Strengthening Parenting Among Refugees in Europe (SPARE): A Collaborative Feasibility and Acceptability Study Among Four European Nations

ABSTRACT

Per the United Nations High Commissioner for Refugees (UNHCR), record numbers of people have been forcibly displaced in the past decade due to life‐threatening conditions. The negative effects of traumatic experiences associated with forced displacement are well known. Strengthening Parenting Among Refugees in Europe (SPARE) is a trauma‐informed version of the evidence‐based parenting programme Generation Parent Management Training Oregon Model (GenerationPMTO), developed and implemented by GenerationPMTO experts in Europe. SPARE is intended to support refugee parents and their children and facilitate their inclusion in a new community. Using mixed methods, we examined the feasibility, acceptability, and initial pre/post outcomes of SPARE for refugee families with children ages 0–18 in Iceland, Norway, the Netherlands and Denmark. Parents in all SPARE groups reported high levels of satisfaction with the programme and provided suggestions for its improvement. Parents also reported positive changes in child adjustment, parent psychological distress, and parenting practices following SPARE, with small to moderately sized effects. Differences in parent‐rated outcomes across countries were noted. Overall, SPARE appears to be a promising and feasible contribution to the dearth of evidence‐based and culturally adapted prevention programmes for refugees in Europe. Further investigations of SPARE and its implementation are needed.

1. Introduction

Europe is experiencing a veritable flood of families seeking refuge from war, gang violence, effects of climate change, earthquakes, poverty, and other catastrophes. Today, around 90 million people in the world have been forced to flee their home country. Of those, more than 20 million are refugees, and overall, 41% of these people are children. About half of the refugees in Europe come from the Middle East, Venezuela, and Myanmar (United Nations High Commissioner for Refugees [UNHCR] ²⁰²³). The war in Ukraine has significantly increased the number of refugees in Europe; from fall 2022, about 7.5 million refugees from Ukraine have relocated to other European countries (UNHCR ²⁰²³).

According to a review by Gewirtz et al. (2022), refugee children arriving in Europe generally experience elevated rates of psychiatric disorders compared to their host peers, and mental health is worse for refugee teenagers and adults compared to others living in European countries. World‐wide, refugees are at high risk of suffering from trauma, poor mental health, and other adjustment problems (Bogic et al. 2015; Gewirtz et al. ²⁰²²; Schick et al. ²⁰¹³). The traumas these families have experienced make it challenging for parents to raise their children as they recover from their losses and adapt to new cultures and contexts, which increases the urgency to make more evidence‐based interventions (EBIs) available (Gillespie et al. 2022). Evidence‐based parenting programs for refugees in Europe are now in great demand (see Gewirtz et al. ²⁰²²). Providing parents with EBIs that strengthen parenting strategies can empower them and help prevent and address their children's adjustment problems (e.g., Cai et al. ²⁰²²; Forgatch and Patterson ²⁰¹⁰; Van Ee et al. ²⁰¹⁶). Strengthening Parenting Among Refugees in Europe (SPARE) is a trauma‐informed intervention modelled after GenerationPMTO adapted to support refugee families in Europe. This study examines feasibility, acceptability, and initial findings from the introduction of SPARE for Arabic‐speaking families in Iceland, Norway, the Netherlands and Denmark.

1.1. GenerationPMTO

GenerationPMTO has more than 50 years of development and research (e.g., Dishion et al. ²⁰¹⁶; Patterson ²⁰¹⁶) with excellent ratings from clearinghouses that evaluate evidence‐based programmes (e.g., Blueprints for Healthy Youth Development, n.d.). The Social Interaction Learning Model, which involves the fusion of social learning, social interaction, coercion theory, and behavioural perspectives, provides the theoretical underpinning of GenerationPMTO (Patterson 2016). Therapists help parents practice core positive parenting components (i.e., skill encouragement, limit setting, monitoring, problem solving, and positive involvement) and supportive tools (i.e., clear directions, active communication, observing and regulating emotions, and promoting school success) (Forgatch et al. 2016; Sigmarsdóttir et al. ²⁰¹⁶). The GenerationPMTO programme exists in individual, group, and telehealth formats.

A randomised controlled trial (RCT) testing the model with multiple method measures and longitudinal assessment found the intervention to benefit child outcomes 9 years post baseline (i.e., teacher‐reported externalising behaviour, police‐reported arrests, and child/youth‐reported internalising problems and deviant peer association). These improved child outcomes were mediated by reduction in coercive parenting and increases in observed positive parenting practices, thereby providing experimental support for the model (Forgatch et al. 2009; Patterson et al. ²⁰¹⁰). Positive outcomes have been shown for pre‐ and elementary school‐aged children, for prevention and treatment of children's adjustment problems, especially externalising and internalising difficulties, and further details can be found in articles, chapters, and parent books (e.g., Dishion et al. ²⁰¹⁶; Forgatch and Domenech Rodríguez ²⁰¹⁶; Forgatch and Gewirtz ²⁰¹⁷; Forgatch et al. ^{2017, 2016}; Rains et al. ²⁰²¹).

The GenerationPMTO intervention has been translated and provided to families in several languages, including Norwegian, Danish, Dutch, Icelandic, and Spanish (Askeland et al. 2019; Parra‐Cardona et al. ²⁰¹⁶; Sigmarsdóttir et al. ²⁰¹⁹). Minor adaptations were made for language, appropriate metaphors, and locally relevant artwork as suggested by experts (e.g., Domenech Rodríguez et al. ²⁰¹¹). Some adaptations have provided the program to immigrant, refugee, and displaced families using a link‐worker to translate and provide cultural interpretation during parent group sessions. Examples include Somali and Pakistani families in Norway (Bjørknes et al. 2012), Karen families in Minnesota (Ballard et al. 2018), and war‐displaced mothers in Northern Uganda (Wieling et al. 2015). After Deployment, Adaptive Parenting Tools (ADAPT) added two components (i.e., mindfulness and emotion coaching) to GenerationPMTO for families who have experienced trauma (Gewirtz et al. 2016, 2018). Mediational models testing the intervention for Somali and Pakistani families in Norway and ADAPT for military families reported benefits to parenting, which in turn mediated improved child outcomes (Bjørknes et al. 2012; Gewirtz et al. ^{2016, 2018}), thereby replicating the basic model.

