The evolving continuum of dysglycaemia: Non‐diabetic hyperglycaemia in older adults

Identifying non‐diabetic hyperglycaemia (NDH) and intervening to halt the progression to type 2 diabetes has become an essential component of cardiovascular and cerebrovascular risk reduction. Diabetes prevention programs have been instigated to address the increasing prevalence of NDH and type 2 diabetes by targeting lifestyle modifications. Evidence suggests that the risk of progression from NDH to type 2 diabetes declines with age, and that a diagnosis of type 2 diabetes in older adults is not associated with the same risk of adverse consequences as it is in younger age groups. The current definition of NDH is not adjusted based on a person's age. Therefore, there is debate about the emphasis that should be placed upon a diagnosis of NDH in older adults. This article will explore the evidence and current clinical practice surrounding dysglycaemia through the spectrum of different age ranges, and the potential implications this has for older adults.

remains the simplest test to conduct, as it does not require fasting or standardised glucose loads, and its measurement can be added to routine blood tests for other conditions.Clinicians recognise the importance of identifying individuals with undiagnosed diabetes and NDH in order to treat modifiable cardiovascular risk factors.As a result, although it was never intended to be used in this manner, HbA1c has become used as a screening test for diabetes.Wilson and Jungner's criteria for an effective screening test can be applied to the use of HbA1c for diabetes screening in the older adult population, as shown in Table 2. 8 Multiple factors can affect the relationship between HbA1c and its representation of average glucose which affect its utility as a diagnostic test.][11] Altered renal function has also been shown to alter the reliability of HbA1c. 12rucially, even when correcting for such factors, numerous studies have shown that HbA1c increases with age, including when controlling for the recognised rise in fasting plasma glucose and 2-h post-load glucose in older adults. 3,13,14A consequence of this may be over-diagnosis of NDH and diabetes in older people, as well as potential over-treatment for those with a diabetes diagnosis.One group assessing the use of HbA1c in older adults in China has suggested that HbA1c is unsuitable for diagnosing diabetes in the elderly. 15n line with Wilson and Jungner's criteria, if there is ever any doubt about the validity of HbA1c as a screening test, then an alternative test should be used.Fasting and stimulated glucose tolerance tests may be more accurate, however for practical reasons, HbA1c will realistically continue to be an important and often-used marker of glycaemia.Understanding how HbA1c is affected by individual patient circumstances is thus of vital importance when interpreting it.Of note, HbA1c is also used as a marker of glycaemia in clinical trials, and the implications of this are important for research into the older adult cohort.This should be kept in mind for the remainder of this article.

adults-up to the age of 65
It is predicted that by 2045 the global prevalence of type 2 diabetes will be in excess of 700 million. 16In England, the

What's new
• Elevations in HbA1c do not appear to have the same impact on outcomes in older adults as they do in younger individuals.• Pragmatic approach towards mild elevations in glucose levels are required in older adults as the consequences for older adults are not fully understood.• Research identifying the impact of mild elevations in glycaemia in older adults will add further to the debate about whether criteria for non-diabetic hyperglycaemia (NDH) should be age-or frailty-adjusted.

T A B L E 2
The Wilson and Jungner criteria for screening.

The Wilson and Jungner criteria for screening
The condition sought should be an important health problem There should be an accepted treatment for patients with recognised disease Facilities for diagnosis and treatment should be available There should be a recognisable latent or early symptomatic stage There should be a suitable test or examination The test should be acceptable to the population The natural history of the condition, including development from latent to declared disease, should be adequately understood There should be an agreed policy on whom to treat as patients The cost of case finding (including diagnosis and treatments of patients diagnosed) should be economically balanced in relation to possible expenditure on medical care as a whole Case-finding should be a continuing process and not a 'once and for all' project Note: Highlighted in bold are two of the potential pitfalls of HbA1c use for the diagnosis of NDH in older adults.
prevalence of NDH is rapidly increasing such that over a third of adults are considered to have it. 17NDH represents the intermediate stage of glucose dysregulation between normal glucose and diabetes. 18The term was developed to attempt to identify those individuals who are at increased risk of developing type 2 diabetes.Identifying and intervening early to prevent progression to diabetes is vitally important as evidence suggests early intervention has the greatest impact on cardiovascular outcomes. 19In the UK, Diabetes Prevention Programs have been developed to initiate lifestyle interventions aimed at preventing and delaying the onset of type 2 diabetes.One of the difficulties faced is a lack of consensus on the definition of NDH.Different professional bodies have different biochemical cut offs using various tests for hyperglycaemia-HbA1c, IFG and IGT (see Table 1).These definitions result in significant prevalence differences and identify different groups of people who will have different risks of developing diabetes-making comparisons across research difficult. 20espite this challenge, data demonstrate that even minor changes to glycaemia, such as those seen in NDH, can have a profound impact on outcomes in adults.A recent meta-analysis showed an increased risk of composite cardiovascular disease, coronary artery disease, stroke and all-cause mortality with each of the different biochemical cut offs used, though there was no analysis of an older adult cohort within this. 216][27] There are less data looking at the regression of NDH to normoglycaemia.However, studies looking at interventions for NDH have shown that lifestyle interventions or medications can be effective at reversing NDH, with more long-lasting effects gained from weight reduction, diet and exercise. 27-29

