Uncertainty and certainty: perceptions and experiences of prediabetes in New Zealand primary care – a qualitative study

Keywords: exercise, lifestyle factors, nutrition, prediabetes, prevention, primary care, public health, type 2 diabetes mellitus.

Prediabetes can precede type two diabetes mellitus (T2DM) and is a condition where blood glucose levels are above normal, but not high enough to be classified as T2DM.¹ Prediabetes is associated with a higher risk for heart disease, stroke, neuropathy, and microvascular complications.² There is international variation in the definition of and the tests used to diagnose prediabetes and T2DM. In New Zealand (NZ), the diagnosis of prediabetes is made when a glycated haemoglobin (HbA1c) level is between 41 and 49 mmol/mol; and T2DM when a HbA1c level is ≥50 mmol/mol.³

Prediabetes is associated with between a 5 and 10% annual conversation rate to T2DM⁴ and eventually a 70% conversion.⁵ Some people spontaneously revert to normoglycaemia.⁶ Prediabetes may also be reversed or stabilised by lifestyle interventions.⁷ Interventions that reduce the risk of conversion to T2DM or increase the rate of reversion to normoglycaemia are important, not only for individual wellbeing, but also to reduce societal disease and economic costs associated with T2DM.⁸

In NZ, there is a high and increasing prevalence of both prediabetes and T2DM.⁹ The prevalence for prediabetes is 30.4% for Māori, and 29.8% for Pacific peoples, with an overall prevalence of 26%.¹⁰ This is in parallel with the high prevalence of obesity in NZ, where about one-third of adults are obese, with obesity more prevalent in Māori and Pacific peoples.⁹ These ethnic differences are, in part, associated with inequitable care, including types of interventions offered and the degree to which culturally safe care is given.¹¹

People with prediabetes in NZ are often identified during the population-wide primary care cardiovascular risk assessment programme and are then managed in primary care. Management is based on NZ Ministry of Health guidelines^12,13 and regional clinical pathways,¹⁴ which advise: (1) Healthy eating; (2) Increased physical activity; (3) Weight reduction; and (4) Medication. There are no specific national NZ intensive prediabetes interventions as in other countries,^7,15,16 although time-limited initiatives have been trialled.^17–20

This study reports the qualitative component of a mixed methods study, which also included a randomised controlled trial (RCT) evaluating the effect of probiotic and breakfast cereal-based prebiotic interventions in those with prediabetes.²¹ Ethical approval was granted by the Central Health and Disability Ethics Committee (17/CEN/88). Participants were given a prediabetes information pamphlet at enrolment. Researchers did not give health advice, nor instigate discussions around prediabetes, but clarified the understanding of prediabetes if asked. In this qualitative component, we sought to investigate views about prediabetes and its management from general practitioners (GPs) and nurses, and people with prediabetes.

A sequential mixed qualitative methods approach was used.²² Mixed methods intentionally collect multiple data sets with the aim of triangulating the data. Similarities and differences across the datasets can expand or elucidate understanding. Four data sets were collected: open-ended free-text survey responses about individuals’ understanding of pre-diabetes and the impact of diagnosis; individual interviews; focus group data from people with prediabetes; and finally focus group data from health professionals. These datasets were gathered at different times during the overall study (see Fig. 1).

Fig. 1.  Methods and participants.

Both GPs and nurses were purposively recruited from three general practices in Wellington, New Zealand. They participated in one of three face-to-face focus groups (one per practice). Semi-structured questioning explored their views about working with people with prediabetes.

The people with prediabetes were recruited for interview/focus group participation from the RCT intervention arm. We purposively recruited equal numbers of participants from each ethnic group (Māori, Pacific and South Asian), as well as a European/other category. We sought participants with Māori, Pacific and South Asian ethnicity as these groups are disproportionally affected by T2DM and prediabetes. Learning about the experiences of groups with the highest rates of these conditions is important to understand the drivers of inequities.

The last 85 of the 153 people recruited to the RCT population were asked to complete an online survey.

Survey, individual and focus group questions were informed by previous work on lifestyle behaviour change and health service delivery,²³ and included participants' views of their health and the study intervention.

Interviews and focus group discussions were transcribed. Each dataset was analysed independent to the others. The analysis used an inductive approach to establish meaning and themes,²⁴ with NVivo 12 (QSR International) used to categorise and code (JH, SM, FH). A consensus of themes and analysis was obtained for each dataset through discussion and testing of examples (EM, SM, FH, CB).

