What affects programme engagement for Māori families? A qualitative study of a family‐based, multidisciplinary healthy lifestyle programme for children and adolescents

Aim It is important that intervention programmes are accessible and acceptable for groups most affected by excess weight. This study aimed to understand the barriers to and facilitators of engagement for Māori in a community‐based, assessment‐and‐intervention healthy lifestyle programme (Whānau Pakari). Methods Sixty‐four in‐depth, home‐based interviews were conducted with past service users. Half of these were with families with Māori children and half with non‐Māori families. The interviews were thematically analysed with peer debriefing for validity. Results Māori families experienced barriers due to racism throughout the health system and society, which then affected their ability to engage with the programme. Key barriers included the institutionalised racism evident through substantial structural barriers and socio‐economic challenges, the experience of interpersonal racism and its cumulative impact with weight stigma, and internalised racism and beliefs of biological determinism. Responses to these barriers were distrust of health services, followed by renewed engagement or complete disengagement. Participants identified culturally appropriate care as that which was compassionate, respectful, and focused on relationship building. Conclusions While Whānau Pakari is considered appropriate due to the approach of the delivery team, this is insufficient to retain some Māori families who face increased socio‐economic and structural barriers. Past instances of weight stigma and racism have enduring effects when re‐engaging with future health services, and inequities are likely to persist until these issues are addressed within the health system and wider society.

1 Indigenous peoples face increased barriers to accessing health care due to the ongoing effects of colonisation. 2 While Indigenous groups and other ethnic minority groups tend to have lower retention rates in healthy lifestyle programmes addressing childhood obesity, previous research has not examined the underlying factors or simply cited 'cultural inappropriateness' of the programme without elaboration. 3 Racism is a known determinant of health and driver of inequities for Indigenous peoples.
What this paper adds 1 Engagement in multidisciplinary healthy lifestyle programmes is affected by experiences of institutionalised, interpersonal and internalised racism for M aori families. 2 Racism which occurs elsewhere in the health system or in wider society may have ongoing effects on subsequent engagement with other health services, and until these issues are addressed, inequities in service engagement between Indigenous and non-Indigenous groups are likely to persist. 3 Compassionate, respectful, appropriate and culturally safe care focusing on positive relationship-building can help mitigate some of the impact of racism on engagement in prevailing health care services.
of children aged 2-14 years. 2 [Correction added on 23 April 2021, after first online publication: The ratio of obesity rates in Pacific children compared with the overall population has been changed from 3 to 4.7 times.] Children living in the most socioeconomically deprived areas are also twice as likely to experience obesity than those living in the least deprived areas. 2 These findings are consistent with international inequities between Indigenous and non-Indigenous populations. 1 Multidisciplinary intervention programmes remain recommended best practice for addressing childhood obesity, 3 and retention is key to positive health outcomes. 4,5 However, previous research also demonstrates inequities in attendance between different ethnic groupsa systematic review of 'barriers' and 'facilitators' to participation within paediatric weight management programmes showed higher dropout rates among Black participants and those on a low family income than White participants in the USA. 6 However, few studies have explored the underlying reasons for differences in attendance among Indigenous and other marginalised ethnic groups, or have cited 'cultural inappropriateness' of the programme as the reason for disengagement, without elaboration. 4,7 Indigenous peoples face increased barriers to accessing health care. 8 Racism is a known determinant of health and driver of inequities for Indigenous peoples, and is increasingly acknowledged as shaping health care interactions and outcomes. 9,10 Institutionalised racism, as an ongoing effect of the colonisation of Indigenous peoples world-wide, is reflected in differential access to social, political and economic resources and poorer health and social outcomes. It can operate without identifiable individual perpetrators, 11,12 and may manifest in poorer material conditions. Interpersonal racism refers to personally-mediated discrimination and may appear in the health system in a more covert and passive form, manifesting as implicit bias, a belief or association about a social group that may be automatic. 13 Lastly, internalised racism refers to the acceptance and internalisation of racial assumptions and stereotypes by groups themselves in society. 9,10,12,13 In NZ and other countries with similar colonial histories, these varying forms of racism are entrenched in the context of colonisation and the historical and ongoing processes of dispossession and marginalisation. 14 This has resulted in ingrained structural inequities within the health system that give rise to inequitable outcomes across a range of areas, such as access to primary health care. M aori report a higher prevalence of racial discrimination and are also more likely to experience multiple forms of discrimination compared with NZ Europeans. 15 Racism is known to be associated with both poorer health outcomes and reduced access to health care and resources. 16 Under the Treaty of Waitangi, and reinforced by commitments to international conventions such as the United Nations Declaration on the Rights of Indigenous Peoples and the United Nations Convention on the Rights of the Child, health professionals and researchers in NZ have an obligation to address health inequities for M aori. 17 One of the guiding principles of the NZ Health Strategy is equitable access to health services, 18 to improve health outcomes for those most affected by conditions such as obesity.
Wh anau Pakari is a family-based intervention programme for children/adolescents with weight issues, with a multidisciplinary team including a paediatrician, physical activity specialist, dietitian, healthy lifestyles coordinator, and a clinical psychologist. 19 It aims to be non-stigmatising, with a focus on healthy lifestyle change rather than weight or obesity. Programme conception and consultation involved key M aori community stakeholders in the region and is ongoing. 19 The programme has replaced the conventional hospital-based model of care within the prevailing health care model in Taranaki, NZ, in order to provide more accessible, appropriate health care. The results of a randomised clinical trial embedded in the service found that participants achieved a modest reduction in body mass index standard deviation score, as well as achieving positive changes in health-related quality of life and cardiovascular fitness. 5 Greater reductions in body mass index standard deviation score were achieved if participants attended ≥70% of programme sessions. 5,20 While service recruitment included approximately equal numbers of M aori and NZ European participants, M aori were less likely to attend the recommended optimal number of sessions. 5 A previous paper describes the experiences of Wh anau Pakari participants with varying levels of attendance.
In NZ, health research should be responsive to the needs and diversity of M aori. 17 The approach of this research team was based on Kaupapa M aori theoretical principles. 17,21,22 It was aligned with social and structural determinants of health approach and developed with the aim of contributing towards the elimination of health inequities for M aori, resisting persistent power imbalances and the continued use of cultural deficit theory to explain inequities between M aori and non-M aori, which attributes poor M aori health to M aori 'culture' or something inherent to M aori as a group. 17,22 The aim of this paper was to explore the experiences of M aori families referred to Wh anau Pakari. Given the suggested relevance of 'cultural appropriateness' by previous studies, 4,7 we also sought to determine whether Wh anau Pakari was perceived to be appropriate for M aori.

