Experiences of parents whose children participated in a longitudinal follow‐up study

Abstract Background Long‐term follow‐up is necessary to understand the impact of perinatal interventions. Exploring parents' motives and experiences in consenting to their children taking part in longitudinal studies and understanding what outcomes are important to families may enhance participation and mitigate the loss to follow‐up. As existing evidence is largely based on investigators' perspectives using Western samples, the present pilot study explored parents' perspectives in a multicultural New Zealand context. Methods Data were generated using semi‐structured interviews with parents whose children had participated in a longitudinal study after neonatal recruitment. Parents' experiences of being part of the study were analysed thematically using an inductive approach. Results Parents (n = 16) were generally happy with the outcomes measured. Additionally, parents were interested in lifelong goals such as the impact of parental diabetes. We identified three themes: (1) Facilitators: Research participation was aided by motives and parent and research characteristics such as wishing to help others and straightforward recruitment; (2) Barriers: A hesitancy to participate was due to technical and clinical research aspects, participation burden and cultural barriers, such as complex wording, time commitment and nonindigenous research and (3) Benefits: Children and parents experienced advantages such as the opportunity for education. Conclusions Parents reported positive experiences and described the unexpected benefit of increasing families' health knowledge through participation. Improvements for current follow‐up studies were identified. Different ethnicities reported different experiences and perspectives, which warrants ongoing research, particularly with indigenous research participants. Patient or Public Contribution No active partnership with parents of patients took place.


| INTRODUCTION
Long-term follow-up is necessary to understand the impact of perinatal interventions but can be a challenge both in terms of recruitment and retention. Systematic reviews of retention strategies in longitudinal cohort studies revealed that reminder letters, resending questionnaires and offering incentives were the most successful specific tactics, 1 while decreasing the burden on participants was the most effective general approach. 2 However, studies investigating retention strategies are based on investigators' perspectives, rather than on those of research participants themselves.

| Design
This qualitative study used a thematic analysis approach 3 and adhered to the consolidated criteria framework (COREQ). 4

| Recruitment
Participants were parents of children who took part in the Children with Hypoglycaemia and their Later Development (CHYLD) study; a prospective cohort study investigating outcomes of children born at risk of neonatal hypoglycaemia. Children in the CHYLD study were assessed at birth, at 2, 4.5 and 9-10 years. [5][6][7] Assessments included measures of growth, vision, brain structure (magnetic resonance imaging [MRI]), cognitive and language development, executive function, academic achievement and psychosocial adaptation. After each wave of data collection parents received a summary of the assessment findings. If the health or developmental assessments indicated children needed further assessments, a letter of referral was sent to their general practitioner with parental consent. A summary of the overall study findings at 2 and at 4.5 years was also sent to participating families at the time these were published. No other interaction between the research team and parents took place between the assessments. A purposive sample of parents whose children participated in the CHYLD study was invited to take part after their children had completed the 9-10-year assessment.
Sampling aimed to ensure the inclusion of a range of ethnicities with a specific focus on Māori, the indigenous population of New Zealand. Potential participants were approached by email, phone or in person.

| Data collection
Participation options for the semi-structured interview included inperson, online face-to-face or via phone, either individually or in a group setting (i.e., focus group discussion). We invited parents to share their views about being part of the CHYLD study, focusing on what they liked and did not like, how being part of the study had affected them and which factors they perceived to be facilitators or barriers to participation. This discussion was guided by a list of topics, which was developed by the authors. Individual interviews were conducted by N. F. and J. R. (no parents opted for group discussion) between March and July 2020 and took between 15 and 45 min. Some data collection took place during the COVID-19 lockdown in New Zealand and all public health precautions were followed.
Respecting the indigenous value of whanaungatanga (kinship), interviews with Māori parents were conducted by J. R., who is of Māori descent (Ngāi Tahu) and who has been involved in recruiting these families to previous data collection waves. All other interviews were facilitated by N. F., who has a background in family psychology, and both N. F. and J. R. had prior experience in conducting and analysing interview data. Interviews were audio-recorded and transcribed verbatim and checked for accuracy by N. F. No notes were taken during the interviews to encourage a natural conversation.

| Data analysis
Qualitative data were analysed thematically to generate an authentic account reflecting parents' experiences with longitudinal perinatal research and to identify factors that may influence participation.
The first step comprised reading the transcripts thoroughly and repeatedly, in line with an inductive approach. 3,8 The content of the data directed coding and theme development. Initial coding was followed by a grouping of the data, using QRS NVivo, to enable the identification of common themes. N. F. and J. R. reviewed and discussed data with all other authors to clarify and refine themes as analysis progressed and reached consensus regarding relevant extracts for each theme. The resulting themes and subthemes were refined and supported by quotations from parents' transcripts.