| NDH in older adults
While the importance of identifying and treating NDH in younger or middle-aged adults may not be disputed, the picture is more complicated in the older adult cohort.One study showed the prevalence of NDH in those aged over 75 in the United States (diagnosed by IGF or IGT) to be nearly 50%. 30,31The prevalence of diabetes and NDH together in this age group as defined by this study was 75.7%. 30When the prevalence of a condition is so high within a population, it is worth considering whether this is pathogenic or part of a spectrum of normality.

| Complications associated with a diagnosis of NDH in later life
Studies have investigated whether a diagnosis of NDH as an older adult has the same implications as it would for a younger individual.An analysis of the Atherosclerosis Risk in Communities (ARIC) study, compared cardiovascular and mortality data between those with diabetes of different durations and those with NDH and normoglycaemia in the United States.The study found that, in contrast to analyses looking at all ages, older adults (ages 66-90) with NDH (HbA1c 39-48 mmol/mol) had a similar risk of cardiovascular disease and mortality as those with normoglycaemia.Among those with diabetes, increased diabetes duration was significantly correlated with mortality. 32It is important to remember that the American Diabetes Association definition of NDH was used in this study which uses slightly lower HbA1c cut-offs.It is possible that a greater difference might have been observed if higher cut-offs had been used, such as those in use in the UK (Table 1).
A further prospective analysis of 5636 participants of the ARIC study confirmed that for older adults without diabetes, there was no increase in mortality with either (NDH-range) elevations in HbA1c or fasting glucose. 33hese data suggest that a diagnosis of NDH later in life may not be associated with the same risk of diabetes complications as it would be if the diagnosis was made in a younger individual.

| Complications associated with a diagnosis of type 2 diabetes in later life
The main purpose of identifying individuals with NDH is to identify those at risk of developing type 2 diabetes.Understanding the implications of a diagnosis of type 2 diabetes on older adults is therefore also of relevance in understanding the importance of identifying NDH in this age group.
A recently published longitudinal cohort study of 36,060 adults over the age of 50 assessed the cardiovascular outcomes of older adults who developed diabetes (selfreported) at different ages over 50 with a matched control group who did not develop diabetes. 34Participants were part of the Health and Retirement Survey (HRS), a longitudinal survey of older adults in the United States.It showed that individuals diagnosed with diabetes at younger age had a significantly increased risk of heart disease, stroke, disability, impairment, and all-cause mortality.This risk significantly reduced as the age of diabetes diagnosis increased, such that the risk of heart disease, stroke, disability and cognitive impairment was comparable in those over 70 years with a new diagnosis of diabetes and those of the same age without diabetes.There was only a slight overall increase in mortality in those over 70 with a new diagnosis of diabetes compared to matched controls without diabetes (HR, 1.08 [95% CI, 1.01-1.17]). 34 review of the Swedish National Diabetes Registry involving over 300,000 individuals with type 2 diabetes and around 1.6 million controls without diabetes, supported these findings. 35In this study, the risk of cardiovascular mortality, non-cardiovascular mortality, heart failure, coronary artery disease and stroke all reduced as the age of diabetes diagnosis increased-such that in this study the mortality of those who were diagnosed after the age of 80 was the same as those without a diagnosis of diabetes. 35 systematic review and meta-analysis of 26 studies, published in 2021, also examined this association, including over a million participants with type 2 diabetes.It showed that younger age of diabetes diagnosis was strongly correlated with microvascular complications, macrovascular complications, and all-cause mortality.An older age of diabetes onset is associated with a lower risk of microvascular complication (5% reduction per year), macrovascular complications (3% reduction per year) and a lower risk of allcause mortality (4% reduction per year). 36hus, a new diagnosis of type 2 diabetes in an older person may have less cardiovascular implications than in a younger adult.By extrapolation, this means a diagnosis of NDH may be relatively less important in older adults as the purpose of identifying NDH is to predict those at risk of progression to type 2 diabetes.
Overall, there is a suggestion that the relationship between glycaemia and the development of adverse consequences is more nuanced in the older adult population.Further studies are required to explore this relationship further, including any possible reasons for this.