1.2. Development and Adaptation of SPARE

The SPARE program, funded primarily by two European grants from Erasmus+ and Nordplus, was developed by a team of GenerationPMTO experts from Iceland, Norway, the Netherlands and Denmark. This team has experience adapting, implementing, and studying intervention effects based on a decade or two of providing GenerationPMTO to families in Europe with high fidelity (Askeland et al. 2019; Sigmarsdóttir et al. ²⁰¹⁹). The group collaborated through many in‐person and virtual meetings to adapt GenerationPMTO for this at‐risk population in Europe. Because Arabic‐speaking people comprised one of the largest groups of refugees in Europe at the time the SPARE program was initiated, an adaptation for this particular group was the first focus of this endeavour.

1.3. Main Cultural and Contextual Adaptations

In a study, comparing two levels of adaptation, Parra‐Cardona and his colleagues found that both versions achieved positive outcomes, but the addition of two sessions addressing culture and change provided greater benefits (Parra‐Cardona et al. ²⁰¹⁷). In the case of SPARE, the adaptation provides the core GenerationPMTO components listed earlier, with two components from ADAPT added to address trauma issues: mindfulness (called observing) and emotion coaching (Gewirtz et al. 2018). Other relevant cultural and contextual adaptations were made with respect to language, methods, illustrations, metaphors, goals, concepts, and the use of link‐workers. Table S1 provides information based on Bernal et al.'s (1995) Ecological Validity Model (see also Domenech Rodríguez et al. ²⁰¹¹). When conceptualising levels of adaptation as existing on a continuum spanning surface‐ to deep‐level adaptations (e.g., Parra‐Cardona et al. ²⁰¹⁷), SPARE adaptations may be placed approximately in the middle.

There are similar and varied contexts for the refugees in the four European countries. Iceland, Norway, the Netherlands and Denmark all use publicly driven specialist and social services to help families needing assistance because of social, financial, and family difficulties. All these countries employ strong child protective services systems, with Norway arguably having the most rigorous systems (Fylkesnes et al. 2017). In Norway, the Netherlands and Denmark, municipalities where the refugees have settled are required by law to offer typical services that are provided to all residents as well as special services designed for refugees to promote efficient integration to society (The Immigration Act; Lovdata ²⁰²²; Ministry of Immigration and Integration ²⁰²⁰; Government of the Netherlands n.d.). In addition, in Norway a mandatory program introduces refugees to Norwegian culture and society through a parental guidance course (The Directorate of Integration and Diversity). In Iceland, there are no specific laws to help refugees adapt to new surroundings (Institute of International Affairs, University of Iceland ²⁰¹⁷). Services for refugees vary depending on the community where they live, and to our best knowledge, no evidence‐based parenting program is offered systematically to refugee groups.

In the opening phase of the SPARE project, a small study was conducted in Iceland and Norway to assess feasibility and acceptability. The intervention was provided to two parent groups with fourteen Arabic‐speaking families, seven in Iceland and seven in Norway. Attendance was good, parents reported satisfaction with their experiences, and they said that the material addressed important family needs. The groups began at the onset of the COVID‐19 pandemic. Lockdowns prevented planned group startups in Denmark and the Netherlands and required changes in intervention procedures for the groups conducted in Iceland and Norway. For further details, see Sigmarsdóttir, Arnesen, and Forgatch (2023).

1.4. Current Study

The present study extends the earlier work with data from a total of seven SPARE groups conducted in Iceland, Norway, the Netherlands and Denmark. The overall aim is to evaluate acceptability, feasibility, and initial outcomes of the SPARE groups for Arabic‐speaking refugee families. The expansion of SPARE to four European nations, with initial outcomes that had not been tested before, provides a broader picture of the intervention in Northern Europe in general. Our main research questions follow: (1) What can be learned from parents' voices, attitudes, and attendance in further planning of SPARE for European refugees? (2) To what extent do parents report acceptance of and satisfaction with SPARE in each country? (3) To what extent do parents report benefit to children's behaviour, parenting practices, and parent psychological distress before and after SPARE groups?

2. Method

2.1. Design

In this paper, we describe findings from the second phase of the feasibility and acceptability study undertaken in collaboration among four European countries. A mixed methods design was used to collect data from a total of seven SPARE groups; two were conducted in Iceland, Norway, and the Netherlands, respectively, and one group in Denmark.

2.2. Participants

Participants were 63 Arabic‐speaking refugee parents with a total of 134 children (up to age 18), representing 45 families. We decided on a wide age range for the children to ensure sufficient participants and to optimise opportunities for the refugee parent population. Although the program is manualised, it gives well‐trained therapists flexibility to adapt to parents of children within a wide age range, for example, by using different role plays based on child age. Parents were from the Middle East and Africa and had fled to Europe from a total of 13 origin countries, most of them because of war (e.g., Iraq, Syria, and Palestine). Thus, the inclusion criteria were as follows: (1) Refugee status; (2) having a child aged 0–18; (3) Arabic speaking. Two thirds of the participants had lived in the arrival country for 1 year or less, and most parents were married. In Iceland and Norway, mothers and fathers were equally represented; all participants in the two Dutch groups were mothers, and four of the six parents in the Danish group were mothers. All parents had completed elementary/primary school, and some had high school and college or university education. Except for the Icelandic participants, annual family income was reported by most parents to be far below the defined poverty line in the European Union. Dutch participants did not report on family income. Parent education and family income were not significantly correlated (r = 0.15). Family demographics are displayed in Table 1.

TABLE 1.

Demographic information for participants in Iceland, Norway, the Netherlands and Denmark.