| Validity and reliability
We took several steps to ensure the validity and reliability of the study findings. First, purposive sampling was employed to ensure a range of ethnicities to enhance the transferability of the results. Second, J. R.
conducted the interviews with the Māori participants, with whom she had a previous research relationship. Third, regular discussions amongst authors were held while interpreting the results to minimize researcher bias. Fourth, the inductive approach to analysing the data involves applying a set of systematic procedures, which support the validity and reliability of the study. 8 Fifth, we used quotations to support our findings, thereby allowing readers to review interpretation.

| RESULTS
In total, 15 mothers and one father took part in the semi-structured interviews. There were seven Pākehā (European descent) families, five Māori, three Indian and one Pacific Island family from a range of socioeconomic backgrounds (deciles 2-10 of the 2013 New Zealand Index of Deprivation). 9 Participating children were aged between 9 and 13 years. See Table 1 for an overview of participants' demographic characteristics. Further details of recruitment rate and reasons for nonparticipation are reported in the COREQ checklist (Appendix A). All parents opted for an individual interview about their experience of participating in perinatal research, which mostly took place via video calling (n = 8), followed by phone (n = 4) and in-person (n = 4). Parents were also asked which child health and wellbeing outcomes they were interested in learning more about, in addition to those tracked by the CHYLD study. Thematic analysis yielded three themes involving facilitators, barriers and benefits of research participation ( Table 2).

| Overview of outcomes of interest
Parents described being satisfied with the information that they received about the outcomes measured, and identified guidance

| Burden of participation
In the context of the stress involved with dealing with the potential health In the Māori-diverse world, in our world, we are safe.
Everything that we do, our practices and our techniques, keeps us safe. Therefore, allowing any European perspective in is not safe. So, you have to be courageous to take a chance on science. Because it is not our world, it doesn't belong to us and it's not in our genealogy.

| Benefits for parents
Parents considered the regular assessments reassuring, as described by  Authors of a birth cohort study recommended the inclusion of altruism as a recruitment focus, especially in studies with few perceived benefits. 13 Ostensible altruistic motives found in our study (i.e., helping others), may have originated from parents' desire to reciprocate for care that has benefitted their child. Indeed, our findings around the importance of engagement and connections imply that reciprocity is of great significance to families. Therefore, it may be more helpful to promote a study's social and altruistic benefits. 14 One parent's wish to use CHYLD research findings for education purposes further corroborates the importance of social benefits. Personal motivators found in the current study included the desire to serve the interests of other family members and the opportunity to learn, but monetary compensation was seen as an unexpected benefit, contrasting with a prior study where financial gains were a major motivator. 15 Informed consent and awareness of the study's objectives were variable and four parents were unaware that blood tests were standard hospital procedure. As stated by policies that govern research with human participants, the consent process should focus on comprehension of what study participation entails, especially when studies intersect with clinical care. 16 Detailed consent forms that promote participants' interests and rights appear to be aimed at protecting the research institution 17 rather than increasing participants' understanding. 18 Indeed, lengthy and technical study information sheets were mentioned as a barrier to informed consent, suggesting that complex study information is not conducive to autonomous informed consent. 16 A potentially effective strategy to reach informed consent is to provide prospective participants with reasons for and against study participation, 19 an approach employed by one parent in the pilot study who discussed what participation entailed with a family member whose child was also enroled. Therefore, informed consent should take into account participants' cultural background, including having research staff of the same ethnic group as the participant, 23 and may require seeking community-level informed consent before obtaining individual consent. 24

Presence of nonparticipants:
Typically, only the participant and interviewer were present. In some instances, other family members of the participant, such as a partner or young child were also present.
16. Description of sample: There was one father and 15 mothers. Ethnicities included Māori (n = 5), NZ European (n = 4), Indian (n = 3), Samoan/NZ European (n = 1), Welsh/Zimbabwean (n = 1), South African (n = 1), Scottish (n = 1). Families from eight urban and eight rural locations were included and the age of the children at the time of data collection ranged from 9 to 13 years.

Interview guide:
All interviews were guided by a set of questions and prompts.
During in-person interviews, participants were given a copy of the questions and prompts. The current project is a pilot project and will inform the next phase. 30. Data and findings consistent: There was consistency between the data presented and the findings.
31. and 32. Clarity of major and minor themes: A distinction was made between major and minor themes.
Subthemes were presented to illustrate differences between participants' experiences within each major theme.