Abstract
Background Evidence and guidelines increasingly support an individualised approach to care for people with type 2 diabetes and individualisation of glycaemic targets in response to patient factors.
Methods We undertook a scoping review of the literature for evidence of factors impacting upon glycated haemoglobin target individualisation in adults with type 2 diabetes. Data were analysed thematically with the themes inductively derived from article review.
Findings Evidence suggests that presence of cardiovascular disease, hypoglycaemia unawareness, severe hypoglycaemia, limited life expectancy, advanced age, long diabetes duration, frailty, cognitive impairment, disability, extensive comorbidity, diabetes distress and patient preference should inform the setting of glycaemic targets.
Conclusion The management of people with diabetes is complex. In clinical practice, many patients will have a variety of factors that should be considered when personalising their care. Approaches to personalised care and glycaemic treatment targets should be undertaken as part of a shared decision-making process between physician and patient. Use of electronic records might enable greater efficiency and more widespread use of personalised care plans for people with diabetes.
Background
As highlighted by the National Institute for Health and Care Excellence type 2 diabetes guidelines, the need for individualised care in diabetes is increasingly important.1 Diabetes prevalence in the UK is such that many patients are unlikely to receive regular specialist input for their diabetes care. Conversely, people with diabetes encounter non-specialists with greater frequency. It is important to consider individual characteristics of people with diabetes before agreeing on appropriate glycaemic targets. Discussing and agreeing individualised glycated haemoglobin (HbA1c) targets as part of care plans are paramount to improving patient experience and care.
The objective of this scoping review was to collect and discuss the evidence on the use of individualised HbA1c targets in people with type 2 diabetes and to identify any existing gaps in knowledge as areas of potential future research.
Methods
This review was prepared according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Scoping Reviews (PRISMA-ScR) statement and registered in the online Open Science Framework database (https://osf.io/snjpr) prior to data extraction.2
Search strategy
MEDLINE, AMED, PsycINFO and Embase were searched from inception to 01 June 2021. We completed a comprehensive search using free text and Medical Subject Headings (MeSH) for various forms of the following terms (in titles and abstracts): individualisation, glycaemic and target. The terms and truncated variants of the terms were combined for study retrieval. Additional articles were identified through backward and forward searching. The final search strategy can be found in supplementary material S1, Table S1.
Study selection
Publications were included if they were in the English language, included adult people with type 2 diabetes and were full text. Studies of any design were included to encompass the variety of factors that impact glycaemic target individualisation. Articles were excluded if they were not reporting an original study; had a small sample size (n<15) or did not cover themes relating to the individualisation of HbA1c targets in diabetes.
Study quality and data extraction
Two authors reviewed study quality of the included quantitative studies using the National Institutes of Health Quality Assessment Tools (QATs).3 The QATs used were study-design specific. For the included qualitative study, the Critical Skills Appraisal Programme qualitative checklist was used.4 Studies were rated as good, fair or poor depending on risk of bias. Data were extracted and tabulated on article characteristics (publication year, country of origin, number of participants and study type) and contextual factors (diabetes type and theme). Data were thematically analysed, with no formal quantitative synthesis taking place due to significant methodological heterogeneity in the included studies.
Results
Fig 1 details the records obtained from the search. After screening, 59 full-text articles were evaluated, of which, 11 were excluded for the reasons documented in Fig 1. Forty-eight studies were included following screening and review against eligibility criteria. Overall, risk of bias was rated low in 26 of the included studies, moderate in 16 and high in five (supplementary material S1, Table S2). The included qualitative study was rated good (supplementary material S1, Table S3).
Process of inclusion of studies and stages.
Study characteristics
The study characteristics are shown in Table 1.5–52 Sample sizes ranged from 28 to 264,687 (median 3,572; interquartile range 533–11,140.
Overview of included studies
Individualising glycaemic targets in response to patient factors
Full-text review of the articles revealed several emergent themes on the use of individualised HbA1c targets. Articles were coded according to theme. Concurrence of themes between articles resulted in the determination of key patient factors where individualised HbA1c targets were beneficial. These factors were:
Individualised HbA1c targets and established cardiovascular disease
Evidence from the UK Prospective Diabetes Study (UKPDS) in people with type 2 diabetes showed that, in their patient population, early intensive glycaemic control resulted in improved microvascular and macrovascular outcomes.5,7,14
The Action to Control Cardiovascular Risk in Diabetes (ACCORD), Action in Diabetes Vascular Disease: Preterax and Diamicron-MR Controlled Evaluation (ADVANCE) and Veterans Affairs Diabetes Trial (VADT) trials were subsequently undertaken to further evaluate the effects of intensive glycaemic control on outcomes in people with pre-existing type 2 diabetes.6,11,12 Three and a half years in, the ACCORD trial was halted due to increased all-cause mortality seen in the intensive control (HbA1c 47 mmol/mol (6.4%)) group. Despite having similar objectives, ADVANCE and VADT trials showed no difference in macrovascular outcomes. Explanations for the differences in outcomes seen in these trials vary and uncertainty remains.