	Iceland	Norway	The Netherlands	Denmark	Total
Families	17	11	13	4	45
Parents	28	16	13	6	63
Mothers	14	8	13	4	39
Fathers	14	8		2	24
Parent age
< 20 years	1				1
21–25 years	5	2			7
26–35 years	9	8	3	4	24
36–45 years	8	5	7	1	21
46–55 years	3		1		3
> 55 years	2				2
Missing data		1	2	1	4
Number of Children	42	32	35	8	117
Mean # of children per family	2.5	3.2	3	2.7	2.6
Missing data		1	1	1	3
Children's gender
Girls	24	12	9	4	49
Boys	18	13	17	4	52
Missing data		7	9	1	17
Children's age
0–3 years	9	6		2	17
4–7 years	15	11	7	3	36
8–12 years	12	7	9	3	25
> 13 years	6	4	8		18
Missing data		4	11	1	16
Family household
Both parents	12	9	9	2	32
Single mother	2	1	2	1	6
Single father	3				3
Mean # of family members	4.1	5.6	3.3	4.3	4.3
Missing data		1	3	1	5
Family living area
Urban	17	8	8	4	37
Rural		2	3		5
Missing info		1	2		3
Time living in arrival country
0–1 years	24	10	9		43
2–4 years	4	2	1	3	10
5 years or more		3	1	2	6
Missing data		1		1	2
Parent education
Elementary school	16	9	2		27
1–2 years of high school		2		2	4
Graduated high school	5	2	4	1	12
1–2 years of university	4		2	2	8
Graduated college/university	2	2	3		7
Missing data		1	2	1	4

2.3. Intervention, Adaptation, and Procedures

The Icelandic and the Norwegian procedures were reviewed and accepted by the Icelandic National Bioethics Committee (VSNb2020010017/03.01) and the Norwegian Agency for Shared Services in Education and Research (817650), respectively. In the Netherlands and Denmark, it was not mandatory to have the procedures reviewed and accepted by a Science Ethics Committee; however, they followed the same procedures as Iceland and Norway. All participants signed the same written informed consent form before giving any information, and each person was assigned a participation number to de‐identify data and ensure confidentiality.

These SPARE groups were conducted following the initial feasibility study in Iceland and Norway. Some groups were run post pandemic restrictions, enabling all group sessions to be conducted in person. Intervention content was standard throughout the four countries (see session order and content in Table 2 and in Supporting Information) and steps in the adaptation procedure are described in Table S1 striking a balance between surface level and deep level adaptations (see Parra‐Cardona et al. ²⁰¹⁷). All countries divided parents into groups of 4 to 15 attendees each, with one or two parents representing each family. Strategies to recruit families varied. In Iceland, participating parents were recruited from one municipality, the capital of Iceland. Recruitment was carried out in collaboration between the PMTO centre in the capital and professionals responsible for refugee welfare located in the city service centres. In Norway, parents were recruited from refugee services in one urban and one rural municipality as part of the national mandatory introduction programme for immigrants (Ministry of Labour and Social Inclusion ²⁰²⁴). Regional PMTO therapists at the Norwegian Center for Child Behavioral Development (NUBU) responsible for the two municipalities, gave information about SPARE to the refugee services, who in turn invited parents to participate. In the Netherlands, parents were recruited at schools specialising in teaching refugee children in two urban municipalities. In Denmark, participating parents were recruited from one urban municipality. Recruitment was carried out in collaboration between a PMTO specialist and a referral team, both employed in the child welfare department of the municipality. Some participants were invited to participate in the group by their family counsellors, as a supplement to ongoing treatment.

TABLE 2.

Parent attendance per session in Iceland, Norway, the Netherlands and Denmark.

Session content	Iceland (n = 28)	Norway (n = 16)	The Netherlands (n = 13)	Denmark (n = 6)
1. The start of the journey	100	83.4	70.9	83.3
2. Clear directions	96.2	65.8	82	100
3. Teaching positive behaviour	96.2	80.2	82	83.3
4. Recognising emotions	96.2	81.8	65.3	83.3
5. Responding to emotions	92.3	83.4	65.3	83.3
6. Setting limits	92.8	73	36.1	66.7
7. Following through	82.8	47.7	65.3	100
8. Foundation for emotion coaching	85.7	80.2	47.2	50
9. Family problem solving	92.3	80.2	43.1	83.3
10. Monitoring and supervision	82.3	67.5	18.1	0 a
11. Emotion coaching	84.6	77.8 a	52.8	66.7
12. Balancing life	80.8	88.9 a	75 a	83.3
Average	90.2	74.7	58.6	73.6

^a Session not conducted for one of the groups in that country.

Each SPARE group was led by two certified PMTO‐therapists and one link‐worker. The link‐workers, who had diverse educational backgrounds, originated from an Arabic‐speaking country, understood the language, and were familiar with the participating families' culture. All materials for therapists, link‐workers, and parents were created in English, and parent material was translated into Arabic by experts within the four collaborating nations. Group leaders were all trained PMTO therapists; most had a background in psychology or social work, and they received at minimum a two‐day additional training in the SPARE programme. As part of training, therapists practiced the additional trauma‐related components and familiarised themselves with the adapted curriculum. They also received information about the refugee group and expanded their skills regarding what to expect and how to deal effectively with trauma. Link‐workers also received training where information about GenerationPMTO and SPARE was presented, and skills were practiced for at least 2 days. Therapists and link‐workers received regular coaching throughout the project.

SPARE was offered through 12 in‐person, 2‐h weekly parent group sessions in which each session contained training in different effective parenting practices as well as an observational activity (see Table 2 for session order and content). Sessions followed a structured agenda in which therapists balanced completing the agenda with responding to issues raised within the group. Within sessions, therapists also discussed challenges refugee families face as they adjust to a new country and how SPARE tools could be applied to address them. Parents were able to communicate in Arabic with each other and through the link‐worker. Parents received a home practice assignment at the end of each session, which was debriefed in the beginning of the next session. Troubleshooting facilitated clarifying procedures and adjusting for specific family circumstances. Between‐session communication varied across countries.

