understanding (0–10)	9.3 (1.2)	2.0–10.0
Relevance of the goals and topics (0–10)	9.5 (.5)	8.0–10.0
Client‐practitioner fit (0–10)	9.2 (1.1)	4.7–10.0
Overall alliance (0–10)	8.9 (1.5)	4.6–10.0
Total (0–40)	36.8 (3.4)	23.9–40.0

Filled in by 12 participants.

TABLE 6

Quality of delivery: the percentage to which trainers applied therapeutic skills and competence specific to FAME

	% Yes	% Sometimes	% No	% N/A
Introducing activities	100	0	0	0
Organizing subgroups	29.0	22.6	0	48.4
Taking distance	51.6	29.0	3.2	16.1
Continuously being on the move between families	74.2	19.4	0	6.5
Zooming in on interactions	54.8	16.1	6.5	22.6
Asking about perspectives/reflections	90.3	3.2	0	6.5
Letting parents take responsibility for children	22.6	45.2	3.2	29.0
Practicing with new behavior	13.3	33.3	33.3	20.0
Emphasizing link between group and home	93.5	0	0	6.5

Abbreviation: N/A, not applicable.

Concerning exposure, four groups of FAME consisted of seven or eight sessions. Two groups were terminated after the introduction session and measurements, as families indicated that they did not want to continue FAME. Sessions lasted 92 min on average, and an average of five families (at least one family member) participated in each session.

Participant responsiveness varied between sessions. Positive interactions varied between 20% and 90% per session (M = 60%). Variations occurred mostly because some sessions included serious themes, whereas, in other sessions, icebreakers and playing with the children resulted in more positive interactions. Overall, participants participated actively in the sessions (M = 78%). Twelve participants filled in the CSRS. Participants evaluated the subscales “respect and understanding,” “relevance of the goals and topics,” and “client‐practitioner fit” as well as the “overall CSRS score” as satisfactory. The average score on “overall alliance” indicated room for improvement.

The scores for quality of delivery varied widely from 100% (introducing activities) to 13% (practicing with new behavior). During several sessions, the components “organizing subgroups” and “letting parents take responsibility for children” were not applicable, as no children attended the respective sessions.

Evaluation by trainers

Factors promoting feasibility

Four of the five trainers, who together were present in all sessions offered, took part in the focus group. They reported several factors promoting the feasibility of FAME, including collaboration with local professionals. Although the manual of FAME facilitated its feasibility, trainers highlighted the importance of flexibility and deviating from the protocol to address specific group needs. Trainers reported that the structure (icebreaker, activity focused on theme, and closure) and core themes of the program (mentalization, recognition of problems, and parental resilience) were useful.

Trainer: The theoretical background of this program is different from parenting programs that focus more on rewarding, ignoring and punishing, and this is really about mentalizing.

Factors interfering with feasibility

Trainers reported several factors that interfered with the feasibility of FAME. It was not always possible to carry out all activities, for example, when participating families had limited time because of other obligations. Moreover, as a result of issues such as relocations, sometimes, only one or two families attended the session. Finally, trainers stated that there were difficulties with organizing a suitable location, as two different rooms had to be available in order to offer separate activities to parents and children.

Another factor limiting the feasibility concerned difficulties in recruiting and motivating families to take part in FAME. It was often difficult to prioritize FAME when families were faced with pressing issues, such as concerns linked to the asylum procedure.

Although offering FAME to parents as well as children allows families and trainers to work with interactions coming up during FAME sessions, it was difficult to address topics such as harsh parenting and daily stressors when children were present. In addition, the presence of children sometimes caused stress for parents as children walked in and out of the room and asked for their attention during the session. Moreover, adolescents did not want to join FAME together with their parents. Trainers noted that, to increase the feasibility, it might help to decide together with the parents whether children should take part in FAME. One trainer noted that it is important to explain in more depth why children are invited to the group, as illustrated by the following quote:

Trainer: How I imagined it at the start, I thought: OK, the group is a sort of reflection of daily life, children are always there. Whether you’re stressed or not, whether you’re having fun or not, they’re present. So as much as possible […] you have to be available, and how do you do that. And that is what you practice within the group. But I wonder if parents are aware of this and if they know what they’re working on at that time.

Evaluation by participants

Complete pre‐ and post‐FAME measurements were available for six participants. For illustrative purposes, their qualitative and quantitative evaluations are described below.

Strengths of FAME

Four participants found FAME useful. Most participants (n = 4) stated coming together with other families allowed for sharing perspectives, ideas, and experiences. For example, participants explained that they learned how others deal with similar problems. Several participants found the concrete exercises and advice offered during FAME helpful. For example, one mother felt relieved after writing down her problems and throwing them in the metaphorical “bucket” (defining insolvable and solvable problems). A father added that the metaphor of the treasure chest helped him to find solutions for his problems. A mother emphasized that FAME increased her awareness of the impact of stress on her children and of the importance of protecting them from stress‐related reactions, as illustrated by the following quote:

I did see a lot of changes. Especially what I learned during the course: if I feel stress, that’s not a good sign for my daughter, and I can’t take care of my daughter that well. […]. If I don’t have a lot of stress I can take better care of my baby. And also that I show the stress less to my daughter. So that I go away from her if I have stress, and not show her.