| Progression of NDH to type 2 diabetes in later life
There is also strong evidence to suggest that a diagnosis of NDH in older adults is not associated with the same risk of progression to type 2 diabetes as in younger adults.A population-based cohort study in Sweden followed up 2575 adults over age 60 for 12 years.22% of those with NDH (defined as HbA1c of 39-47 mmol/mol) reverted to normoglycaemia, and 13% progressed to type 2 diabetes.23% died during follow-up. 2In this study, regression to normoglycaemia was thus more likely than progression to type 2 diabetes.
A cohort study assessing electronic health records in the UK found that among more than 14,000 adults with NDH (defined in line with NICE criteria-see Table 1), the risk of progression to diabetes reduced with increasing age.In this study, the risk of progressing to type 2 diabetes was highest in those diagnosed with NDH aged 45-54, HR1.20 (95% CI 1.15-1.25).The risk reduced with age with an HR in the aged over 85 group of 0.65 (95% 0.60-0.71). 1 A retrospective analysis of the English Longitudinal Study on Ageing (ELSA) defined NDH as an HbA1c between 39 and 47 mmol/mol, or a fasting glucose between 5.6 and 7.0 mmol/L. 372027 adults over 60 years old were included for analysis which included data collection at baseline and after 10 years.Four hundred and seventynine participants were found to have NDH according to HbA1c measurement.After 10 years, 180 of these had regressed to normoglycaemia, and 59 progressed to diabetes, while 56 died.The remainder continued to have an HbA1c in NDH range.When looking at NDH as defined by IFG, 225 individuals were found to have NDH at baseline.After 10 years, 131 regressed to normoglycaemia, 35 progressed to diabetes and 31 died.This analysis again showed that regression to normoglycaemia was more common than progression to diabetes in older adults (over 60) diagnosed with NDH-regardless of whether NDH was defined by HbA1c, or by IFG. 37uch data are not completely consistent however.Another analysis of six prospective studies found an inconsistent relationship between age and progression from NDH to diabetes. 38The authors postulated that this may be because of the diversity of the populations studied.The two studies that showed an increase in progression to type 2 diabetes with increasing age had very slight increases and very low overall progression rates. 38ogether, these data suggest that in older adults, the diagnosis of NDH is not as important for predicting the risk of progression to type 2 diabetes.Importantly, these studies use a range of NDH definitions and, in the most part, the findings correlate well.

| Frailty
Older adults are a heterogeneous population.While some will be healthy, others will have complex health needs.Physiological changes occur with ageing, including frailty, which describes physiological decline and loss of reserve across organ systems that leaves an individual vulnerable to stressors which trigger deteriorations in health or functional status. 39,40The presence of frailty itself is associated with increased mortality. 41Those living with frailty and diabetes are unlikely to survive long enough to benefit from the long-term results of tight glycaemic control, while being more susceptible to the adverse consequences of hypoglycaemia. 42ssessment of frailty during consultations is important to guide therapeutic decisions and is perhaps even more relevant than just looking at someone's age. 39It has been suggested that glycaemic targets in the management of type 2 diabetes should be adapted depending on frailty status. 9This is partly due to the recognised harms of overtreatment, and the subsequent consequences of hypoglycaemia in this population.A meta-analysis has shown an increased risk of micro-and macrovascular complications, falls, fractures and mortality with increasing episodes of hypoglycaemia in older adults. 42The harms of hypoglycaemia in individuals living with frailty are compounded by the loss of hypoglycaemia awareness and reduced physiological reserve.
The recognition of the harms of over-treatment in the management of older adults with diabetes adds further weight to the view that normal glucose levels tend to gradually increase in the healthy older adult population.Although there are mechanisms to adjust treatment targets for diabetes in response to advancing frailty, there is no such alteration to diagnostic cut-offs for diabetes or NDH in response to changing physiology in older adults.