In all countries, SPARE was conducted either in English (Iceland) or the country's language (Norway, Denmark, Netherlands) while the link‐worker in each group translated simultaneously to Arabic. Slight variations between countries included missing sessions. After each session, parents received a brief phone call (or WhatsApp message) from the link‐worker reminding them of their homework and helping them resolve problems. Session order and content can be seen in Table 2. Further description of the SPARE program can be seen in Supporting Information and in an earlier paper by Sigmarsdóttir, Arnesen, and Forgatch (2023).

2.4. Measures

Native Arabic speakers with fluency in English at a high professional level translated and back translated all parenting measures from English to Arabic. Parents completed questionnaires on parenting and child behaviour before and after the 12 group sessions and satisfaction questionnaires at the end of each session. Parents completed questionnaires in their preferred language in paper–pen format and link‐workers assisted as needed, with a research team member present.

2.5. Attendance

Therapists recorded parents' session attendance.

2.6. Satisfaction Survey

A 20‐item questionnaire adapted from Adaptive Parenting Tools (ADAPT; Gewirtz et al. ²⁰¹⁴) was administered to parents at the end of each session. Each item was rated on a 5‐point scale ranging from ‘1 = Not at all true’ to ‘5 = Very true’. Items load onto three subscales: parent satisfaction (α = 0.58 to α = 0.85 across sessions), positive group experiences (α = 0.76 to α = 0.96), home practice satisfaction (α = 0.58 to α = 0.89). Additional information is presented in Appendix S1. At the end of the 12 SPARE sessions, parents rated their overall satisfaction, how helpful or useful they perceived specific session content, as well as the parent material, home practice assignments, and the facilitators' teaching. Finally, parents were asked for suggestions regarding ways to improve SPARE, and if they would recommend the course to others.

2.7. Focus Group Interview

Upon SPARE completion, participants from each group who attended the last session participated in a semi‐structured group interview based on a template by Parra‐Cardona et al. (2016). The interview lasted about 1.5 h, and each group participated in the interview together to maintain procedural consistency in all countries. Interview discussions covered three grand‐tour themes regarding parents' experiences with SPARE: (1) relevance of talking about and practicing specific parenting strategies (e.g., encouragement, limit setting, observing), (2) how it was to participate in the group, and (3) relevance of focusing on differences in cultural backgrounds as parents in a new country, including status as a refugee. To mitigate bias, interviews were conducted by trained graduate students who had not participated in SPARE. Interviewers strove to elicit responses from all participant. Each group's link‐worker assisted in simultaneous interpretation of questions and responses between the interviewers and parents. Parents responded to questions and to each other's examples. Given the feasibility focus, qualitative data from all groups were combined and analysed holistically to obtain a general overview of participants' views and experiences. Variations in participant responses across countries are highlighted (see Results section).

2.8. Therapist Fidelity

Four different SPARE sessions covering two core components (encouragement and limit setting) were chosen to assess therapist fidelity (i.e., ‘introduction to encouragement’/session 3, ‘troubleshooting encouragement’/session 4, ‘introduction to limit setting’/session 6, and ‘troubleshooting limit setting’/session 7). Therapists self‐rated their intervention fidelity for the four sessions on a questionnaire consisting of 10–15 statements that correspond to the content of that session. Each statement is rated on a 5‐point scale ranging from 1 (false) to 5 (very true).

2.9. Eyberg Child Behaviour Inventory

The Eyberg Child Behaviour Inventory (ECBI) is a 36‐item parent‐report questionnaire of children's adjustment problems. The ECBI's two subscales rate (i) frequency and intensity of behaviours on a 1 (never) to 7 (always) scale, and (ii) whether parents find particular behaviours problematic (yes = 1; no = 0). The ECBI has good psychometric properties (Burns and Patterson 1991, 2000; Robinson et al. ¹⁹⁸⁰); with internal consistency in this study α = 0.93 to α = 0.96 for the total sample, and α = 0.99 (Norway) to α = 0.62 (Denmark) for each of the four countries. Additional information is presented in Appendix S1.

2.10. SCL‐10

Parents reported on their own psychological distress (anxiety and depression) with the Hopkins Symptom Checklist‐10 (SCL‐10), which includes 10 statements about the past week's level of distress rated on a 4‐point scale from 1 (not at all) to 4 (extremely). Cronbach's α was 0.91 (pre‐test) and 0.85 (post‐test) for the whole sample. Additional information is presented in Appendix S1.

2.11. Parent Practices Interview

To measure parenting skills, a three‐subscale version (Tømmerås et al. 2018) of the self‐assessment Parent Practices Interview (PPI) was administered to parents (Webster‐Stratton ¹⁹⁹⁸). We found acceptable internal consistency for the positive parenting scale at pre‐test (0.76) and post‐test (0.80) and for the harsh discipline scale at pre‐test (0.77) and post‐test (0.80). Because the inconsistent discipline scale had low internal consistency at pre‐test (α = 0.17) and post‐test (α = 0.50), results for that subscale were omitted from further analyses. Test–retest reliability of parents' ratings before and after SPARE was r = 0.41. Additional information regarding PPI is presented in Supporting Information.

2.12. Data Analysis

Focus group interviews were audiotaped, transcribed, and analysed according to qualitative content analysis (Graneheim and Lundman 2004). Patterns were identified, analysed, and reported deductively (see Bingham ²⁰²³), which is in line with the three grand themes in the semi‐structured interview guide originally developed by Parra‐Cardona et al. (2016). Transcriptions were read and re‐read to get initial ideas of the material and a sense of parents' perceptions. We organised and structured these data along the three original themes focusing on: (1) relevance of SPARE materials, components, and skills (e.g., encouragement, limit setting, observing); (2) experiences of SPARE participation; and (3) cultural differences.