Weaknesses of FAME

Two participants did not find FAME useful. Both stated that although important subjects were addressed, there was insufficient time to discuss these subjects in depth, and too much time was spent on subjects that did not concern parenting and on problems that could not be solved. In addition, several participants (n = 3) reported that although they felt supported by the other families and trainers during FAME, they did not experience a change in social support outside the group. Two participants added that their future was too insecure and/or there were no solutions offered for (continuous) stressors. A mother added:

[Nothing changed.] This is due to my procedure. Four families have been deported from the asylum center. This causes a lot of stress, I cannot control it. This isn’t going any better now.

Impact of FAME

Several participants (n = 4) reported an impact of FAME. For example, some experienced a decrease in worrying, feelings of helplessness, stress, or anger. Others noted that they were more patient or calm. Moreover, a mother explained that FAME increased her understanding of her children’s perspectives.

Table 7 shows the mean total PHQ‐4 and SCORE‐15 scores for each participant at t 1 and t 3. The participants who filled in the pre‐ as well as the post‐FAME questionnaires took part in most sessions of FAME. Two of the six participants who completed the PHQ‐4 measurements reported improvements in depression and anxiety scores from t 1 to t 3, as indicated by the RCI. The other four participants reported no significant change in depression and anxiety scores. One of the five participants who completed the SCORE‐15 reported an improvement in family functioning from t 1 to t 3. The other four participants reported no significant change in family functioning.

TABLE 7

Reliable change indexes on PHQ‐4 and SCORE‐15 scores per participant (N = 6) from t 1 to t 3

Participant no	PHQ‐4 t 1 (M)	PHQ‐4 t 3 (M)	RCI PHQ‐4 t 1–t 3	SCORE‐15 t 1 (M)	SCORE‐15 t 3 (M)	RCI SCORE‐15 t 1–t 3
1	2	4	.83	— a	—	—
2	6	5	−.41	27	33	.96
3	9	6	−1.24	36	30	−.96
4	9	7	−.83	21	26	.80
5	12	5	−2.89*	59	46	−2.07*
6	12	7	−2.07*	54	46	−1.21

Abbreviations: PHQ‐4, patient health questionnaire for depression and anxiety; RCI, reliable change index; SCORE‐15, systematic clinical outcome and routine evaluation.

Data on the SCORE‐15 are missing as the family only consisted of mother and baby.

Clinically significant decrease.

DISCUSSION

This study is the first to investigate the feasibility of FAME, a preventive multifamily program for asylum seeker families.

Program integrity

We found that several deviations to the program protocol occurred. Two groups were discontinued after the introduction session. As a result, exposure to FAME differed per family, as the number of families attending the sessions, and the number of sessions per group varied widely. Overall, families participated actively in the program. According to the trainers, factors promoting the feasibility of FAME included collaborating with local professionals and using the FAME manual. Factors limiting feasibility included constraints in time, attendance, and physical space.

Evaluations, family functioning, and distress

A small group of participants took part in all measurements. Several participants stated that they had appreciated gathering with multiple families. Although not all participants noted an impact of FAME, some had developed new coping strategies and were more aware of the impact of stress on their child(ren). The quantitative results offer an indication that FAME might coincide with a decrease in anxiety and depression. However, only one of the participants reported a statistically significant improvement in family functioning. This might be an indication that FAME, in its current form, has a limited impact on the improvement of family functioning in asylum seeker families.

Methodological challenges

Several methodological challenges arose during the implementation and evaluation of FAME. As challenges in the recruitment, inclusion, and retention of families were anticipated, several measures were planned to limit the impact of these challenges, as described in the published study protocol (Van Es et al., 2019). For example, an introduction session was offered to inform families about FAME and to exchange mutual expectations, as suggested by Asen and Scholz (2010). In addition, trainers and interpreters went door to door to invite families to take part prior to each session. Researchers visited the living locations on multiple occasions to offer participants several opportunities to take part in the measurements. However, some challenges could not be overcome, such as participants facing deportation, families relocating, and new families moving into the center.

Implementation challenges

Trainers noted that they found it difficult to compromise between adhering to the study protocol and deviating from the protocol to adjust to the families’ needs. Although higher program adherence is expected to lead to better outcomes, several studies have indicated that some adaptions to a program can lead to more positive program evaluations and better outcomes possibly because trainers adapt the program to the participants’ needs (Kösters et al., 2017; Durlak & DuPre, 2008). Studying program integrity as well as program outcomes could shed light on whether adaptations have led to more positive evaluations and outcomes (Perepletchikova & Kazdin, 2005; Perepletchikova et al., 2009). Unfortunately, this study does not allow for drawing any conclusions on the impact of program integrity on program outcomes because of the small sample size and low response rate.

