Views of children with diabetes from underserved communities, and their families on diabetes, glycaemic control and healthcare provision: A qualitative evidence synthesis

Children and young people with diabetes (CYPD) from socio‐economically deprived and/or ethnic minority groups tend to have poorer glucose control and greater risk of diabetes‐related complications. In this systematic review of qualitative evidence (qualitative evidence synthesis, QES), we aimed to explore the experiences and views of clinical encounters in diabetes care from the perspectives of CYPD and their family/carers from underserved communities and healthcare professionals in diabetes care.


| INTRODUCTION
In the UK, 36,000 children and young people (CYPD) under the age of 19 are living with diabetes. 1The majority, 94.6%, have Type 1 diabetes while 3% have Type 2 diabetes (29,871 0-19 year olds). 2,3Good glycaemic control in CYPD is linked to optimal self-and family/carer management.Self-management is defined as "the ability of the individual in conjunction with family, community, and healthcare professionals to manage symptoms, treatments, lifestyle changes, and psychosocial, cultural, and spiritual consequences of health conditions". 4As "self and family/carer management" is not a defined concept familiar within the health service, the term is used in this context to refer to the "self-management" of children where the responsibilities fall to larger or lesser extents (depending on age) on parents/carers.
For both Type 1 and Type 2 diabetes, these selfmanagement behaviours include healthy eating, physical activity, adherence to treatment regimens, administering insulin in the majority, and self-monitoring of blood glucose. 5CYPD and their families can be supported to achieve good glycaemic control through self-and family/ carer management education programmes that include: dietetic support to help in carbohydrate counting (Type 1 diabetes) for those managed with mealtime insulin; the use of diabetes technologies such as continuous glucose monitoring and continuous insulin infusion pumps; lifestyle and exercise advice; and psychological interventions that support emotional health and well-being. 6YPD from socio-economically deprived and/or ethnic minority groups tend to have poorer outcomes in terms of glucose control, compared to CYPD from less socio-economically deprived and/or white UK populations, which puts them at greater risk of diabetes-related short-and long-term complications.[7][8][9][10] Evidence suggests that poorer outcomes are associated with lower levels of engagement with health services and reduced access to the technology that might assist with medication and glucose monitoring adherence.8,10,11 Indeed, there is considerable variation in the delivery and uptake of diabetes care processes, engagement with structured

What's new?
This is the first evidence synthesis of the views and experiences of children and young people with diabetes (and their families) who are from underserved communities.Descriptive themes led to the development of analytical themes that serve to better understand and strengthen the therapeutic alliance in the context of concordance with self-management of diabetes in CYPD.

What is already known?
• Ethnicity and low socio-economic status are associated with poor glycaemic control and longterm outcomes for children with diabetes • Little published data exist regarding the views and experiences of children and young people with diabetes from underserved communities (and their families and their healthcare providers) on their health care, their glycaemic control and managing their diabetes What this study has found?
• Themes identified included self-perception, identity, understanding of diabetes and its treatments, relevance of healthcare services, children taking on responsibility and support for families • An exploration of how the analytical themes interact reveals how concordance with treatment may be affected in underserved communities.
What are the implications of the study?
• The findings can be used to inform the development of diabetes services for children and young people from underserved communities.
education, diabetes technologies and achievement of diabetes-related health targets in underserved communities. 2Underserved communities, for the purpose of this evidence synthesis, are defined as young people living in the UK or Europe from non-white ethnicities and/ or people in the UK or Europe living with poverty or living in disadvantaged areas.Where an area is referred to as deprived, it can mean that people have a lower income but it can also mean they have fewer resources or opportunities.][14][15] Health inequalities are defined by the NHS as "unfair and avoidable differences in health across the population, and between different groups within society". 16To improve the outcomes of the most disadvantaged CYPD, it is important to address the challenges of health inequalities and the impact of a range of socio-cultural and economic factors.
Interventions to improve glycaemic control, and reduce anxiety for parents for both adults and children have been developed, they include Diabetes Self-Management Education advice and instruction, psychological counselling, self-monitoring, specialised (tailored) clinical care, occupational health one-to-ones with young adults with goals and aims, telecare (health-specific responsive feedback), parental mentoring, assistance with parenting and behaviour and interventions may use a combination of approaches. 17However, there are vanishingly few studies of these interventions aimed at children from underserved communities, and the majority have recruited and enrolled homogeneous adolescent samples from lower or the lowest risk groups.][20][21][22] In the past, approaches to improving diabetes selfmanagement have focused the attention and attributed blame on the behaviour of CYPD and their careers, rather than on the nature of the therapeutic relationship.This approach may be disempowering or hamper the integration of diabetes into daily life. 23,24Patient-centred approaches with a focus on the patient as a whole person, recognising the importance of individual preferences, social context and empowerment have been called for. 25nabling the sharing of power and decision-making about the interventions could enable CYPD and their parents to develop capacity for self-responsibility for their diabetes, tailored to fit the context of their lives.A review of knowledge and awareness of empowerment processes between patients and HCPs have shown equality, trust and respect are crucial in healthcare services to avoid judgemental attitudes and paternalism. 26oncordance is an evolving concept in health care in which the consultation is characterised by a partnership approach in which patients and HCPs work together to make decisions about treatment that align with patients' preferences, lifestyle and health beliefs. 23The aim of concordance in the context of diabetes management is to elicit a trusting relationship in the clinical encounter that supports self-and family/carers through facilitating a personalised treatment approach for the CYPD including discussion and mutual agreement between the HCP and the CYPD. 27These patient-centred factors should continue to be considered as the child develops and adopts responsibility for self-management of diabetes. 28he aim of this qualitative evidence synthesis (QES) is to identify and synthesise existing primary qualitative research exploring the experiences, beliefs, and views of CYPD from underserved communities, their family/carers and those who deliver diabetes healthcare services.We will look into CYPD and family/carers' experiences of the clinical encounter.This work will inform the development of the diabetes self-management interventions within the "Diversity in Diabetes" study (NIHR202358) which has an over-arching goal of codesigning a bespoke programme of support more sensitive to the needs and preferences of this patient group to better understand and strengthen the therapeutic alliance in the clinical encounter and to foster greater self-and family/carer management of diabetes in underserved communities.

