Falling into a deep dark hole: Tongan people’s perceptions of being at risk of developing type 2 diabetes

Abstract Background Prediabetes is a precursor for type 2 diabetes. Compared to the New Zealand/European and other population groups (24.6%), the prevalence of prediabetes is higher within Pacific groups (29.8%). The diagnosis of prediabetes presents a potential opportunity to intervene to prevent progression to type 2 diabetes. Objective To develop an understanding of how being ‘at risk’ of developing type 2 diabetes is perceived by Tongan people with prediabetes living in Auckland, New Zealand. Methods The Kakala and Talanga Tongan methodologies underpinned this study. Twelve one‐on‐one, semi‐structured interviews with Tongan patients who had prediabetes from a primary health‐care clinic in Auckland, New Zealand, were conducted. Thematic analysis was used to identify recurrent themes from the data. Results Participants were not aware of their prediabetes diagnosis, emotions associated with the diagnosis reflected fear and disbelief and a perception of imminent danger. Family history informed perceptions of the risk of developing type 2 diabetes. Participants could not differentiate prediabetes from type 2 diabetes, and recollections of being ‘back in the Islands’ of Tonga were consistent with healthy lifestyles. Conclusions Prediabetes appeared to be poorly understood and was believed to be irreversible, which could discourage behaviour change, social and physical improvements in health. Appropriate culturally tailored messages to accompany a prediabetes diagnosis, including cause and management, would be beneficial for Pacific peoples.

Tongan people comprising 20.4% of Pacific peoples in NZ. 4 Pacific peoples are at high risk of T2D both within their homelands and as migrants to NZ. 5 Overall, Pacific populations in NZ are more likely to have a non-communicable disease, compared to NZ/European people. [5][6][7] Prediabetes described by the World Health Organization is a state of higher than normal high blood sugars (41-49 mmol/mol) but not high enough to meet the diagnostic criteria for T2D (>50 mmol/ mol). 8 The chronic and non-acute nature of prediabetes can lead to late diagnosis and delayed treatment. Established risk factors associated with prediabetes include ethnicity (particularly for ethnic groups in NZ, being of Māori, Pacific Islander or South Asian descent) 5 and increasing age. Other risk factors include a sedentary lifestyle, poor nutrition (and overweight), high blood pressure and family history of T2D. 9 Although prediabetes can lead to T2D, it can be managed through lifestyle changes such as diet and physical activity. 9 The diagnosis of prediabetes presents a potential opportunity to intervene to prevent progression to T2D. The challenge of engaging with people who have been diagnosed with prediabetes is that they are likely to feel well, therefore may see little need to change their behaviours. 10 This ambivalence may reflect a pervasive belief that T2D is largely pre-determined, an inevitable consequence of ageing or genetics. 11 People who are aware that they are at risk for developing T2D have the best chance of preventing the onset of T2D by changing their diet and increasing their physical activity. 12 Maintaining good health while being at risk for a chronic condition is argued to be contingent on access to credible information on practical options to manage risk. 11 Yet, people who live with an elevated risk profile are often least likely to have access to and/or engage with health information and system support. The NZ health system is designed to deliver quality, accessible health care to a diverse population. 13 Despite an intention to be accessible, the majority of information for supporting people with long-term conditions, such as diabetes, is via didactic and Western biomedical health information approaches, for example pamphlets or in-person consultations explaining the importance of dietary changes. Research suggests that access to credible, culturally relevant health information is essential, but not the only element required to support behaviour change. 14 Misinterpretation of prediabetes information between patient and clinician during a consultation could lead to inaccurate knowledge about the definition, diagnosis and treatment. [15][16][17][18][19] Any misunderstanding of what prediabetes is and being at risk of developing T2D may affect the likelihood of initiating any necessary behaviour changes such as changing diet or increasing physical activity. 18 Some argue that the risk needs to be communicated effectively to mobilize behaviour change. [19][20][21] An individual's perception of being 'at risk' of developing a life-threatening disease inevitably shapes how they think and feel and what they do. 21 The literature pertaining to risk perception of individuals at high risk of developing T2D and those diagnosed with prediabetes highlights several key points. Firstly, there are inherent differences in knowledge and understanding of risk and prediabetes between clinician and patient. 15,21,22 Often, health-care providers base diagnoses on physiological markers and use medical terminology that patients do not necessarily understand, whereas a patient's perspective of being at risk is based on illness perception, 23 social factors such as culture, family history and present lifestyle, their emotional state and ability to understand medical information at that point in time. 16,17 Studies find that promoting patient-centred dialogue about prediabetes is important because it ensures that patients can comprehend risk on a good level of understanding to make decisions for their health. 19 What constitutes patient-centred information, of course, is variable. Secondly, other studies found that patients who were at risk of developing T2D were unaware of their prediabetes diagnosis 24 leading to guilt, distress and fear. 15,18,21 However, when the diagnosis was made known, patients felt that this was not serious because prediabetes was asymptomatic. 18 This led to ambiguity about the meaning and seriousness of the disease. Thirdly, other studies reported that patients with prediabetes acknowledged that family history of T2D directly impacted their understanding of being at risk of developing T2D. 18,25 One study revealed that having a family history of T2D prevented positive behaviour change in patients due to belief in a pre-determined outcome of T2D. 25 On the other hand, a diagnosis of prediabetes motivated and encouraged other patients to change their eating habits and increase physical activity. 25 Those who were at high risk of developing T2D were more likely to draw upon the negative experiences of family members who lived with T2D to initiate positive lifestyle changes for themselves. The adverse experiences of seeing family members take diabetes medication to manage their condition and avoid limb amputations and subsequently death were actually detrimental actions to prevent T2D. It is evident that patients at high risk of developing T2D comprehend prediabetes in multiple ways.
Though it is almost always based on family history of T2D, this is an interesting proposition that despite strong familial connections, personal risk of developing T2D might be an opportunity to mobilize collective or personal action to change.
This study was designed to examine how Pacific people, specifically Tongan people with prediabetes, understand and conceptualize the concept of risk as it pertains to T2D. Pacific people experience a higher risk of developing T2D; understanding perceptions of risk may provide clues for future behaviour and social group-level change interventions.