Deviations from the protocol in this study were mostly due to the absence of children and trainers taking time to address themes brought up by participants. Parents and trainers were often hesitant to discuss certain sensitive topics in the presence of the children. Other studies describing multifamily therapy have addressed this issue by conducting separate groups for parents and children, for example, only including the children at the start and end of each session (Fristad et al., 2009). Calvo et al., (2014) noted that it is more suitable to offer parallel groups to adolescents and their parents, as adolescents face developmental challenges such as differentiation from their parents. Therefore, a clear explanation should be offered on why children are invited, worries that parents have concerning the attendance of their children should be addressed, and parents should be involved in deciding whether or not to invite the children.

Asylum seeker families face several continuous stressors, such as uncertainty, concerning their asylum status and financial difficulties. Fazel and Betancourt (2018) stated that, as a result, it might be difficult to prioritize other concepts such as mental health. Although several participants found FAME useful, some added that FAME could not address these important continuous stressors, as the program does not offer any practical solutions. This finding highlights the importance of managing expectations (FAME does not offer practical solutions) and offering clear psycho‐education (FAME aims to offer ways to cope with worries and stress caused by the problems). When offering a program to refugee families, it is important to assess whether key elements of the program match the needs of the families (Weine, 2011). FAME addresses families’ needs in a culturally sensitive, flexible manner. However, for some families, FAME might not be offered at the right time, as other priorities, such as moving house and asylum procedures, require their attention.

Limitations

Several limitations of the study must be noted. Firstly, because of the small sample and low response rate, we cannot draw conclusions about the effectiveness of FAME. Secondly, we were not able to explore why two groups discontinued after the introduction session. Thirdly, social desirability bias may have occurred due to (co‐)trainers conducting the follow‐up interviews/focus groups. Finally, to our knowledge, no standardized and culturally appropriate measures exist to assess family functioning in the wide range of cultural backgrounds of the study participants. There is a need for the development of culturally sensitive assessments of family processes (Weine et al., 2008).

Recommendations

When planning an intervention in the context of trauma and resettlement, several challenges will exist. As Weine (2011) stated: “The required efforts would be substantial, but then so would the pay‐offs” (p. 426). Our findings indicate several implications and recommendations for the implementation of FAME. Firstly, we recommend collaborating with local professionals, who can offer information on how to address families, can aid with inviting and motivating families to join the introduction session, and can be trained to offer FAME themselves. Moreover, to improve the feasibility of FAME, several conditions should preferably be met when selecting locations, including sufficient space, time, and a minimum number of participating families. Attendance rates can be improved by ensuring that there is a match between the needs of the asylum seeker families and the aims of FAME. The implementation of FAME can be improved by putting more emphasis on psycho‐education and explaining more clearly why a certain group is invited to participate, why FAME is offered, and why parents as well as children are asked to participate. Finally, flexibility in meeting families’ needs will likely contribute to higher retention.

Recommendations for future research include flexibility in research methods to accommodate challenges inextricably linked to the conducting research with asylum seekers in the setting of asylum centers. We made several adaptations to the original research protocol to increase the feasibility of the study. Because of the language barrier, we preferred using the more visual child version of the SRS for studying parents’ evaluations of therapeutic alliance. We also decided not to use videotapes, as participants might be hesitant to give their permission to use these measurements. In addition, the feasibility of a pre–post‐test design is highly impacted by offering an open group where families can join FAME in a later session or have to stop early because of relocations. Offering a program to a group of culturally diverse refugee families in resettlement calls for modifications in response to learned lessons and contextual changes (Weine, 2011). Future studies should include qualitative measures in order to give room to participants’ voices and increase our understanding on whether and how FAME addresses their needs. By using qualitative data, researchers can explore whether other factors asylum seeker families are faced with, such as news concerning the asylum procedure or stress concerning relocations, have an impact on the outcomes.

CONCLUSION

Few studies have focused on the evaluation of multifamily programs for asylum seekers families (Slobodin & de Jong, 2015), and to our knowledge, this is the first to evaluate the feasibility of a multifamily program designed specifically for asylum seeker families in the Netherlands. Several challenges to the feasibility of the implementation and evaluation of FAME were encountered. When offering a preventive family‐focused program, we advise to (a) invest in the practical organization of the group whilst also allowing flexibility to address families’ specific needs; (b) offer clear expectation management and psycho‐education; and (c) evaluate the group in an accessible manner, allowing flexibility in the assessments, as it concerns a preventive, open group that is prone to contextual changes.