| METHODS
We registered the protocol for this evidence synthesis on the PROSPERO database (CRD42022316028) 29 and followed ENTREQ reporting guidance. 30

| Research aim and inclusion criteria
We used the PERSPECTIF framework (Perspective, Setting, Phenomenon of Interest, Environment, Comparison, Timing) to frame the research question (see Table 1) 31 which was: To examine, experiences, views, beliefs of, and attitudes to the factors influencing selfmanagement of diabetes by CYPD from underserved communities, their family/carers and those who deliver diabetes healthcare services (doctors, specialist diabetes nurses, psychologists, nutritionists).Setting was paediatric diabetes healthcare services within "free at point of care" healthcare systems in high-income countries as defined by the World Bank. 32Factors of interest included, but were not limited to: engagement, acceptability of the healthcare system, aspects that help or hinder good self-management, aspects of specific interventions delivered to improve self-and family/carer management (e.g., introduction of additional support such as education programmes), acceptability of/or access to treatment regimens to improve glycaemic control, for example, insulin pump (continuous subcutaneous insulin infusion CSII), continuous glucose monitoring systems (CGMS).We included primary research using qualitative methods and systematic reviews for the purposes of citation searching.Determination of some inclusion criteria was an iterative process 33 and studies where the underserved communities were not representative of, and therefore likely to encounter different challenges to those to be found in the UK; for example, first nation populations, indigenous or aboriginal peoples in Australia, Canada, New Zealand or US, were excluded.Studies from countries that had recruited from an underserved population similar to the UK were included, for example, people migrating from Iran, Syria, Turkey and Sudan.Furthermore, we only included studies from countries where health care is free at the point of delivery through a government-funded system such as found in many countries in mainland Europe.

| Search methods
We ran Search strategies developed by an information specialist (SD) (February to March 2022) in MEDLINE, CINAHL, PsycINFO, Embase, Web of Science, International Bibliography of the Social Sciences (IBSS), Social Services Abstracts, Social Care Online and Sociological Abstracts.Terms for <children and young people>, <dia-betes> and <underserved communities> were used together with a filter for qualitative evidence. 34Full details are in Data S1.We limited the search to English language from the year 2000 onwards to ensure relevance to current health systems.Forwards and backwards citation searches were run on all included studies and relevant systematic reviews.

| Selection of studies
Titles and abstracts were screened to remove irrelevant records using Rayyan software. 35Relevant records were retrieved for full-text screening.Screening was done independently and in duplicate by two reviewers (TM, AS) using pre-stated inclusion criteria with disagreements solved by discussion.

| Data extraction and assessment of study quality
One author extracted study details using a standardised data extraction form and a second author (AS) checked them.The CASP (Critical Appraisal Skills Programme) tool was used to assess study rigour by one author (TM) and a second (AS) checking. 36

| Thematic analysis
Inductive thematic synthesis was undertaken following the three-stage process developed by Thomas and Harden. 33,37,38Stage 1: Data coding.Open, line-by-line coding of the first-order (verbatim quotes from respondents) and second-order (authors' views) constructs was done by two reviewers (AS, TM) in duplicate using NVivo.12. 39tage 2: Development of descriptive themes: TM, AS met to discuss and consolidate the meaning of the codes.Stage 3: Development of analytical themes: As the development of descriptive themes "remains" close to the primary studies and in accordance with Thomas and Harden, 38 the reviewers endeavoured to generate "new interpretive constructs explanations or hypotheses".This notion of "going beyond" the descriptive themes generated from the synthesis of primary study findings is what distinguishes it from other types of literature review and informs the development of subsequent research and intervention.

