The study was a process evaluation that was conducted in parallel with a randomised controlled trial. The process evaluation was guided by the framework developed by the Medical Research Council Guidance by Moore et al. . Three components, context, implementation and mechanism of impact, were used to describe the context for the implementation, the way in which the implementation strategy was carried out and potential mechanisms for achieving implementation (Table 1).
The study was conducted in a medium-sized municipality in Sweden with a population of 51,000. By the end of 2017, 17% of the municipal population was foreign-born , with Somalis making up one of the largest immigrant groups. The municipality’s social services had experienced considerable difficulty reaching out to immigrant parents and were therefore determined to find ways to engage hard-to-reach groups in parenting programmes . Thus, this study was conducted together with the social services.
The Ladnaan Programme has two components: the Connect parenting programme  and societal information (Figure 1). The Ladnaan programme was delivered four times to four separate groups, and each programme had 12 sessions, which were attended by 12–17 participants and were delivered by two group leaders (one female and one male).
Connect is a 10-session parenting programme derived from attachment theory . Its primary aim is to promote children’s mental health and strengthen the parent-child attachment relationship . The Connect programme was chosen based on findings from previous research  showing that Somali-born parents were expressing a need to strengthen their relationships with their children. Each Connect programme session lasted one hour, and each societal session lasted 2–3 hours. The Connect programme comprises 10 sessions that are based on nine attachment principles: 1) All behaviour has a meaning, 2) Attachment is for life, 3) Conflict is a part of attachment, 4) Autonomy includes connection, 5) Empathy – the heartbeat of attachment, 6) Balance our needs and the needs of others, 7) Growth and change are part of relationship, 8) Celebrating attachment, and 9) Two steps forward, one step back (see Figure 1). In each session, parents are introduced to one attachment skill. The session is presented through reflections, case examples, role plays and exercises in which group leaders illustrate attachment principles in-depth. After each Connect session, parents receive translated and summarised hand-outs about the session.
The societal information component comprised two of the 12 sessions of the Ladnaan Programme and was developed from a previous study on the need among Somali immigrant parents for parenting support . The societal information component of the current programme contained three parts: 1) Child Welfare Services, which aimed to provide parents with an overview of Swedish Child Welfare Services; 2) Parenting Styles, which was aimed to introduce different parenting styles, such as democratic and authority parenting; and 3) the United Nation’s Convention on the Rights of the Child (CRC), which aimed to provide parents with knowledge about the international human rights of children and the promotion of children’s agency. The societal information sessions were delivered as workshops in which group leaders gave a short introduction of the theme and proceeded with discussions and questions from the parents.
All group sessions for this study were conducted in Somali except for the single session on parenting styles that was delivered by a Swedish-speaking professional from Family and Child Welfare Services and interpreted by one of the group leaders. The programme was delivered with a culturally sensitive approach by group leaders who had cultural competence relating to both Swedish and Somali society. The culturally sensitive approach was developed from the findings of a previous study  and involved tailoring the parenting programme to the parents’ needs, culture and context. The programme was named Ladnaan (a Somali word referring to a sense of health and wellbeing), and the term “parent empowerment” was to reflect the fact that the programme was meant to empower the parents rather than control them. In addition, Somali metaphors and proverbs were used during the sessions to make the content comprehensible and to affirm certain content.
The implementation strategies focused on the implementation of the Ladnaan Programme, the recruitment and training of the group leaders and lecturers, and the reaching and retaining of parents.
Recruitment and training of group leaders
The main task for group leaders was to deliver the Connect sessions. Initially, nine group leaders (five males and four females) with a Somali background who were employed by social services, the integration office for newly settled refugees and schools were recruited to ensure that the parenting programme would be sustainable even after the study had concluded (Table 2). Mixing the gender of group leaders in each session (one female and one male group leader) aimed to attract both mothers and fathers to the programme. The group leaders received four days of standardised Connect training provided by Connect instructors qualified to train group leaders .
Two of the group leaders and one lecturer from the municipal Family and Child Welfare Services department delivered the societal information sessions.
Strategy for reaching and retaining the target group
Strategies for reaching and retaining parents included the following components: 1) informal information meetings (held in small groups and at the individual level); 2) diploma for parents who participated in the parenting programme; 3) times for meetings adapted to the needs of the parents; 4) session venue in the neighbourhood where most families lived; 5) reminders; 6) babysitting services; and 7) assistance with reading and answering letters from government agencies.
Parents first received information about the parenting programme at informal meetings held in Somali associations and schools and at the Workers’ Educational Association. The meetings were arranged together with the Somali associations and key individuals (Somalis from the municipality who were well respected within the Somali community). Parents who were interested in receiving more information were phoned by the external facilitator (member of the research group), who gave them detailed information about the parenting programme.