SPSS version 29 was used to examine quantitative data, including normality distribution, homoscedasticity, and possible outliers. Missing data were handled per scoring instructions for each measure. Means and standard deviations were calculated for parent satisfaction ratings and therapist ratings on parent attendance. Means and ranges were calculated for therapist fidelity scores. To examine whether scores differed from pre to post, t‐tests were used for the total sample. However, due to the low sample size in each country (n < 30), we used the Wilcoxon signed‐rank test, a non‐parametric test, as an alternative to the t‐test. Kruskal–Wallis tests were conducted to explore differences in ECBI, SCL‐10 and PPI scores between countries.

3. Results

In total, 63 parents (39 mothers) provided data from at least one of the 12 sessions. The seven groups from the four countries were unequal in terms of group size, varying between n = 4 (one Dutch group) and n = 15 (one Icelandic group), which might impact tests of homogeneity.

3.1. Feasibility and Acceptability

3.1.1. Attendance

Twelve SPARE sessions were provided for all groups except for one group in Norway and one Dutch group (10 sessions), and the Danish group (11 sessions). Average parent attendance for all countries ranged between 90.2% (Iceland) and 58.6% (the Netherlands). Data are summarised in Table 2.

3.1.2. Satisfaction Survey

Means and standard deviations of parents' session ratings are displayed in Table 3. As shown in Table 3, all subscales showed consistently high ratings across countries and sessions.

TABLE 3.

Means and standard deviations of satisfaction ratings across SPARE sessions.

	M (SD)
Subscale	Iceland	Norway	The Netherlands	Denmark
Parents' satisfaction	4.9 (0.42)	4.5 (0.47)	4.7 (0.19)	4.5 (0.48)
Positive group experiences	4.7 (0.40)	4.3 (0.57)	4.7 (0.28)	4.3 (0.34)
Home practice satisfaction	4.5 (0.56)	4.3 (0.53)	3.9 (0.68)	3.6 (0.63)
Home practice was hard	1.5 (0.87)	1.8 (1.0)	2.1 (0.88)	1.7 (0.83)
Experienced negative emotions in session	1.1 (0.27)	1.0 (0.16)	1.3 (0.57)	1.3 (0.49)

Note: 1 = not at all true; 2 = slightly true; 3 = somewhat true; 4 = quite true; 5 = very true.

3.1.2.1. Overall Satisfaction

Parents in the four countries found the group helpful overall (means 4.6–4.9), rated their use of parenting techniques positively (means 4.1–4.8), liked the information, and felt that group facilitators were understanding and responsive to their situation. Parents across countries noticed positive changes in their children's behaviour and no or minimal negative changes in their children's behaviour. SPARE components and content overall were rated as helpful (means 3.6–5.0). Parent material, home practice assignments, and facilitators' teaching abilities were rated between 3.8 to 4.9. In unison, parents stated they would recommend SPARE to other parents (M = 5.0, SD = 0.00). More material on overall satisfaction with the intervention can be found in Supporting Information.

3.1.3. Focus Group Interviews

Three themes were highlighted: Parents' perceived relevance of SPARE materials and components; experiences of group participation; and cultural differences between their countries of origin and their arrival countries. Generally, parents in all countries had positive experiences and offered valuable suggestions for program improvement. Below we provide comments and examples from all groups. An overview from all countries as well as main participant suggestions about how the program could be done differently can also be found in Supporting Information.

3.1.3.1. Parents' Perceived Relevance of SPARE Materials and Components

Parents from all groups emphasised the usefulness of the parenting tools and the opportunities to practice them through role play. Illustrative quotes include the following: “Feelings, monitoring, limit setting, praise, and the incentive chart; all was useful.” “It was a good feeling to give them encouragement and see how it gave them confidence. The children's safety was built up over time.” One mother said: “I knew these ideas, but they had faded away. I think it's very important to compliment them, to reward good behavior.” Role play helped parents to practice at home. One mother said: “This makes it easy, because if you already have the example, then you understand it better, and then you are more confident because you have practiced well.” A father said: “I know how to handle situations with my child and how to talk with her. I got to practice during the course.”

In Iceland and in the Netherlands, parents were enthusiastic about setting limits, especially the steps within the time out procedure and how it helped them stay calm. One mother said: “The procedure helps me to stay calm and not to argue with the children.” Another parent said, “It was kind of a punishment, but in a gentle way. Very gentle, and we're not used to that.” Some parents talked about difficulties with setting limits and would have liked even more time on this topic (Norway) and one father (Denmark) said: “I used it once. I do not think it was relevant. I had other ways. She is not that stubborn; hence time out was not necessary.”

Regarding monitoring, several parents underlined the balance between keeping an eye on the children without being overly controlling and how it gave them a new perspective. They learned to understand their children's moods and behaviour, to intervene when they thought something was wrong, and to talk to the children about the problem. Some parents in Iceland thought control should be discussed at the very beginning of the course: “We learned to see what behaviors to focus on when we observed [the children] and the children learned from us and began to observe each other and report back to us. We then managed to stay calm and direct them and say: No … do it like this.” Problem solving was regarded as especially helpful in negotiations between siblings of different ages and needs: “We, as a family, try together to find the solutions that we can use to solve the problem.” Some parents of older children (Iceland) had created a tradition for the family to always eat dinner together. The routine was an important time when they could discuss problems that all family members were facing.

Across countries, all groups were satisfied with the trauma‐focused components, observation exercises, and emotion coaching, described them as relaxing, pleasant, and helpful, and said it helped them calm down in difficult situations with their children. A father in Iceland said: “Mindfulness made me think about how I used to be when I was in the military. I was trained to be very tough and strong in Iraq and the meditation helped me to think how bad it was to be that way. Now I see how much I need to change myself.” A mother from a Dutch group said: “The exercises have made a difference. I am calmer at home. And I use breathing exercises before falling asleep.” A few described the audiotaped voice in the sound clip as being too mechanical or a bit boring in longer exercises. Regarding emotion coaching, a parent in Denmark said: “I have learned to wait when my child is angry. I wait till she is calm, and I am calm, and then we talk. It is much easier to talk when we both are calm.” Another said the biggest difference was: “Today I understand my child much better. Before, I did not understand why she became angry. That made me tense and angry. Now I know what to do … And I am more relaxed … I feel I am competent as a parent.”