Table 1 shows a summary of participants and Table 2 a summary of identified themes.

Table 1.  Summary of participants.

Table 2.  Summary of themes.

GPs believed progression to T2DM was unpredictable. This created uncertainty about when and how to intervene. Some described people with prediabetes whose raised HbA1c had stayed the same for years. Others thought progression was inevitable, irrespective of what was done.

…she had prediabetes and she didn’t want it to progress to diabetes so very early on she went on low dose Metformin … (but) your genetics get you in the end. (FG2, GP2)

Nurses felt their efforts should focus on those with higher levels of HbA1c and as a discipline, they seemed to be more optimistic that they could successfully support lifestyle change.

Yep 49 to 50 (HbA1c), I think more time needs to be spent with this group of patients because I think if we can catch them early and talk to them about lifestyle changes … the nurses hold a really important part in this process … (FG2, Nurse1)

GPs and nurses alike said prediabetes was not a priority in their work, often because it was just one condition among others that people with prediabetes had. In everyday practice, GPs and nurses struggled with fitting in the prediabetes conversation, judging when and how to raise it and what to cover.

…they’ve got the anxiety and asthma and we just throw in a wee bit of diabetes or prediabetes. In 15 minutes and that includes ‘oh hello, how’s it going, see you, bye’ and writing notes. And the script. (FG2, GP5)

Many GPs discounted actively intervening when the HbA1c had just entered the prediabetes range, saving their efforts for those about to progress to T2DM. Some did not tell people when their HbA1c was in the early 40s range, whereas others simply mentioned and minimised it.

(with prediabetes)… they’re not anywhere near the steep part of microvascular complication curve. Do you actually need to do anything more? Is this solving it, trying to address a problem that’s not there. Or do you wait till they dip a toe into diabetes and then you intervene? (FG3, GP4)

Both GPs’ and nurses’ uncertainty about prediabetes progression resulted in them being unsure about the best way to explain prediabetes and what to emphasise. They did not want to exaggerate the relative importance of prediabetes, but also wanted people to feel sufficiently concerned so they would make lifestyle changes.

…they’ve got to want to change. Yep so you can’t give them too much information because it all just comes in it’s just overload so I don’t know what works. (FG2, Nurse1)

Explanations ranged from the low-key: ‘blood sugars are a bit higher’ (FG3, Nurse1) to the assertive: ‘if you do not make some steps or make some changes it is very, very likely that you will go on to develop diabetes’. (FG1, GP2)

Broad advice was given around making ‘lifestyle changes’ and ‘taking action’. Both GPs and nurses focused on dietary advice, such as reducing portion size and foods high in fat and sugar, while increasing plant consumption and physical activity. They believed working with Māori and Pacific peoples about diet and weight loss was particularly challenging as cultural patterns of eating were incompatible with the more constrained diet needed to avoid prediabetes.

… there are other kinds of cultural food that they want to eat for breakfast which has lots of (sugar) like it’s coco rice. (FG1, Nurse1)

Many people with prediabetes were felt to be in the pre-contemplative phase of the motivation cycle.²⁵ Finding the right motivator might work, such as reminding them about another family member with T2DM consequences or staying healthy for the grandchildren. The uncertainty about whether GPs’ and nurses’ efforts were effective was reinforced when people cycled through changing and reverting lifestyle behaviour. As a fall-back, they ‘plugged away’ with prediabetes education, acknowledging this was an interim strategy.

…people go in bursts, they’re fine at the beginning and then it just eases off and then you know you go through the cycle again. (FG2, GP6)

Some people did not know they had prediabetes until invited to join this study. Several worried that they should have been informed earlier. Some were frightened by having prediabetes, but many were relatively untroubled. Others were uncertain about the importance of prediabetes and wondered about the volume and complexity of the information received or conversely that it was too vague.

… when I got told I had prediabetes … I had an appointment with the nurse. She was talking about exercise and watching your diet and that sort of thing, they’re all very vague sort of things. (FG4, 137, Asian)

Many of those who were Māori or Pacific felt certain they would develop T2DM, with some surprised they had not already developed it. This was particularly so for those with whānau (family) affected and they described health-related consequences of T2DM such as amputations, blindness and early death. A few wondered if they could stop the progression.