Methods
The method for conducting the interviews has been described previously. 21 Briefly, authors 1 and 2 undertook 64 in-depth interviews in the home with past participants of Wh anau Pakari of varying levels of engagement, with equal numbers of interviews with families with M aori and non-M aori children (Table 1). Interview participants were mainly parents, grandparents and/or caregivers of children involved in Wh anau Pakari (five children/adolescents themselves participated in the interviews) and were from a range of socio-economic backgrounds (deciles 1-10 of the 2013 NZ Index of Deprivation). 23 Interviews centred on families' experiences of engaging in Wh anau Pakari and the wider health system. Interview recordings were independently transcribed, and participants were offered their transcripts to check for accuracy and acceptability. Participants were offered a koha (gift, contribution) as a sign of reciprocity for their time and for sharing their experiences.
The transcripts were coded and analysed in MAXQDA software by author 1 using thematic analysis, identifying common patterns across participants as well as any potential differences between groups. All authors collaborated to finalise the themes and develop the framework, with agreed respectful parameters allowing the authors to debate, challenge and refine interpretations of the data. The researchers agreed to apply the 'Give-Way' rule if there was disagreement over the interpretation of the data concerning M aori participants, and the final decision involving cultural interpretation of M aori participants' experiences would pass to a M aori researcher. 22 Dissemination of study findings to participants was via a summary feedback video, which was preferred by participants over a written report or feedback hui (meeting). Ethical approval for the study was obtained from the Central Health and Disability Ethics Committee (New Zealand) and all study participants gave written consent.

Results
This study's findings aligned with Camara Jones' framework for understanding racism and its impact on health service engagement. 12 The core themes identified in the results reinforce this framework and have been grouped accordingly: Institutionalised Racism; Interpersonal Racism; and Internalised Racism ( Table 2). The extent of programme engagement for M aori participants appeared to be related to the experiences of and responses to racism. This study also identified what was considered culturally appropriate care for participants.