Parents' experiences of participating in the group were generally positive and they expressed gratitude to the facilitators and link‐workers. The support, shared experiences, and feelings of safety in the group were highlighted: “We learned a lot in the course, but we also felt relieved to hear that others were struggling with similar problems. I was not alone, and it gave me peace to hear that” (Iceland). A parent from Denmark said: “It is much easier to cope when you know others have some of the same difficulties.” A mother in the Netherlands stated: “We are with this group. We have the same pain, the same problems, and the same suffering.” In Norway, the usefulness for all parents was noted: “I think it should be for all new families to go to a group like this. It should be mandatory, not only for refugees, but for all parents.” In sum, the recognition and feelings of being connected were highly valued by participants.

3.1.3.2. Cultural Differences

In Iceland, Norway, and the Netherlands, parents described cultural similarities and differences connected to monitoring, limit setting, and communicating with children. In Denmark, few differences were described, but one said: “In our culture we do not admit when we have problems. It's not a tradition.”

In the Netherlands, parents emphasised differences related to time management: “Time is important here. We have not learned that: arrive on time, eat on time, sleep on time. In my country, family time is more important than going to bed early.” Other parents described how in their countries of origin, monitoring is number one in raising children, but in Iceland there is much more freedom; parents don't always know what the child is doing.

Participants talked positively about how parents in the Netherlands listen to their child and ask for ideas on specific subjects: “I think it's important, how you can talk with your child. With that, you give the child more confidence, his opinion matters, parents listen to his opinion. That is also important to him.” In Iceland, one parent noted how European culture may affect children's behaviour: “Children are starting to take advantage of Child Protective Services and threaten the parents! They tell their parents that if you don't buy me an iPad, I will go to school and tell my teacher that you hit me. Fortunately, we know better today; we know that the Icelandic government does not work like this. The children now know that we are watching them, and they like it.”

When asked to reflect on cultural differences, several parents in Norway highlighted limit setting and physical punishment, fearing that Child Protective Services would remove their children: ‘… when you arrive in a new country, a lot is different, and you need to learn about the differences. After we got this program, we have learned about how we can set limits, which are allowed in Norway, and we know what is not allowed. That has been useful, to know about the laws and rules about what is allowed in Norway. We feel safer about how we can be good parents in Norway’.

In summary, several parents reported that the SPARE tools helped them regain control, set limits, and find a balance between the different cultures, which made them feel safer as parents in a new country.

3.2. Therapist Fidelity

Overall, the average mean therapist fidelity score was 4.5 (range 4.2–4.7) across the four rated sessions (i.e., introduction to encouragement/session 3, troubleshoot encouragement/session 4, introduction to limit setting/session 6, and troubleshoot limit setting/session 7). Average mean fidelity scores across countries ranged between 4.4 (Norway and the Netherlands) and 4.6 (Iceland and Denmark). Scores were most consistent across the four countries for the session Troubleshoot Limit Setting (range 4–5). Conversely, scores were less consistent for Troubleshoot Encouragement across countries (range 1–5).

3.3. Child Behaviour, Parent Psychological Distress, and Parenting Practices Outcomes

3.3.1. ECBI: Intensity of Child Behaviour and Parents' Perceived Child Behaviour Problems

ECBI data are presented in Table 4. Parents from the four countries' combined sample (n = 52) reported a significant mean decrease in pre/post intensity scores. A Wilcoxon signed‐rank test revealed a statistically significant reduction in intensity scores for the Icelandic sample (n = 28), which carried the effect for the whole group; z = −2.59, p = 0.010, with a moderate effect size (r = 0.35). The median score on the ECBI Intensity scale decreased from pre‐ (Md = 85) to post‐SPARE (Md = 77). None of the other countries showed significant change. Note that all pre/post scores were below the clinical threshold depending on child age for all groups (ranges between 110 and 135) except at pretest for the Danish sample (Md = 140).

TABLE 4.

Paired samples t‐test of ECBI, SCL‐10, and PPI; pre‐post differences in mean scores for total sample.

	Pre	Post
Measure	M (SD)	M (SD)	t (df)	p	Cohen's d
ECBI intensity scale (n = 52)	85.4 (33.3)	76.7 (28.5)	2.36 (51)	0.022	0.33
ECBI perceived problem (n = 50)	8.6 (10.1)	5.0 (6.7)	2.98 (49)	0.005	0.42
SCL‐10 (n = 53)	17.2 (7.2)	15.1 (5.5)	2.38 (52)	0.021	0.33
PPI positive parenting (n = 36) a	60.3 (13.6)	67.2 (12.6)	−4.53 (35)	< 0.001	−0.76
PPI harsh discipline (n = 33) a	21.3 (7.7)	19.4 (7.8)	1.51 (32)	0.142	0.26

Note: Higher ECBI and SCL‐10 scores = higher level of symptoms. Higher PPI scores = more positive parent practice, harsher discipline. Bolded p values = significant < 0.05.

^a Norway did not collect PPI data.

Kruskal–Wallis tests revealed statistically significant differences in intensity scores across the four countries at both pre‐ and post‐test χ ² (3, n = 58) = 18.11, p < 0.001. At pre‐test, Denmark (n = 4) reported the highest median score (Md = 140), the Netherlands (n = 11) reported a median value of 95, Iceland (n = 28) 85, and Norway (n = 15) the lowest median score at 49. Also, at post‐test, differences across countries in intensity scores were large, with Denmark reporting the highest (Md = 116.5), the Netherlands recording a median value of 96, Iceland reported 77, and Norway reported the lowest at 46.5, χ² (3, n = 55) = 12.72, p = 0.007.