| Grade of evidence
The certainty of the descriptive themes was assessed with the application of GRADE CERQual (See Table 3 and Data S1). 40,41 3 |RESULTS

| Description of studies
We found 3782 records after removing duplicates (see Figure 1) and a further 67 from citation searching of included studies and systematic reviews.3][44][45][46][47][48] Reasons for exclusion are listed in Figure 1.All included studies were based in northern Europe, (Denmark, Sweden and the UK), see Table 2. Respondents were children and young people aged 4-19 years (n = 67) with Type 1 diabetes, 44,45,47,48 Type 2 diabetes 42 or a mixture of Type 1 and Type 2 diabetes. 43,46Two studies included parents (n = 16), 44,45 one study included healthcare professionals (specialist nurses and endocrinologists [n not reported]). 45The families and CYPD included in the studies were selected because they lived in areas of known deprivation 42,43,46 or because of their immigration status. 44,45,47,48thnicities included South Asian heritage 43 ; Black Caribbean heritage 42,43 ; Black African heritage 43 ; parents who had immigrated from Somalia, Morocco, Iraq, Lebanon, or were Turkish or Kurdish or had a "non-Nordic" background 44,45,47,48 or white British. 42,43Socio-economic status (SES) was specifically described as separate from ethnicity in three studies, all from the UK, which recruited in socio-economically deprived areas. 42,43,46The studies were set in paediatric diabetes outpatient clinics 43,45,47,48 or in the community 42,46 or special educational sessions organised by paediatric diabetes outpatient clinics. 44One study reported views of adults and children with diabetes and we extracted only constructs attributed specifically to CYPD. 46Some included measures to facilitate communication and relevance of the research such as interpreters, 47 and used CYPD co-investigators. 43

| Stages 1 and 2: Development of descriptive themes
The iterative coding and synthesis process led to the development of 11 descriptive themes that ranged from aspects of self-perception and identity, understanding of diabetes and its treatments, to how the families felt supported by each other, their caregivers and schools (Table 3).Most themes were graded as "moderate certainty" because of combinations of being drawn from relatively few studies or participants, availability of primary and secondary constructs and reasonable alignment of the aim of the papers to that of our research aim.The themes "Family finances" and "Diabetes self-care in a school setting" were graded as low certainty because there were few studies with thin data (fewer than 10 constructs) contributing to them.The theme "Perception of diabetes and glycaemic control" * Citation screening was done for relevant systematic reviews, and for included studies and primary research addressing the diabetes and underserved communities based in the UK was graded high certainty because there were a reasonable number of studies, and many constructs.(Table 3 and Data S1).

| Stage 3: Development of analytical themes
In the final stage, analytical themes were developed by AS, TM and JW who were involved in the process of identifying and interpreting concepts found within and across the dataset and going beyond the content of the original studies.Through further interpretation of the interrelationships between the descriptive themes, the analysts (AS TM JW) developed three analytical themes that are considered to be integral to engaging with the "diabetes discourse" and strengthening the therapeutic alliance.The "diabetes discourse" is considered the salient knowledge on what diabetes is and how it should be managed in the context of the biomedical model. 47The themes are 1.Alienation of CYPD; 2. Empowerment of CYPD and family; 3. Integration of diabetes (into self and daily life).
Theme 1: Alienation of CYPD (this theme was drawn from descriptive themes: self-perception and identity, HCP delivery, faith and culture, perception of the clinical consultation, social interaction and peer support, taking responsibility and family support).
Alienation is a recurrent theme for CYPD and their families.Alienation in the context of diabetes self-and family/carer management is affected by communication in the clinical encounter.Alienation of CYPD is also impacted by the extent of perceived family support and in relation to the extent of peer support, acceptance and identity validation for CYPD.During the clinical encounter, these factors contributing to alienation could be reinforced as a result of the CYPDs' / family carers' perception that HCPs followed a narrow diabetes discourse such that factors related to the family and socio-cultural and emotional context of the CYPD were not adequately addressed.
The responsibility for self-management behaviours was negotiated with family/carers who may either be reluctant to cede responsibility for self-management to the CYPD, or CYPD may perceive family/carers to be overinvolved in their care which makes it difficult to elicit an environment, supportive of self-management.There are also social pressures that may undermine the CYPD in taking responsibility for self-care.For example, the data from CYPD across the studies suggested that selfmanagement behaviours such as insulin injection, blood testing and having to be continually attached to an insulin pump can disrupt self-identity and social activities with peers and serve in alienating the CYPD.This alienation is also contingent on the extent of CYPDs' adaptation to social norms, and the influence of their families' cultural and religious background.For example, CYPD were reluctant to disclose their diabetes to others due to perceived shame and stigma that was culturally and socially based, which in turn made support seeking from peers and family difficult to negotiate.In addition, there are implications regarding the school environment such that a lack of awareness and prompting from staff, provision of suitable rooms offering privacy for insulin injection may also instil a sense of alienation in CYPD.
Theme 2: Empowerment of CYPD and family (this theme was drawn from descriptive themes: taking responsibility, family support, social interaction and peer support, perception of diabetes and treatments and faith and culture).
Patient empowerment is defined as helping patients discover and develop their inherent capacity to be responsible for their own health. 25Empowerment may be contingent on the extent to which the CYPD feel supported in taking responsibility for managing their diabetes and the extent to which their families' perception of diabetes and its treatment aligns with the "diabetes discourse".For example, there were data pertaining to the descriptive themes to suggest that CYPD from minoritised backgrounds were subject to culturally determined discourses where disease perceptions sometimes differ from the dominant discourse of the host country.Such data highlight the tension that some CYPD experienced between the dominant diabetes discourse of the clinical encounter and cultural perceptions of sickness and treatment relating to their own background.Furthermore, some families, despite having long-term experience of managing a child with diabetes and receiving educational support, still seemed to experience conflicts between their perception of "being a good parent" and diabetes management.These conflicts illustrate the emotional challenges families experienced with accepting the condition and its management.Review data also suggested that families were anxious about perceived social consequences and resulted in feelings of shame and stigma.For example, there were fears about how diabetes would affect CYPD's marriage or employment prospects which sometimes dominated over concerns about the potential for long-term health complications.Finally, empowerment of CYPD and family/carers was unlikely to develop if the responsibility for managing diabetes was seen to lie solely with the HCP and health service.This was affected by HCP's communication style which sometimes negatively affected motivation to learn about and take responsibility for self-management.Indeed, despite expressing satisfaction with the healthcare professionals, some parents described doubts about whether their children were given the best treatment, and some felt a need to consult social media and/or HCPs from