Institutionalised racism: Substantial structural barriers and socio-economic deprivation
Institutionalised racism was evident through the wide range of adverse events affecting many M aori participants and their families, which affected their capacity to engage with the service and the wider health system. M aori participants and their families frequently reported both acute and chronic life stressors and difficult socio-economic conditions (Table 2).
We didn't have a vehicle. And we were going to car-pool with my aunty who had to take her son as well. And, um, she didn't have enough for all of us, you know. I just felt shit that I couldn't take him […] Unless I hitch hike with all of them.
Systemic racism was demonstrated through the range of experiences evident within single families. For example, M aori or Pacific participants who were socially assigned by other people as NZ European experienced racism differently in the health system and wider society. Participants acknowledged that they were accorded societal advantage or disadvantage depending on how they were socially assigned by others.
Um, when the two boys were going to school, cause one was dark and one was white, they used to get teased quite a bit because they didn't think they were brothers and that's really upsetting because why should they be different because one's dark and one's white, you know, because [SON 2] actually identifies with his culture just as much as [SON 1] does, and it's almost like he's worked twice as hard whereas [SON 1]'s dark and he doesn't care, you know […] I know a lot of people who name their children, um European names so they get ahead, and they are still do that today and it's really upsetting if that happens and disappointing.
Interpersonal racism: Cumulative effect of weight stigma and racism While weight stigma was experienced by both M aori and non-M aori participants, M aori participants reported compounding effects of weight and race stigma within their interactions with health professionals and felt targeted because of their weight and being M aori (Table 2). When asked about discrimination, one participant emotionally recounted her reaction to being negatively stereotyped by a health professional: Much of the interpersonal racism experienced by participants both within and outside the health system was centred around implicit bias and stereotyping, as a result of systemic racism. Participants were sometimes reluctant to articulate this as racism and instead described experiences of being judged or treated unfairly in the community (Table 2).
… he said "Mum, I'm the only brown boy at school." I said "no you're not. I'm sure there's another one." […] he said "oh no, [classmate] is brown too, that's right," and I said "see, you're not the only one," but there's some kids there, white kids, who, I'm just assuming are a little bit racial. That might be because of the way their family, you know, obviously it's how their parents kind of probably are, and I know that their parents possibly don't have brown associations. Sixty-four interviews total, 11 interviews involved 2+ family members.
Half of the interviews were with families with M aori children. Five interviews included a child/adolescent participant in addition to their parent/caregiver. Maximum total of 74 potential interviews for funding and resource reasons. A total of 136 families approached, of which 53 were uncontactable, 7 were living out of the region, and 12 declined (see COREQ checklist for reasons). ‡ Total ethnicity output (more than one ethnicity selected). Twelve participants identified with more than one ethnicity. § Twenty-four families invited total. ¶ Forty-two families invited total. † † Fifteen families invited total. ‡ ‡ Fifty-five families invited total.
However, participants also identified more explicit instances of racism occurring within and outside the health system (Table 2), resulting in suspicion of a range of government-provided services in addition to the health system (Table 3).
Cops [police] may judge us, but they won't even know that they're judging us. Like, really in their intentions or hidden agendas, they judge us because of what I look like. They're driving, but really, they're going to pull you over. 'Oh, we're just doing a random check'. I've done it, sort of driven past cops with this [hat and hooded jumper] on and then when you drive back past them with no hat and they don't even look at us. Yeah, my mate… I was driving with him and he, he saw a cop coming and he goes 'bro, take your hat off, take your hat off' and I go 'aye, why is that'? 'Coz some cops are coming, take it off', and he took it off and then it made a differencecop didn't look at us.

Internalised racism and biological determinism: Perception that ancestry determines outcome
Many participants with M aori children self-attributed overweight and obesity to family genetics, or more broadly to their ancestry (Table 2). There was a perception that M aori and Pacific peoples are 'naturally big' but that this was not necessarily a concern. This perception of children being 'born big' or 'solidly built' was evident across both M aori and non-M aori families, but it was  These internalised ideas shaped attitudes about genetic propensity towards overweight and obesity, the perceived potential effectiveness of healthy lifestyle change, and therefore the perceived value of engaging with healthy lifestyle services. Additionally, the internalisation of racist stereotypes highlights the relationship between external racism and internalised identity ( Table 2).
…There are a lot of us that, yeah, there are a lot of bad people that are brown and that too so the good ones that are brown will actually have it all too because that's what happens I suppose.