| Pacific qualitative methodologies
This study used the Kakala 26 and Talanga 27 Tongan methodologies. The Kakala methodology is the process of garland making in the Tongan context. There are three stages in this process: 1. Toli: to pick flowers needed to make the garland, 2. Tui: to weave the flowers in a respectable fashion, and 3. Luva: to present the garland to the intended wearer. Each stage of the Kakala methodology is equivalent to the research process of recruiting participants, data analysis and dissemination of findings. 28 The Talanga is the manner of interactive talking with a purpose between people. 27 These methodologies were essential to adhere to during the research process for cultural consistency, respectful sharing of experiences and authentic dialogue.

| Recruitment
Eligibility criteria required participants to be of Tongan descent and with a clinical record of being diagnosed with prediabetes (HbA1c between 41 and 49 mmol/mol) with no subsequent T2D diagnosis. Participants were purposively sampled from a Pacific primary health-care clinic located in South Auckland, NZ. The health coach who is responsible for managing recall for people with diabetes is the prediabetes educator in the clinic assisted with recruitment. The health coach provided information about the study to all eligible patients about the study. Those who were interested in participating in an interview contacted the clinic to arrange a convenient day, and time for the interview was scheduled.

| Transcribing and Translating
Recorded interviews were transcribed verbatim in the language it was conducted (eg Tongan or English). All transcripts in Tongan were translated to English for analysis. The first author completed the translation process. The transcripts were then reviewed by an independent Tongan scholar to ensure that the English translations aligned with the Tongan transcript version.

| Data analysis
Transcripts were de-identified, imported into and managed using NVivo version 12. Braun and Clarke's six-stage thematic analysis method was used to elicit critical themes. 29 The first author became familiar with the data by reading and re-reading the printed hard-copy transcripts. The transcripts were imported to NVivo for further analysis. Using NVivo, the textual data were coded inductively, line-by-line generating several codes. A codebook was generated to detail key concepts. The codebook was reviewed by the research team to refine specific themes and ensure consistency of understanding. Codes that were similar in meaning were grouped into a theme. Key themes were identified by their significance, based on the codes. A thematic map was then created to illustrate connections and defined the key themes.

| Participant characteristics
A total of twelve patients who have recently been diagnosed with prediabetes were interviewed. Table 1 presents the demographic characteristic of the sample. Most participants were female (nine), and three participants had other comorbidities, including stroke, hepatitis B and muscular dystrophy. Eleven participants were born in Tonga, and all eleven had a family history of T2D. Tertiary (University level) 5

| Themes
There were four key themes from the interviews: (a) no awareness of prediabetes diagnosis, (b) emotions associated with prediabetes diagnosis, (c) lack of differentiation between prediabetes, being at risk and T2D, and (d) 'back in the Islands' of Tonga was consistent with a healthy lifestyle. Each theme is described below with supporting quotes.

| Emotions associated with a prediabetes diagnosis
The common emotions associated with views of being at risk and having a prediabetes diagnosis were disbelief and perceptions of being in danger and fear.

| Disbelief
With minimal awareness of their prediabetes diagnosis prior to the interview, participants expressed disbelief when told of their

| Perception of being in danger
All participants expressed that being given a diagnosis of prediabetes (and therefore at an elevated risk of developing T2D) was feeling like they were in a state of danger, reinforced with a warning. Most participants described prediabetes as a sign of being on the verge of hitting

| Fear
Participants experienced the effects of T2D vicariously through seeing family members live with and eventually die from the disease. This familial experience played an important role in both shaping and exacerbating the fear associated with a prediabetes diagnosis.