Measures in place to address cultural variation Aim
Approached n = 24 13 agreed Semi-structured interviews Phenomenology.Analysed using the meso, micro and macro model.
None reported.All adolescents spoke Swedish.
To identify factors important for adolescents to take responsibility for self-care.
Approached T A B L E 3 Descriptive themes and certainty GRADE CERQual.

Theme and description
Example (primary and secondary constructs)

Papers contributing to this Theme GRADE CERQual level of certainty 1 Diabetes self-care in school setting
Adolescents pointed out that too little time was scheduled for lunch to allow them to take the test, queue up, be served, and eat: "I do not have time to eat, and I also have to queue; terribly boring" (G, 17) 47 Boman 2015 47 Aspects of school organisation that affect the ability of CYPD to carry out self-management behaviours at school.
Hinder 2012 46 Sharpe 2021 43 Low 2 Faith and culture Despite their expressed satisfaction with the Danish healthcare professionals, immigrant parents described insecurity and worry as to whether their children were given the best treatment and a need to consult media, doctors and others from their own background. 45man 2015 47 How faith and culturally based factors affect self-and family/carer management of diabetes such as acceptance of diabetes and trust and dialogue with health services Povlsen 2008 45 Povlsen 2009 44 "I had a DKA [diabetic ketoacidosis) over Ramadan.My mother was away and I was at home with my older sister.I wanted to experience fasting and the feast at the end.However, I ended up in hospital" (P11, aged 18, T2). 43arpe 2021 43 Turner 2015 42

Moderate 3 Family finances
The impacts of poverty and family disruption were sometimes multiple and mutually reinforcing.Karl, for example, lived with his mother in a 2-bedroomed council house (IMD score 58); his brother lived with his father."I ask if Karl goes to any activities outside school.He says he used to go to boxing but his dad lost his driving licence and they had to stop for three months and did not start again".Field notes from visit to Karl, age 11, 46 Hinder 2012 46 Factors relating to the financial and social position of the family in relation to providing resources for self-and family/carer management of diabetes Povlsen 2009 44 Sharpe 2021 43 Turner 2015 42 Low 4 Family support "You're better off here with this disease, because of the chances of getting insulin and all that; the material things are available here.But down there (in Iraq) I would not be alone.I would have my mother, sister, aunts; family around me.This means that if I was tired, they could relieve me; take over". 44man 2015 47 Practical and emotional support from parents and other family members to the CYPD, and support for the parents of CYPD receive in relation to self-and family/carer management of diabetes.

Theme and description
Example (primary secondary constructs)

HCP and health service delivery
Information from medical records.Most families were described as having limited knowledge and/or compliance, but the records also demonstrated a lack of follow-up on identified problems.For example, one girl was tested and found to have limited knowledge on diabetes, but was not offered re-education; children with fear of injections or known failure to take the prescribed insulin were introduced to a new regimen requiring more daily injections.Some were educated at a level that seemed poorly adjusted to their background, such as being instructed and given a handout in Danish prescribing graduated doses of fast-acting insulin according to blood glucose levels and carbohydrate intake. 45man 2015 47 Factors impacting on HCP in information provision including challenges in language and communication.
Bowman 2017 48 The team expressed obvious frustration to the fact that many immigrant families have difficulties in accepting that diabetes is a 'true' chronic disease and that this influenced their motivation for education and learning: "What they want to know, is when the child will be cured.Some say it every time". 45nder 2012 46 Povlsen 2008 45 Povlsen 2009 44 Sharpe 2021 43