But we do get judged, but I don't let that beat us though […]
Although not directly linked to health system use, the experiences of these participants are the result of the ongoing effects of both believing and challenging internalised racism, which often led to a distrust of and disengagement with services (Table 3).
I think that's why I, in the end, I don't know if I could trust them [hospital service] with any help so I don't go to them anymore.
What is 'appropriate' care?
Despite the past negative incidents of care many participants had experienced, participants were clear about what constituted 'good care'. Care that was respectful, compassionate and dignified appeared to overcome some of the past negative experiences. Many participants reported receiving this type of care in Wh anau Pakari, as well as in other areas of health care. When asked directly, participants reported that Wh anau Pakari was culturally appropriate, and responses tended to be centred around the positive relationships developed with the delivery team rather than any tangible aspect of the service itself.
Do you mean culturally? Yeah of course or I wouldn't have been… we would have opted out otherwise. Yeah, we wouldn't have been going, we'd have been making up all these excuses, well I know I would. You would have been 'hey, oh just don't worry about going today, I'll say something', that sort of thing. Yeah, nah, I did, I really enjoyed it. Like I say, the staff, they were awesome… Our experience with you guys was awesome, and with everyone else. Those are pretty straight up answers.
Participants also spoke of respectful care they had received elsewhere in the health system. Likewise, it centred around the compassion and kindness of health-care professionals, irrespective of their ethnicity.
Yeah, no, the M aori lady… get another M aori to approach another M aori aye… yeah she kind of became someone that we could turn to if we needed anything, which was awesome, but she'd also throw some of it back in our face as well, where it got to the point where I was 'I don't think I need you actually'. Aye, I think I might go to this side of the fence where I've got an awesome um white lady who is prepared to sit there, listen and help, you know, help me and my daughter big time.
This data demonstrates positive experiences with relationshipbuilding in the health system and are examples of how respect and compassion are central to appropriate care.