Parent reports of perceived child behaviour problems also decreased significantly. A Wilcoxon signed‐rank test revealed a statistically significant reduction in problem scores for the Icelandic sample (n = 28); z = −2.05, p = 0.040, with a small effect size (r = 0.27). The median score on the ECBI Problem scale decreased from pre‐ (Md = 4.5) to post‐SPARE (Md = 3). The Icelandic and Norwegian samples rated child behaviour problems far below the clinical threshold (range between 6.5 and 11 depending on child age) at both pre‐ and post‐test. The Dutch and Danish parents rated problem behaviour above the clinical threshold at pre‐test, but within the threshold range at post‐test.

3.3.2. SCL‐10: Parent Psychological Distress

Parents reported a significant reduction in distress from pre‐ to post‐test for the total sample across countries (see Table 4). The Wilcoxon signed‐rank test did not reveal statistically significant changes in SCL‐10 scores for any of the four countries separately. Except for the Netherlands, all countries reported lower levels of distress over the 12 weeks of SPARE.

3.4. Parenting Practices

3.4.1. Positive Parenting

Data for parenting practices are displayed in Table 4. Mean positive parenting scores increased significantly from pre‐ to post‐test for the total sample. Scores from the Icelandic sample carried the effect for the full group. A Wilcoxon signed‐rank test revealed a statistically significant increase in Positive Parenting for the Icelandic sample only from before and after SPARE; z = −3.46, p < 0.001, with a medium effect size (r = 0.48). The median score on the Positive Parenting scale increased from pre (Md = 66) to post (Md = 74).

3.4.2. Harsh Discipline

Mean scores showed no significant pre‐ to post‐test change for the total sample. Similarly, a Wilcoxon signed‐rank test did not reveal significant change for any of the three countries separately. However, median scores decreased from pre‐test to post‐test for all samples (Iceland, 19.5 to 16.5; the Netherlands, 31 to 26; Denmark, 23 to 20).

4. Discussion

Refugees are faced with a plethora of challenges that include creating a secure home in their new environment, adapting to the local culture, and coming to terms with the traumatic experiences that drove them from their homelands. For families with children, parents must also help their children adjust and recover. Raising children in the best of circumstances is demanding; parenting for refugees is even more challenging. The SPARE intervention was developed to help refugee parents support their families as they adjust to life in their arrival country.

In the first phase of development of this SPARE adaptation, findings were based on one group each completed in Iceland and Norway. Attendance was good, parents reported satisfaction with their experiences in group, and they indicated that the material addressed important family needs (Sigmarsdóttir, Arnesen, and Forgatch ²⁰²³). The current study extends that work with seven SPARE parent groups conducted in Iceland, Norway, the Netherlands, and Denmark. These findings provide a broader picture of the SPARE program in Northern Europe as more countries were included and initial outcomes from all of them, which had not been tested before, were presented. As stated previously, the adaptation of SPARE struck a balance between surface and deep level adaptations (see Parra‐Cardona et al. ²⁰¹⁷), along with minor adjustments based on findings from the first phase of SPARE (Sigmarsdóttir, Arnesen, and Forgatch ²⁰²³).

The current findings indicate that SPARE was both feasible and acceptable, thus replicating parents' reports from the earlier study. Overall attendance of roughly 75% is also in keeping with reports from prior studies (Gewirtz et al. 2014; Sigmarsdóttir, Arnesen, and Forgatch ²⁰²³). Parents from all four countries expressed positive attitudes toward SPARE and provided valuable suggestions for making improvements. Combined group outcomes point toward greater reported child adjustment, less parent psychological distress, and more positive parenting practices, particularly in Iceland. Although child and parent outcomes were positive overall, they varied across countries, and sample sizes were small.

Variations between countries merit further exploration to elucidate factors that contribute to these differences and to develop strategies to strengthen positive outcomes. Differences between groups in attendance and participant numbers per group in the four countries might be explained by several factors. Levels of support provided to the groups and communication with parents between sessions differed. In Iceland, participants received reminder phone calls, whereas in the Netherlands, they received text messages via WhatsApp. Different systems at each site were employed to recruit families to the program, and competition with other legally required parenting programs for refugees may have affected recruitment and attendance across countries. In Iceland, no other evidence‐based parenting programs were offered to refugee families. The mandatory requirement for parenting programs for refugee families (as in Norway) could be a factor of influence, and even the size of the country. Our team is further exploring recruitment procedures for SPARE and other programs for refugee families. It has been shown that well‐functioning systems within families' communities contribute positively to participant outcomes (Simenec and Reid 2020).

Despite variations in the delivery of SPARE across countries, parents reported high levels of satisfaction with the program and the parent materials, indicating high levels of acceptance by this group of refugee parents. These results are consistent with other feasibility studies founded on adaptations to the GenerationPMTO intervention (e.g., Gewirtz and Taylor ²⁰⁰⁹; Gewirtz et al. ²⁰¹⁴; Parra‐Cardona ²⁰¹⁹; Wieling et al. ²⁰¹⁵), including the Iceland and Norway SPARE study by Sigmarsdóttir, Arnesen, and Forgatch (2023).

In focus group interviews following SPARE, parents' statements were positive, informative, and in keeping with their satisfaction ratings assessed during the intervention. Overall, parents liked the strategies they learned and shared positive group participation experiences, with some non‐systematic differences between groups. Parents wanted more sessions and more time in each session to discuss their experiences, including cross‐cultural differences in parenting practices. Although the facilitator manual reminded practitioners to include this in each session, it could have been emphasised. This finding aligns with reports by cultural experts that parents tend to be more satisfied and find programmes more helpful when they can safely discuss their own experiences (e.g., Domenech Rodríguez et al. ²⁰¹¹; Parra‐Cardona et al. ²⁰¹⁷). Along similar lines, parents would like facilitators to use examples that are relevant to their culture and the age of their children, which corresponds with findings reported by Parra‐Cardona (2019) regarding an adaptation of GenerationPMTO for immigrant Latino/a families in the United States. In that study, participating parents reported that material directly addressing their challenges and examples helped the most. Finally, some parents wanted the observation exercises to be read by a different voice. We have considered these important suggestions, and some have already been incorporated. Taken together, parents' reflections on SPARE support its feasibility and acceptability, while also providing interesting suggestions for potential changes to the programme.