Moderate 6 Perceptions of the clinical consultation
Too often, participants' behaviour and attitudes have been wrongly characterised by clinicians as a marker of their disengagement from the service.The markers are hardly ever directly discussed by the healthcare team or voluntarily disclosed in clinical appointments by CYPD but loom large over the relationship.Participants said: "At first, I did not really take it [diabetes] that seriously, I suppose, but then when you realise it's going to affect you every day, then I started to take it more seriously".(P9, aged 15, T1).Some parents perceived that the diabetes team provided good help and support, while others felt that the professionals did not really care about them. 45man 2015 47 CYPD and families' views and experiences of clinical consultations and challenges self-and family/carer management to between appointments.Povlsen 2009 44 Hinder 2012 46 Interviewer You sound slightly hesitant to tell your friends.
Sharpe 2021 43 Participant I think it's just because the minute you say like, obviously they associate diabetes with like overweight, bad health … I did not want them sort of knowing all of that side of things.(Participant 11) 42 Turner 2015 42 Moderate 11 Taking responsibility for self-management (by CYPD) I want my mother to know that I can, that I can do it by myself; she does not need to nag all the time.I do understand that all she wants is to take care of me, but she does not need to nag.(G, 18G) 47 Boman 2017 48 Hinder 2012 46 CYPD negotiations with HCPs and parents in taking/ceding responsibility to self-manage diabetes for self-and family/carer management of diabetes.G5: "I like to do things my way, I want to have the control, and I want to know I'm in control".In one of these consultations, the adolescent was successful in her self-care and the doctor confirmed her autonomy: D4: "Yes, you are old enough now.You can do it by yourself, that's for sure". 48vlsen 2008 45 Povlsen 2009 44 Turner 2015 42
own background or country of Continuing struggles with accepting the condition, and stigma and a lack of trust can compromise attempts to support and empower CYPD and families and can make it more difficult for them to receive from their social network.
Theme 3: Integration of diabetes (into self and daily life) (this theme was drawn from descriptive themes: perception of diabetes, self-perception and identity, perceptions of diabetes treatments, social interaction and peer support and family support).
Integration of diabetes refers to the extent that having and managing diabetes is incorporated into the daily lives of CYPD and their sense of self.The integration of diabetes is contingent on ameliorating perceived alienation and increasing empowerment in CYPD to enhance their engagement with self-and family/carer management behaviours beyond the clinical consultation.The focus of integration of diabetes into CYPDs' identity is to elicit a sense of competence and volition such that CYPD are armed with skills to successfully negotiate their family and social environment and undertake self-management behaviours whilst embarking on their journey from adolescence to adulthood.
Integration requires family/carers ceding responsibility to enable CYPDs to self-manage their condition.Strategically, this might partly be achieved through parental style such that attitudes expressed within a "consultative" and supportive style may promote empowerment as opposed to a parental style perceived as "reminding", "nagging", and "controlling" which can cause frustration and discontent in CYPD.The synthesis also showed that CYPD who had integrated diabetes as part of their selfidentity expressed hope that they could further improve the quality of their self-management. 47This journey to integration could also be supported by HCPs through broadening the clinical encounter to explore the socio-cultural and emotional factors related to integration as the CYPD negotiates responsibility with both their family/carers and HCPs.Thus, determining the degree to which CYPD integrate diabetes into their self-identity is important to optimise psychological and emotional well-being and elicit the motivation to enact self-management behaviours.

| Concordance and self-management of diabetes
It is proposed that the three analytical themes are interdependent and provide a conceptual framework for understanding the socio-cultural and economic determinants of self-and family/carer management of diabetes.In addition, the themes may be used to explore engagement with the "diabetes discourse" in the clinical encounter to foster greater concordance with treatment.See Figure 2.

| DISCUSSION
The aim of this qualitative evidence synthesis was to identify existing primary qualitative research exploring the experiences, beliefs, and views of CYPD from underserved communities, their family/carers and those who deliver diabetes healthcare services.The findings have informed the development of three analytical themes.(1) Alienation of CYPD; (2) Empowerment of CYPD and family; (3) Integration of diabetes into daily life.These themes may F I G U R E 2 From alienation to concordance.How the analytic themes work independently and synergystically to decrease alienation, empower and increase integration to achieve greater concordance in self-family/carer management of diabetes.serve as a framework to inform and strengthen communication between CYPD, HCPs to foster greater concordance in diabetes The desired outcome of concordance is to elicit a trusting relationship in the clinical encounter that supports self-management outside of the clinic environment.