It's like a feeling of fear because as time goes on it's like I will get it. (Female, 52 years)
To be honest, fear. I was scared to, at the time I got the result when it said that I've got diabetes.

(Female, 35 years)
When discussing the implications of prediabetes developing into T2D, more than half of the participants expressed that they feared T2D because they did not want to have insulin injections and commented that they knew of family members whose limbs were amputated as a result.

| Understanding prediabetes
Participants understanding of prediabetes was constructed upon experiences of family history of T2D. As a result, a few participants could not distinguish a diagnosis of prediabetes from T2D.
One male participant expressed that being at risk was clearly having T2D and not being able to return to good health from the diagnosis.  Treats such as fizzy (soda) drinks, chocolate biscuits and potato chips were rare luxuries for the participants who grew up in Tonga.

| 'Back in the Islands' of Tonga
Participants described drinking coconuts and water rather than fizzy.
Their mode of transportation and range included walking to and from school, church and the city town centre. Some participants recalled working as farmers on their land growing crops to feed their families.
These daily activities constituted their exercise routine.   37 Strachan and colleagues found that for many, it was inevitable that the progression of T2D will occur, regardless of health support since the risk factors associated with prediabetes were primarily embedded in the participant's everyday lives. However, studies report that prediabetes is reversible with lifestyle modifications, 18,38 including evidence that lifestyle interventions could lead to a 27% reduction of diabetes incidence over 15 years. 12 Lifestyle modifications have been shown to delay the development of T2D by three to four years, 39 and in patients who exercised, and with every kilogram in weight loss, the risk of T2D has been found to reduce by 16%. 40 In addition to structural and environmental factors, health literacy and tailored health promotion initiatives can be instrumental in reducing risks associated with T2D. 41 The prediabetes screening appointment presents an opportunity to actively engage patients with practical, realistic and culturally appropriate lifestyle adaptions to stave off T2D. 42 The importance of culturally relevant health literacy and education tools should not be underestimated in communicating important information.

| D ISCUSS I ON
Understandably, many participants held a poor understanding of the difference between prediabetes and T2D. This may be due to several factors, including the relatively new and possibly ambiguous term 'prediabetes' as well as asymptomatic nature of both prediabetes and T2D. 43 Studies found that patients who were diagnosed with prediabetes did not see the seriousness of the prediabetes to make immediate changes. 15 It is essential, therefore, that patients with prediabetes know that the condition can lead to T2D and that with changes to diet and physical activity, the risks can be reduced. 43 The findings of our study reflect concern that there is no simple Tongan word for prediabetes or term to describe being at risk of developing T2D. In this instance, risk and prediabetes were described as a concept accompanied with a description that was relatable to Tongan Participants went on the traditional Aboriginal diet for seven weeks. As a result, participants lost an average of eight kilograms in weight. 47 There is therefore potential that reverting to or increasing traditional foods in everyday diets and lifestyles incorporating greater physical activity could potentially reduce the risk of developing T2D. However, the challenge associated with the rise of globalization worldwide creating obesogenic environments still poses difficulties and challenges for Pacific peoples in NZ and the Pacific region.y 48

A strength of this study includes conducting interviews in both
Tongan and English languages. The use of Pacific methodologies to underpin the research design ensured that Pacific worldviews were reflective of the participants. The generalizability of these findings, however, cannot be applied to all Pacific peoples in NZ or the Tongan people living in the Kingdom of Tonga. This study was conducted in a Pacific primary health-care provider located in an area with a large enrolled Pacific population which allowed for purposive recruitment of participants. Due to the small number of participants, and the sampling method, the perspectives are not generalizable. It is a limitation that recruitment was from one clinic and there is potential that the way prediabetes is communicated in this clinic is not reflective of other clinics. The sample was limited to people who were engaged with the health system and had a diagnosis, and therefore, the findings are not a reflection of the perceptions of those who are managing their prediabetes, and those not engaged with the health system.

| CON CLUS IONS
To improve awareness of a prediabetes diagnosis within Tongan populations, primary health-care screening and appropriate communication of being at risk of developing T2D are vital. Developing feasible, culturally relevant ways of communicating about prevention and management of prediabetes and T2D within the Tongan community is essential to support social and behaviour change.

E THI C S
This study received approval from The University of Auckland Human Participants Ethics Committee on the 2nd of June 2017 for three years (reference 019242). All participants provided informed consent before taking part.

ACK N OWLED G EM ENTS
The authors would like to thank the primary health-care clinic, health coach and study participants.

CO N FLI C T O F I NTE R E S T
All authors have no conflict of interest to declare.