Discussion
This study found that the engagement of M aori families in Wh anau Pakari was influenced by the effects of institutionalised racism, manifesting as socio-economic deprivation and other differential barriers to access, and interpersonal and internalised racism and stigma experienced throughout the health system and society. Responses to these experiences included disengagement from Wh anau Pakari, regardless of where the experience occurred. However, respectful and culturally appropriate care with an emphasis on positive relationship-building may be a way to partially mitigate and resist past experiences of racism and weight stigma. Our data suggest that at the system and service level, health care that practices manaakitanga (the process of showing respect, support, and care for others) and aroha (love, compassion, empathy, kindness) as guiding principles can promote engagement. However, this is not always sufficient to retain families who are dealing with multiple complex challenges while also being faced with many barriers to making healthy lifestyle changes. 24 Previous research has also suggested that there are multiple factors hindering engagement external to the service provision; 21 this study reinforces the need to address wider social issues to enhance engagement and retention in services such as Wh anau Pakari.
This study's findings are consistent with the limited previous literature on barriers to engagement in lifestyle interventions for Indigenous and other marginalised ethnic groups, as well as literature on health care access more generally which commonly identifies socio-economic factors and racial discrimination as key barriers for Indigenous peoples. 8,25 The high rates of deprivation for M aori as a result of institutionalised racism are likely to contribute towards differential access to health services such as Wh anau Pakari between M aori and NZ Europeans. While 'personal circumstances' are frequently identified as factors influencing attendance in multiple studies, 4 a study of participant retention in a family-led weight-management programme for Pacific children with obesity specifically identified unpredictable life events such as deaths, illnesses and employment changes as key barriers which affected participation. 7 Although some of these stressors are unpredictable, many are the result of household deprivation in participant communities. For example, it is difficult to address weight issues when food insecurity is a bigger threat to health and wellbeing due to the social conditions in which children and their families live. In NZ, almost one in five children live in households that are experiencing moderatesevere food insecurity. 26 These factors are external to a healthy lifestyle service and were consistently identified as determinants of disengagement as families' priorities are forcibly and understandably changed.
While it was not identified as a factor affecting engagement in a systematic review of barriers and facilitators of engagement and retention in childhood obesity interventions, 4 our study demonstrates that previous occurrences of racism have an enduring ability to influence seemingly unrelated interactions within the health system. Racism in NZ has previously been identified as a key determinant of health due to the ongoing effects of colonisation, contributing towards health loss and inequities between M aori and NZ Europeans 9,15 with a 2018 study demonstrating a dose-response relationship between the number of forms of discrimination experienced and negative health impacts. 15 Our study suggests that for some M aori participants, various experiences of racism and weight stigmaregardless of where they occurredaffected subsequent engagement with the health system. Practically, this has important implications for clinical careif participants have had stigmatising or discriminatory experiences elsewhere, this may affect engagement with other unrelated services.
Indigenous-led services are critical in order to mitigate inequities. 27 Ensuring that prevailing or 'mainstream' health-care services are culturally appropriate and reflect the communities they serve is also important, in order to complement Indigenous-led service provision. Previous literature on engagement in healthy life-style programmes has cited a lack of 'cultural appropriateness' as a reason for lower engagement, without further elaboration. 4 In our study, the most important facilitators of engagement for Indigenous groups were not concrete components in a 'check-list' of appropriateness or competency. 25,28 Critically, the acceptability of Wh anau Pakari for all participants, but particularly for M aori, stemmed from positive, respectful, trusting relationships between the programme deliverers and programme participants, as defined by the participants themselves. 21 When participants chose to engage with Wh anau Pakari despite historical experiences of weight stigma and racism, they cited the relationships developed with programme deliverers as the reason this engagement was successful. Previous studies have identified relationships and social connectedness as essential for acceptable services and enabling engagement with Indigenous groups, 25,[29][30][31] especially in healthy lifestyle programmes. 4 This is aligned with the strengths-based concept of cultural safety, which refers to health care delivery with a focus on critical analysis of and reflection on the power imbalances in health care interactions between Indigenous patients and non-Indigenous health care providers. 27,28,32 A context of cultural safety can facilitate appropriate care. 27 Past evaluations of the service have suggested that Wh anau Pakari is considered culturally acceptable. 33 The programme emphasis on general health and wellbeing rather than obesity is also likely to contribute towards the acceptability of the programme for M aori. 29 A strength of this study is the strong representation from M aori participants and those who chose not to engage with the service at all. Our high recruitment rate was due to a concerted effort to overcome common barriers to research participation; this included text reminders, home-based interviews, a mixed M aori-NZ European interview and research team, establishing rapport with participants before commencing, and a koha (gift) which recognised the time and effort given by participants as a sign of reciprocity. The research process was designed to be a positive, respectful experience for participants. The analysis was undertaken with M aori and non-M aori researchers who had agreed parameters for respectful contributions. 22 The research may not be generalisable to other groups in NZ and globally to other Indigenous populations; however, it does identify a number of issues that have commonalities in Indigenous peoples' experiences globally. A limitation is the ability of this study to capture the heterogeneity of M aori experience. Half of the interviews were with participants who had M aori children involved in Wh anau Pakari, and this also included interviews with NZ European parents and caregivers of M aori children. While this perhaps influenced how participants perceived their experiences, it also reflects the lived realities of M aori children growing up in contemporary NZ. Finally, the majority of participants were mothers or female caregivers, which may have affected the results. For the small number of male parents who participated, answers did not appear to differ from female parents' answers.

Conclusions
In conclusion, experiences of racism at institutionalised, interpersonal and internalised levels affects engagement for M aori families engaged in a healthy lifestyle programme. Past negative experiences elsewhere in the health care system affected engagement with the service. While participation in the programme was identified as a positive experience for most participants, this was insufficient to achieve sustained engagement for some families due to external competing priorities. Racism that occurs elsewhere in the health system or in wider society may have ongoing effects with regards to subsequent engagement with other health services, despite their acceptability, and until these issues are addressed, inequities in service engagement between Indigenous and non-Indigenous groups are likely to persist. While it may not be enough to address the effects of racial discrimination more generally, compassionate, respectful, culturally safe care focusing on positive relationship-building can help mitigate some of the impact of racism on engagement in prevailing health-care services.