| Developing concordance with self-management of diabetes
A key finding is that the "diabetes discourse" may dominate in the clinical encounter and mostly exclude the CYPDs' subjective experience.A particular point of concern for CYPD during the clinical encounter, is the perceived narrow biomedical focus on glycaemic control without support to achieve it and to integrate the complex behaviours required into different lifestyles, for example seeing CYPD as having lives outside of the diabetes consultation.Indeed, Sharpe et al. 43 suggest that engendering trust in the clinical relationship ought to be prioritised to allow for "honest and frank discussions on lifestyles, behaviour and identity, which are not currently actively taking place in clinical appointments".However, sometimes CYPD may actively seek positive feedback and reassurance form HCPs. Therefore, such invitations provide an opportunity for the HCP to offer encouragement to sustain and intensify their striving for improvement. 48Finally, good communication can take place in the clinical encounter without compromising the biomedical aspects, if issues of concern to CYPD and family/carer are also addressed. 48Even when biomedical aspects are communicated well from the patient perspective, the socio-cultural factors may not be adequately acknowledged and addressed in the consultation and may reinforce feelings of alienation that are grounded in CYPDs' social and family lives.Some HCPs simply refrained from bringing up certain topics in diabetes education sessions on the assumption that they were not culturally acceptable, despite their relevance to diabetes care. 45In summary, communication of the biomedical aspects of managing diabetes in the clinical encounter is important but not to the exclusion of, addressing socio-cultural and emotional issues of concern to CYPD and family/carer rather than focusing solely on clinical concerns.
Person-centred care and shared decision-making approaches that promote collaboration between individuals with diabetes and HCPs have the potential to mitigate health inequalities, including the lower uptake of diabetes technology, fewer prescriptions of intensive therapy plans, and worse psychosocial outcomes. 49,50Furthermore, making efforts to know the CYPD, to provide opportunities to engage CYPD as active partners in a person-centred approach, have been shown to improve concordance between care provider and patient regarding treatment plans in other illness contexts. 51It is also suggested that HCPs should address CYPD family/carers' practical and emotional problems, to minimise conflicts related to diabetes management in order to elicit motivation for self-care as well as promoting clinic attendance. 46here were data in the synthesis to suggest that HCPs expressed interest in the families by being "open, curious and not prejudiced" and/or by trying not to go beyond their limits with regard to perceived privacy and/or cultural norms.HCPs also endeavoured to become better acquainted with CYPD in an effort to empower them through empathising with their lives and perspectives and by asking them to describe themselves and link that to their motivation for self-management. 48Furthermore, other HCP strategies such as normalising feelings of frustration and acknowledging negative emotional responses to having a chronic disease by offering emotional support can help manage crises and enable CYPD come to accept their condition and ongoing management. 46ocio-economic factors play a pervasive and persistent role in child health and development. 52However, it is not clear how ethnicity impacts on childhood diabetes independently of socio-economic status.Such complexity could explain why not all CYPD and family/ carers attribute non-adherence with self-and family/carer management to their migrant and ethnic background.For example, Boman et al. 47 suggest that belonging to an underserved community does not necessarily mean that it is the membership in itself that is associated with vulnerability.Language may also be an additional or a standalone challenge to communication other than the ethnic/ religious background of the CYPDs' family.Furthermore, it can also be difficult for some CYPD to acknowledge and articulate relevant aspects of their own culture, because they are inherent and/or assimilated since birth or may not be different, with regard to identity and social interaction, to that of peers from non-ethnic backgrounds.It is also known that health inequalities can compromise support seeking from and empowerment of patients from minoritised communities. 26We were unable to draw directly from the included studies any substantial themes about what CYPD, their carers or the HCP caring for them would like to see in terms of support for self-management as there were no direct first-order constructs on this.In some cases we could infer the type of support that was lacking, but views were not all aligned.For example some parents expressed a lack of trust in the healthcare services while others in the same study found the support to be acceptable, and stated what they would see. 44 recent overview of on supporting selfmanagement for CYPD identified four broad strategies psychoeducation, including many delivered as digital games; psychological counselling to families and children or to adolescents as individuals; self-monitoring using mobile phones and apps; and telehealth, which incorporated self-monitoring with feedback and contact with services, again using apps and websites.The effects of these interventions on healthcare outcomes were mixed, with some studies reporting improvements and others finding the interventions had no effect 17 however none of the reviews included had a focus on children from underserved communities.Evidence from reviews of interventions for adults with diabetes from underserved communities show a positive effect of developing or adapting interventions on healthcare outcomes.4][55][56] In 2010, a review of how culture was addressed for managing diabetes or asthma in children from ethnic minority groups reported the following adaptations: translation of materials into native language of ethnic groups; accommodation of beliefs, myths and practices of ethnic groups; acculturation of interventionists; and adopting a collaborative approach to intervention development with ethnic groups. 19A review of strategies to deal with food insecurity in families with diabetes, listed new practices to support populations with diabetes management including screening for food insecurity as part of routine care, nutrition counselling and budget management, tailoring of medical management to reduce incidence of hypoglycaemia, involving patients in decision making process, assess coping strategies of patients and referral to community resources related to food and housing, such as affordable grocery stores. 57ssues encountered by adults from minoritised populations with diabetes and their medical care included the participants' strong adherence to cultural norms, religious beliefs, linguistic diversity, low health literacy levels, different beliefs about health and illness, belief in expert and professional support, low accessibility of culturally appropriate services/information, and low concordance with Western professional advice. 58Taken together this evidence suggests that for underserved communities diabetes health care could be enhanced with measures to improve communication in clinical encounters, and a power sharing approach could help to achieve concordance.This qualitative evidence synthesis suggests that underserved communities are likely to face challenges in the adaptation of diabetes education to their own context and require culturally competent consultations and informed education and support in and beyond the clinical encounter.It is also recognised that CYPD have a greater risk than their peers, living without diabetes, of psychological conditions such as anxiety and depression, and behavioural conditions such as conduct disorders and attention deficit hyperactivity disorder. 59Therefore, these factors should be given due consideration especially for CYPD from underserved communities. 43,46This point is emphasised by the UK National Institute for Health and Care Excellence (NICE) guidelines for CYPD which recommend that "children and young people with Type 1 diabetes and their families or carers should be offered emotional support after diagnosis tailored to their emotional, social, cultural and age-dependent needs". 59It has been suggested that incorporating an explicit focus on ethnicity is important because ethnic identities have implications for health, independent of other socio-economic factors. 60,61owever, addressing these diverse, socio-economic factors within four 30 to 45 minute-, clinic consultations per year (based on typical UK consultation times) is challenging.There is therefore, a need for more support for HCPs to address socio-cultural factors in clinical consultations.Some HCPs may be reluctant to bring up topics related to ethnicity, culture and socio-economic disadvantage as they may be sensitive and evoke strong emotions.However, training in communication skills, active listening and cultural competence is available.By gaining insight and confidence in navigating and responding to the varied needs of patients, HCPs will influence the way care is given and could have an impact on reducing disparities in healthcare outcomes.