This study extended the earlier SPARE trial by collecting pre/post parent report data on parenting and child variables. Because the numbers of group participants across countries are small (ranging from 4 to 15), the findings must be viewed as preliminary. When participant findings were combined from all countries, results were in the desirable direction, some statistically significant, with moderate effect sizes in child adjustment, parent psychological distress, and positive parenting practices improvements. This mirrors results of prior studies supporting positive effects of evidence‐based parenting programmes (e.g., Cai et al. ²⁰²²; Forgatch and Patterson ²⁰¹⁰; Van Ee et al. ²⁰¹⁶) and highlights the importance of providing evidence‐based parenting programmes to refugee families.

Contexts within a hosting nation may contribute to bias in parental reports. For example, in Norway, parents reported very low levels of child behaviour problems, well below participants in the other countries and below what would be expected from a group of European refugees (Gewirtz et al. 2022). Such reports may reflect the fact that Norway employs very rigorous child protective service systems (Fylkesnes et al. 2017), a reputation well known among refugee parents and mirrored by participant reports. This context could be holding parents back from reporting problems and motivating them to report their family functioning more positively, which may not reflect their reality. Ultimately, this may prevent refugee parents and children from receiving much needed support and thereby jeopardise their future prospects. Thus, it will be important for communities and organisations supporting refugee families to create conditions in which parents and children feel comfortable enough to be forthcoming about their experiences. In this regard, link‐workers may play a critical role; as they are in direct contact with parents and can help to establish a sense of safety.

Relatedly, measurement of parenting and well‐being among refugees and other minoritised populations needs to be psychometrically sound and culturally sensitive. Although the measures used in the present study have been validated psychometrically within various populations (e.g., Abrahamse et al. ²⁰¹⁵; Axberg et al. ²⁰⁰⁸; Kjøbli et al. ²⁰¹²; Strand et al. ²⁰⁰³; Tambs and Moum ¹⁹⁹³; Tømmerås et al. ²⁰¹⁸), it is unclear to what extent they are sensitive to refugee families' contexts specifically. For example, due to the highly positive baseline ratings by participants in Norway noted earlier, and the small Dutch and Danish samples, detecting significant changes in outcomes from pre‐ to post‐ SPARE proved to be difficult, except in Iceland, which carried positive effects for the entire sample. These are important considerations for future research on SPARE and related programmes.

Other differences include discrepancies in the number of sessions completed in each group, parents' attendance, and the child's initial level of problems. A valid measure of fidelity to delivery of the program's content and process is essential. In the present study, therapists self‐rated their adherence to the program with a wide range in scores, varying from the lowest score (1) to the highest (5). Future studies might consider evaluating fidelity more rigorously (e.g., by using the Fidelity of Implementation rating system; FIMP; Sigmarsdóttir, Rodríguez, et al. ²⁰²³). Although the current study design and sample size do not allow for examination of potential moderators of outcome (e.g., length of stay in arrival country, child age, parent/child levels of difficulties at baseline, etc.), these data indicate that it is important to deliver the full program with high fidelity and to train parents in using all its core components.

4.1. Limitations

In this phase of the work, we emphasised creating and adapting the program for Arabic‐speaking refugees in Europe. Resources for data collection were scant, which created challenges. Not all countries reported on all measures, and there were several missing items on questionnaires in some countries. To ensure uniformity in data collection procedures across countries and groups, future studies will follow a data collection manual to strengthen the validity of the outcomes. Another limitation concerns low participant numbers in some countries, suboptimal attendance in some sessions, and some SPARE sessions that were not conducted for some groups. Employing a principal project manager to support and supervise project coordinators at each site and facilitate troubleshooting might help overcome these issues. The small sample size and low attendance in some groups limited generalisability and statistical power and may have affected the range of reliability coefficients across sessions, combined with the low number of items on some of the satisfaction measure subscales (i.e., Parents' and Home Practice Satisfaction). Additionally, the low internal consistency of the inconsistent discipline subscale of the PPI, and thus its omission from further analyses, restricted the scope of the present findings. Finally, the potential for bias in parental responses because of issues about anonymity serves as a limitation. For example, although the satisfaction questionnaire was translated from English into Arabic, link‐workers had to read items and write answers for parents who were not literate. The same anonymity concern also could have affected the focus group interviews, which were conducted by trained students and translated by the link‐workers. When full anonymity is impossible, refugee parents might be prone to responding positively toward the intervention, following expectations, and avoiding criticism.

4.2. Conclusion and Future Directions

Evidence‐based parenting programs for refugees in Europe are needed (Gewirtz et al. 2022). If implemented with fidelity within well‐established social and community systems, effective programs supporting the health and well‐being of refugee families could facilitate their inclusion in new surroundings and mitigate the effects of traumatic experiences. The current findings lend further support to the promise that the SPARE program holds for Arabic‐speaking families relocating to Northern European countries. The program could benefit from devoting more time in every session discussing challenges refugees meet in the new country and by providing examples more relevant to the participants.

Possible future trials evaluating the effects of SPARE are pertinent, requiring a high degree of implementation fidelity, culturally sensitive measures, and robust recruitment and implementation supports. We will continue to develop the program and increase its relevance based on participating parents' suggestions, as well as current and future research findings. Importantly, given the feasibility and acceptability of SPARE, further investigation and adaptations for other refugee populations, including but not limited to Ukrainian families resettling in Europe, are clearly warranted.

Author Contributions

Margrét Sigmarsdóttir led the work, coordinated the writing team, collaborated with sites, designed the study with other authors, and was lead in the writing process. Anne Arnesen carried out statistical analyses, participated in designing the study and the writing process. Anett K. Apeland participated in the writing process and oversaw qualitative data analysis. Bergljót Gyða Guðmundsdóttir participated in statistical analyses and in the writing process. Ella Tacq participated in the writing process and collected data. Helge Hallmann participated in the writing process and collected data. Marion S. Forgatch participated in designing the work and in the writing process.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: Supporting Information.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.