The key individuals and the external facilitator suggested that the awarding of a diploma to parents could be a way to facilitate recruitment; therefore, parents were informed that they would be given a participation diploma if they completed the parenting programme. The sessions were held late afternoon in a neighbourhood where most of the parents lived, at a time and place chosen by the parents.
Text-message reminders were sent to parents by internal facilitators the day before each session. Childcare was offered to parents who required it so as to increase the chances of both parents attending the sessions. Parents were also offered assistance with reading and replying to letters from authorities. According to the Somali associations, many parents in the community often sought their assistance to read government letters.
Involved in the study were group leaders, a lecturer, Connect instructors and internal facilitators who participated in the implementation process (Table 2), as well as parents who participated in the parenting programme. The participants involved in the implementation of the programme included eight females and six males. Their roles and characteristics are described in Table 2.
The parents who participated in the parenting programme (n=58) were between 30 and 70 years of age and had children aged 11 to 16 years. The majority (65%) of the parents had lived in Sweden for <5 years.
For this process evaluation study, both qualitative and quantitative data were collected and analysed . The combination of qualitative and quantitative data allowed for a focus on different aspects of the implementation of the programme. Data were collected during the delivery of the four rounds of the programme and after its completion.
Ethical approval for this study was obtained from the Swedish Regional Ethical Review Board in Uppsala, Sweden (Dnr 2014/048/1). All participants were informed about the aim of the study, that their participation was voluntary and that data would be treated in a manner that respected their confidentiality.
Qualitative methods of assessment
The qualitative data were collected through focus group discussions (FGDs), field and reflection notes and observations of the societal information sessions. A semi-structured interview guide was used in the interviews, which was tested with the first FGD. No further revisions were made on the interview guide. Six FGDs were conducted with group leaders, a lecturer, Connect instructors and internal facilitators. The FGDs with the group leaders, lecturers and internal facilitators concerned their experiences with delivering the intervention, including their contact and interaction with participants, and contextual factors that affected the implementation. The FGDs lasted from 60 to 105 minutes. Five FGDs were conducted in Somali and one in Swedish.
To capture the fidelity of the Connect programme, each session was video-recorded, and the recordings were sent to the Connect instructors. For this study, we did not analyse the video recording but held focus group interviews with the instructors who had been supervising the group leaders each week throughout the sessions. Connect instructors were asked about their experiences of delivering training and supervising the group leaders, as well as about the fidelity of the programme. The FGDs with the instructors were conducted in Swedish and lasted approximately 60 minutes.
Moreover, to capture the fidelity of the societal information sessions, two observations and reflection notes from three leaders of the societal information sessions were collected. The observation and reflection protocols contained similar information and were about overall impressions and interactions with the parents, the most common participant questions and session duration.
Field notes were taken during the implementation of the parenting programme to examine the contextual factors that facilitated or hindered the implementation.
Quantitative methods of assessment
The quantitative data included a questionnaire and an attendance list. The questionnaire measured parents’ satisfaction with the intervention. The Consumer Satisfaction Questionnaire (CSQ) was used . The questionnaire was administered to participants two months after intervention completion. The CSQ was a standardised instrument which was adapted for this study, meaning that the questions focused on the current intervention. The questionnaire was also translated into Somali in accordance with the five steps of translation process : 1) forward translation, 2) backward translation, 3) expert panel back-translation, 4) pre-testing and 5) final version. The CSQ comprised 22 items addressing the impact of the parenting programme on the parents’ knowledge, confidence with parenting, and parent-child interactions and relationships. Parents were asked to rate each item on a scale from 1 (e.g. satisfaction with the programme or improvement of problems or parent-child relationships) to 4 (dissatisfaction with the programme or worsened problems or parent-child relationships). The 22 items were summarised to yield a total score between 22 and 88, with low scores indicating higher satisfaction. Dose was assessed using the attendance list.
All FGDs were transcribed verbatim and analysed inductively with content analysis . This method was chosen to understand how the phenomenon (i.e. implementation strategies) was perceived and experienced by the participants. The transcribed FGDs with the Connect instructors were analysed separately. All transcribed data were read by the first author several times to capture the participants’ overall experiences. Phrases, paragraphs and words that captured the key concepts were then highlighted and coded into initial codes. The next step involved sorting codes based on their similarities. Co-authors reviewed the coding, and when consensus was achieved, the codes were then sorted according to the components of the process evaluation guidance.
The field notes, observations and reflection notes were analysed using deductive content analysis . The analysis started with a reading of the text and then placed the text according to the components of the process evaluation guidance.
The CSQ was analysed in SPSS version 24 . Descriptive statistics (frequencies and percentages) were calculated to describe the data.