| Limitations of findings
The studies included in this synthesis comprise a range of data collection approaches such as direct observation of consultations and analysis of medical records, which support the adequacy of the findings and are not solely reliant on the quality of a semi-structured interviews.The certainty of the descriptive themes is somewhat supported by the application of GRADE CERQual which demonstrated moderate confidence across most themes.Issues with regard to confidence were mostly around adequacy of data with some themes arising from four or fewer studies.a comprehensive search studies met the inclusion criteria and although two contributed to four of these papers each from those groups was a distinct piece of research with unique participant samples.Furthermore, some descriptive themes were developed from papers that had limited relevance to our research question, for example the sample in Hinder et al. 46 were aged 5 to 88 and included many adults although data was extracted pertaining only to quotes attributable to CYPD.Turner et al. drew from a broad population with only 4 participants identified from underserved communities. 42,46This particularly affected the theme of finances where insufficient evidence was found to include it in the analytical themes, the importance to families of adequate financial resources to provide the means to support healthy lifestyles and stable family structures is acknowledged.Therefore, future primary research should include questions relating to financial security and the family budget and such data may need to be included in the future development of the themes as more evidence becomes available.
The majority of the observations in this synthesis are from CYPD and families with Type 1 Diabetes.Two the seven studies included CYPD with Type 2 diabetes, Sharpe et al. 2021 included 2/20 children and Turner recruited all 12 respondents with Type 2 diabetes. 42,43Hinder et al. included 5 children but we do not know which type diabetes. 46The remaining 4 studies had 49 respondents and all children had Type 1 diabetes. 44,45,47,48We saw no themes that related specifically to the type of diabetes.It is reasonable to assume that much of the work to engage families and CYPD from underserved communities will be similar irrespective of diabetes type, but there will be differences specific to the management of these different diseases.Further studies could show how clinical services or interventions could incorporate such differences.
Data were also limited with regard to the perspective of the HCPs by including only 4 HCPs and 1 diabetes team, which limit making generalisations about HCP's perceptions.However, one study was based on a phenomenological analysis of video-recordings of 12 paediatrician-CYPD consultations which served to complement CYPD perspectives of the clinical encounter derived from interview data. 48The majority of the respondents in the synthesis had a diagnosis of Type 1 diabetes and there were a wide range of ethnic minorities represented, but the extent of socio-economic deprivation is less certain.There are also limitations on the extent that underserved communities are affected by access to and effects of the newest technology, such as hybrid closed-loop pumps, which have come out since the papers included in this reviews were published.

| Future work and management of diabetes in CYPD from underserved communities
The present QES has highlighted that the relationship between health services provided for CYPD from underserved communities in the UK and Europe is an under-researched area.The findings from this QES have informed the design and conduct of ongoing primary qualitative work that aims to explore the management of diabetes in CYPD and family/caregivers from underserved communities and the healthcare professionals who care for them.This primary qualitative research, together with this QES and two systematic reviews will inform the development, in partnership with CYP and their families, of culturally acceptable intervention packages. 17,62,63These will be based on models of community liaison keyworkers, targeting education and behaviour change, addressing the specific personal, family, cultural and community barriers to optimal glucose control faced by these groups with Type 1 or Type 2 diabetes.

| Analyst reflexivity statement
AS is a qualitative methodologist with a background in health psychology including self-management of diabetes and practitioner-patient communication.TM has a background in evidence synthesis of quantitative and qualitative data.In acknowledging their respective areas of expertise, the primary analysts have made complementary contributions to data analysis and interpretation of data such that any inherent bias in findings would be balanced by their different theoretical and methodological perspectives.

AUTHOR CONTRIBUTIONS
Records excludedTotal n = 3682 From databases (n =3636 ) From citation screening (n = 46) Full-text reports assessed for eligibility From databases (n =146) From citation screening (n = 21) Total n = 167 Reports excluded Total n = 160 n = 54 Focus is not underserved n = 41 Focus is not children or young people n = 22 Systematic review n = 18 Qualitative methods not used n = 14 Population not relevant to UK setting n = 4 Focus of paper is not relevant n = 4 Setting low or middle income country n = 2 Data analysed as quantitative n = 1 No 1st or 2nd order constructs Reports meeting eligibility criteria and based in UK or Europe From citation screening (n = 1) From databases (n = 6) databases (n =3782 ) From citation screening (n = 67) Total n = 3849