How long can you be prediabetic before you become diabetic or [can] you never become diabetic? (FG2, 053, NZ European)

Despite these uncertainties, most people recounted accurate information they had been told. This included the numbers in the HbA1c range, blood sugar elevation, relationship between prediabetes and T2DM, and need for lifestyle changes. They knew the reasons why they had prediabetes such as family heritage, dietary patterns, being overweight or obese and lack of physical activity. Very few described knowing nothing about prediabetes: ‘(I) don’t have a clue’. (Survey, 084, Māori).

Those who were Māori or Pacific felt the current public health messages were not effective.

Yes it’s the number one disease (T2DM)… and it’s been sort of publicised… all the marae- there’s posters up but there doesn’t seem to be a lot of change in people. (FG1, 024, Māori)

People described what they should do about prediabetes and differentiated between this and what they had done.

… if you change your diet you have positive impacts on being prediabetic. But a lot of people don’t change their diet because it’s just too much effort. (FG2, 050, NZ European)

Some described intentions to change soon.

… (changing) some of my habits such as having coffee without sugar, drinking coke no sugar. (Survey, 103, NZ European)

People with prediabetes discussed at length their uncertainties about lifestyle change. Many gave examples of failed attempts, often regarding weight loss.

And I was doing triathlons for a while (and) lost the weight and then I put the weight back on – … I lost 30 kilos I gained 15. (Int, 078, Māori)

Cultural mores were a strong influence, particularly for Māori and Pacific peoples. But there were also lifelong patterns of unhelpful eating, irrespective of ethnicity.

Unfortunately, I love food and our culture, yes we love food… With our culture- any big event … (if) there’s not enough food- it’s not a very good event. (Int, 029, Pacific)

Older Māori women participants actively championed change within whānau. Their efforts were not wholly well received, with seemingly light-hearted but pointed jokes expressed. Despite this, they persisted because they knew the long-term consequences.

I’m saying to them, I’m getting to them saying look you’re all over 50 now come on we’ve got to last, you’ve got a grandchild coming up so let’s do something about that, you want to be there until they’re married. (FG3, 078, Māori)

The viewpoints of all those who participated in the study had elements of both uncertainty and certainty about prediabetes and its management.

In general, GPs and nurses are uncertain about prediabetes. There was a lack of conviction that prediabetes interventions work. Limited time and possibly skill to implement interventions may reduce the ability of GPs and nurses to support individuals to achieve or sustain lifestyle change.²⁶ This contrasts with approaches used in large-scale international nation-wide prediabetes initiatives where this is a dedicated activity.^27,28

Prediabetes is a problematic diagnosis for many GPs, and current NZ guidelines/advice are unclear. GPs felt dissatisfied with the conversations they had about prediabetes and questioned if it was the best use of their time when other conditions were more important. They wanted tools to discern who would benefit from a more intensive approach. The lack of clear information reinforced their certainty about not acting at the lower HbA1c range or not necessarily telling people about their prediabetes diagnosis. Australia appears to have developed a more nuanced approach to prediabetes by establishing categories of those with prediabetes who are most at risk and targeting them for higher intensity interventions.²⁹

Nurses, in contrast to GPs, wanted people with prediabetes to have lifestyle counselling. Nurses were keener to do this work than GPs, and seemed more optimistic they could make a difference. Overseas and NZ studies have found the nursing skill set is a good match for prediabetes lifestyle counselling.^30,31 Coppell et al. report a NZ primary care nurse-led prediabetes brief intervention, which resulted in those in the intervention group successfully losing weight.¹⁸ The participants made dietary changes^32,33 in relation to goals set and the intervention was cost-effective.³⁴

GPs and nurses did not routinely advise people with prediabetes about weight loss, hinting that it would not be well received. Other NZ research on weight-loss conversations found GPs frame weight loss as ‘healthy eating’.³⁵ Some GPs and nurses explicitly acknowledged the more restrained forms of eating advised for those with prediabetes contravened the cultural norms of manaakitanga and hospitality held by Māori, Pacific and other ethnicities. More work is needed to find culturally relevant ways to modify food practices, as well as using findings from research already undertaken,^19,36–38 and from this, upskilling GPs and nurses with culturally sensitive nutrition education. There also needs to be a greater range of disciplines involved such as dietitians (particularly Māori and Pacific dietitians).