ADULTS
We have highlighted the key differences in the impact of dysglycaemia between middle-aged adults and older adults.These include the comparatively high prevalence of NDH and type 2 diabetes in older adults by current definitions; evidence of reduced impact of mild dysglycaemia in older adults-both in terms of progression to diabetes and diabetic complications; and evidence of increased harm of over-treatment of diabetes in older adults.Mild alterations in glycaemia may therefore be a normal part of the ageing process, and there is on-going debate around whether it is appropriate for older adults to be labelled as having NDH. 43,44aving a 'diagnosis' can motivate people to bring about positive change, such as changes to lifestyle, but also has the potential for adverse consequences including denial, resistance to treatment and anxiety. 45It is important to consider the unintended consequences of investigating, and making a diagnosis-in the same way, doing a CTscan can uncover unexpected benign incidental findings.Important factors to consider include the psychological impact on the individual, whether the diagnosis is likely to result in a beneficial change in management, and impacts on health economics including clinician time and cost of investigations.Ultimately, the management of NDH or early type 2 diabetes in a robust older adult without frailty would be lifestyle advice, including exercise and weight management strategies, advice which is already widely considered to be beneficial for the population as a whole.However, the same diagnosis in an older adult living with frailty may even have negative consequences, as any suggestion to calorie restrict or induce weight loss might even worsen sarcopenia and exacerbate frailty. 46Some have thus suggested that clinicians should think carefully before informing a patient about an NDH 'diagnosis,' due to the possible harm that this could cause for very little benefit. 47Feasibly, limiting the term NDH to individuals under a certain age (e.g.under the age of 65) could be a more practical solution and a way of acknowledging the differences between younger adults and older adults in the evidence, however, this blanket approach could be seen as paternalistic, and deny physically fit older adults the opportunity of addressing potential risk factors.
When considering how this relates to the screening criteria outlined in Table 2, there are clear question marks surrounding the use of HbA1c to screen older adults for NDH.The test itself is likely to over-estimate the true value of the average glucose in an individual and therefore overdiagnosis is likely to occur.In addition, as mentioned, the evidence presented shows that the natural history of NDH in older adults is not adequately understood.In fact, it is likely that a 'diagnosis' of NDH will have no consequence at all for the majority of older adults.
How clinicians respond to a new diagnosis of NDH is also important.Figures from the National Diabetes Audit in 2021 suggest that over half of all new referrals to the UK Diabetes Prevention Program are over the age of 65, and 11.4% of referrals are over the age of 80. 48 The evidence presented in this article suggests that the majority of people in these age ranges would not have progressed to type 2 diabetes anyway, and for those that do develop diabetes at this age it would likely have very little consequence.At present the UK Diabetes Prevention Program has no upper age limit for referral but instead uses the discretion of the referrer as to whether a referral is appropriate.Further guidance for clinicians on the appropriateness and value of such interventions would be useful to ensure resources are targeted effectively.
There is precedence within the endocrine system for such a relationship between declines in hormone levels associated with healthy ageing (as opposed to those with recognised hormone deficiencies), and benefits of hormone replacement.It is well recognised that growth hormone levels decline with age, but treatment with growth hormone does not confer the benefits that might be expected in younger adults. 49Similarly, testosterone levels in men are known to decline with age, yet replacement even in men with symptoms consistent testosterone deficiency, provides minimal benefit, while exposure to the potential side effects remain. 50

| CONSIDERATIONS AND FUTURE RESEARCH
The importance of identifying younger and middle-aged adults with NDH, and intervening to halt its progression to type 2 diabetes, remains essential and has clear benefits for both individuals and society.However, the use of this term becomes problematic in the older adult population, particularly if uniform cut-offs are applied across the spectrum of age ranges.
As outlined, question marks persist about the relevance of identifying NDH in older adults, as there is limited evidence of harm in relation to diabetes-related complications or mortality for those diagnosed with NDH.Although HbA1c remains useful in assessing trends in glycaemia over time in an individual, there are also concerns about its use as a diagnostic test for NDH in this population, and even whether diagnosing NDH is worthwhile at all in this older population, particularly in those who are experiencing frailty.As such it may be time to disassociate the term NDH entirely from older adults or consider age-or frailty-specific diagnostic ranges.This could enable resources to be directed better to those who are likely to benefit the most.More research into the reasons why older adults appear to have fewer classically recognised adverse outcomes from mild glucose dysregulation outside of the diabetic range would be useful, as too would longitudinal follow-up data in such populations.Qualitative studies assessing the views of older adults about the use of the term NDH, including how they perceive it, will add insights to our understanding about whether diagnosing someone as having NDH may have a positive or negative impact on individuals.
Further research both into how normal ageing affects glucose levels, and how frailty affects glucose levels will have further implications in the management of people with type 1 and type 2 diabetes.In the management of these conditions, we are often trying to achieve a 'normal' glucose, or at least a glucose level that prevents complications, but this is difficult to achieve when it is still unclear what a 'normal' glucose is in this population.The harms of over treatment in older adults with diabetes are well known, and it might be the case that due to a lack of evidence, we are trying to achieve sub-physiological glucose levels.Even with a diagnosis of diabetes established in the older adult, a pragmatic approach towards the individual's life expectancy, taking into account co-morbidities, and frailty level should help place any elevated HbA1c into context.

| CONCLUSION
There remains a lack of understanding and consensus regarding NDH in older adults and it is clear that further research in this area is required to understand the implications of a diagnosis, for example, in relation to its risk of progression to type 2 diabetes, and any microvascular, macrovascular or mortality outcomes.It is conceivable that HbA1c thresholds for diagnosis of NDH and for diabetes should be raised in conjunction with increasing age or frailty status.Until then, a pragmatic and individualised approach to older adults with NDH should be employed.

1 |
NDH in young and middle-aged