T A B L E 2 9 Black African n = 2 4 White British n = 8
Characteristics of included studies.Setting: Paediatric diabetes outpatient clinic CASP = 10/10 Adolescents: n = 13 age: 13-18 years.Female n = not stated.Male n = not stated First or second-generation immigrants to Sweden Born in Sweden n = 9, born in Iraq n = 3, born in Somalia n = 1 Diabetes: Type 1 Setting: Paediatric diabetes outpatient clinic Boman 2017 48 Sweden Setting: Paediatric diabetes outpatient clinic CASP = 9/10 Adolescents: n = 9 age: 13-18 years.Mean age 15 years.Female n = 6.Male n = 35 Participants used continuous insulin infusion and 4 used multiple daily injections Healthcare professionals n = 4 Non-Nordic background.All could speak Swedish Diabetes: Type 1 Setting: Paediatric diabetes outpatient clinic Povlsen 2008 45 Denmark Setting: Additional education sessions from the diabetes healthcare teams directed at immigrant families with Arabic speaking parents CASP = 5/10 Children: n = 11 age: 4-17 years.Female n = 6 Male n = 5 Parents: n = 8 parents of the children.Healthcare professionals (specialist nurses and endocrinologist) Interpreter Turkish or Kurdish immigrants to Denmark 1 of 11 of the mothers was 2nd generation.Children were therefore mostly second-generation immigrants with 1 third generation Diabetes: Type 1 Diagnosed for 1-9 years Paediatric diabetes outpatient clinic Povlsen 2009 44 Denmark Setting: Additional education sessions from the diabetes healthcare teams directed at immigrant families with Arabic speaking parents.CASP = 9/10 Arabic speaking parents (n = 4 mothers and n = 4 fathers) of children with Type 1 diabetes Children were not interviewed but were: aged 7 to 16 n = 4 boys and n = 3 girls.Children aged 7-16 years Immigrant population: Parents of children with Type 1 diabetes who immigrated to Denmark from Iraq (n = 4, Morocco n = 1, Lebanon n = 1).At least one parent from each family needed an interpreter Diabetes: Type 1 The children were diagnosed after they immigrated to Denmark.Setting: Additional education sessions from the diabetes healthcare teams directed at immigrant families with Arabic speaking parents.Glostrup, Denmark Hinder 2012 46 Setting: Home, school or work UK Four counties in the North of England CASP 10/10 Children and Adults N = 30) (Aged 5 to 88 years).Children: n = 5 aged 5 to 16 years Specifically regions of high social deprivation in the UK.Ethnicity: White: n = 22, South Asian n = 7, African n = 1 Ethnicities not specified for children Diabetes: Type 1 n = 15 Type 2 n = 15 The type of diabetes was not specified for the children.Setting: Home, school or work Sharpe 2021 43 UK North London and East London Setting: Paediatric outpatient clinic CASP = 9/10 Children and adolescents n = 22 age 10-19 years.10-13 years n = 9 14-17 years n = 4 18 and over n = Black Caribbean n = 2 Dual heritage n = 4 South Asian n = 7 White British n = 6 White other =1 Diabetes: Type 1 n = 20 Type 2 n = 2 Setting: Paediatric outpatient clinic Turner 2015 42 UK Setting: Healthcare, home, and school.CASP = 9/10 Adolescents: n = 12 Age 14-17 years n = 4 Age 16-17 years n = 4 Age 18-19 years n = Asian n = 3 Black Caribbean n = 1 Parental unemployment: Living with one parent unemployed n = 4 Living with two parents unemployed n = 4 Living with two employed parents n = 3 Diabetes: Type 2 diabetes diagnosed for 2 years.Setting: Healthcare, home, and school and peer supportInterviewer And how was it at school with friends and that?Boman 2015 47 CYPD navigating social life and fostering support for selfmanagement of diabetes from peers in social context.Participant I did not really tell them.Unless they saw me doing an injection, I never really let them know … it was more to keep it quiet so … they thought I was a bit weird, a bit off.(Participant 9)

3
IntegraƟon into CYPD and family/carer Concordance with treatment in CYPD and family/carer I IN NC CR RE EA AS SE E I IN NC CR RE EA AS SE E D DE EC CR RE EA AS SE E

T A B L E 1
Inclusion criteria based on the PerSPEcTiF framework. 19,57

4.2 | Culturally informed education and support in and beyond the clinical encounter
Conception of the review; Sabi Redwood, Aidan Searle, Julian Hamilton-Shield.Design of the review; Theresa H. Moore, Aidan Searle, Sabi Redwood, Julian Hamilton-Shield, Jessica Wheeler.Coordination of the review; Theresa H. Moore, Aidan Searle.Search design and organisation of search results; Sarah Dawson.Selection of studies for inclusion in the review: Theresa H. Moore, Aidan Searle.Collection of data for the review; Theresa H. Moore, Aidan Searle.Assessment of the methods; Theresa H. Moore, Aidan Searle.Development of descriptive themes; Theresa H. Moore, Aidan Searle.Assessment of the certainty in the body of evidence; Theresa H. Moore, Aidan Searle.Development of analytical themes; Aidan Searle, Jessica Wheeler, Theresa H. Moore.Comment on the manuscript; Theresa H. Moore Sabi Redwood IL SG