People with prediabetes were generally knowledgeable about the condition, and this may simultaneously reflect their volunteering to take part in the RCT study, and heightened awareness of prediabetes generated through study processes. They repeated the need to change diet and exercise and some mentioned weight loss. Although a few had made changes, many were uncertain and unconvinced they could change their behaviours. Some wanted more detailed advice, believing the information given was too vague, particularly about their personal risks for developing T2DM. The need for more precise information aligns with other studies about: nutrition guidance;^39,40 risk of prediabetes severity;^5,41 how to make and sustain lifestyle change decisions;^42,43 and advice specifically tailored to particular information needs.⁴⁴

There was variability in when and how the prediabetes diagnosis was given, with some people inadvertently finding this out, thus reducing the opportunity for them to change lifestyle behaviour earlier.⁴⁵ Not being told the diagnosis means peoples’ illness perception of prediabetes is not being elicited, particularly how social, emotional, cultural and health literacy factors impact on understanding.⁴⁶ Similarly, the goal-setting process, which is said to be a ‘central strategy for changing health behaviour’⁴⁴ and which others^33,47 have found important for motivating prediabetes lifestyle change, does not appear to be routinely undertaken.

Māori and Pacific peoples, bar some older women, felt certain they were on an inevitable path to developing T2DM and that nothing would stop this. A similar description of inevitability is reported by Faletau et al. in a NZ study aptly entitled ‘Falling into a deep dark hole’.¹⁷ Their study reported participants who were Tongan being fearful and in imminent danger from prediabetes, often conflating it with already having T2DM. For them, prediabetes and T2DM was experienced as a whānau, with interventions best delivered in a collective or community programme.⁴⁸ Co-designed initiatives by Māori or Pacific providers to change one’s diet are likely to be much more successful the cultural meanings of food and manaakitanga.⁴⁹

Older Māori and Pacific women in this study were adamant about the need to make significant dietary and exercise changes, and they championed these approaches with whānau. It may be better for Māori and Pacific people’s organisations to take on this championing work rather than individuals alone, as this risks leaving them disenfranchised from their whānau. Beaton et al. reported the role of Māori Health Organisations, in which kuia, senior older Māori women, play a strong role in spearheading this work.⁵⁰

As typical of qualitative studies, this study has sought to explore the views of a limited number of participants in some depth. The focus group interviews were confined to one geographic region and there were no rural participants. Some participants (GPs and nurses and people with prediabetes) were from the same practices. Although we could not be certain data saturation was reached, there was considerable commonality, particularly in the certainties and uncertainties described by all. The individuals with prediabetes in this study were taking part in a prediabetes intervention trial, and it is likely that the uncertainties reported in this study may be even greater in those not engaged in such a trial.

For GPs, nurses and people with prediabetes, there is uncertainty and certainty about prediabetes and its management. All were certain that prediabetes is increasingly common, and that sustained lifestyle change is difficult. But the following uncertainties were unresolved:

1. Can prediabetes be reversed or stabilised and in what circumstances?

2. Who with prediabetes progresses to T2DM and who does not, and where should people with prediabetes and GPs and nurses put their efforts?

3. Given there are highly respected older Māori and Pacific women who want to promote lifestyle change, how could Māori and Pacific organisations and leadership be better mobilised/supported?

4. Which co-designed culturally appropriate prediabetes initiatives should be further investigated, trialled and evaluated?

5. How important is weight loss and how do you get a population to lose weight rather than targeting individuals?

6. What other public health measures could be used to reverse/halt prediabetes?

Addressing and minimising these uncertainties for both health professionals and those with prediabetes would go a long way to improving appropriate and active patient-led, whānau-led management of a common condition that can be stabilised to prevent or slow the T2DM epidemic. Resourcing effective cultural and community-based leadership, unambiguous messaging and utilisation of best evidence already known within and beyond primary care services would all help reduce uncertainty, improve health, and start addressing significant health inequities.

The data that support this study cannot be publicly shared due to ethical or privacy reasons and may be shared upon reasonable request to the corresponding author if appropriate.

The authors declare no conflicts of interest.

This study was funded by the Health Research Council of New Zealand, the Ministry of Health New Zealand, and the Healthier Lives National Science Challenge (grant number 16/724).