In the ACCORD trial, patients experiencing severe hypoglycaemia, whether in the intensive or standard glycaemic control arms, were noted to have increased mortality rates. These data have been echoed in the Outcome Reduction with Initial Glargine Intervention (ORIGIN) trial that showed severe hypoglycaemia was associated with an increased risk of a composite major adverse cardiac event (MACE; cardiovascular (CV) death, non-fatal myocardial infarction (MI) or stroke), all-cause mortality, CV death (separate from the composite event) and arrhythmic death in people with CV risk factors and type 2 diabetes.8,13
Since the results of these trials, pharmaceutical treatment options have advanced dramatically. Newer agents such as glucagon-like peptide-1 receptor agonists (GLP-1RAs) and sodium-glucose co-transporter-2 inhibitors (SGLT2is) demonstrate cardiovascular and mortality benefits over older treatment options studied in ACCORD, ADVANCE and VADT trials with comparable efficacy on HbA1c levels.53 Despite this, the legacy of the ACCORD trial has meant that clinicians must practise caution in applying intensive HbA1c targets to those at risk of CV disease due to the additional risk of severe hypoglycaemia associated with achieving intensive glycaemic control.10 Alongside the use of individualised HbA1c targets, aggressive modification of all CV risk factors (such as blood pressure and lipid modification) is crucial in reducing long-term CV mortality in people with diabetes. In those with risk factors for CV disease, or with pre-existing CV disease, adjustment of HbA1c treatment targets to avoid severe hypoglycaemia should be considered alongside consideration of switching to GLP-1RAs or SGLT2is in those with increased CV risk, heart failure or chronic kidney disease (CKD).9
Individualised HbA1c targets and advancing age and diabetes duration
The care of older adults presents unique challenges. There is conflicting evidence on whether clinicians over- or under-treat diabetes in the elderly.22,24 Those with advanced age are more likely to have a longer duration of diabetes, higher risk of hypoglycaemia, higher levels of CV comorbidity, higher levels of inpatient and outpatient service utilisation, and inappropriately intensive treatment for their diabetes.14–16,19–21,23,24
Follow-up data from the ACCORD, ADVANCE and VADT trials suggest elderly patients with a longer duration of diabetes are unlikely to gain macrovascular benefits from intensive glycaemic control and may be exposed to excess risk of severe hypoglycaemia, increased morbidity and mortality.6,11,12,18
These findings are expanded upon in a study by Monami et al: follow-up over 6 years in those with a longer diabetes duration (10 years and over) showed that mortality only increases when HbA1c levels are greater than 68 mmol/mol (>8.4%).25 Similarly, over the same follow-up when considering those aged 71 years and over, mortality was shown to increase only when HbA1c levels rose above 68 mmol/mol (>8.4%).
Despite guidance on glycaemic target individualisation and lack of macrovascular benefits, a European study by Strain et al on factors affecting physician glycaemic target-setting behaviours for elderly patients (aged 70 years and over) showed that rigid, particularly aggressive, uniform glycaemic targets are still commonly used in line with national performance indicators.17,54 It is likely that a historical lack of consensus among guidelines and difficulty in accounting for the possible risks and benefits of adjusting glycaemic targets are contributory factors.47
Bearing in mind the diminishing returns of improving glycaemic control, safe, effective treatment of elderly patients with a longer duration of diabetes requires constant re-evaluation of the expected gains in macrovascular risk reduction versus the expected risks of intensive glycaemic control. A relaxed glycaemic target between 58 and 69 mmol/mol (7.5%–8.5%) aimed at avoiding hypoglycaemia and uncontrolled hyperglycaemia with an individualised care plan should be considered in older patients with established diabetes of long duration.17 Excess mortality would be better addressed through modification of other reversible CV risks, such as lipid and blood pressure control in these patients.55,56 There is limited follow-up data in younger people (aged under 60 years) with type 2 diabetes of long duration. Further work is needed in determining appropriate HbA1c targets in this group.
Individualised HbA1c targets and frailty, disability, cognitive impairment and comorbidity
Frailty, disability, cognitive impairment and comorbidity are often seen to be interrelated, with a degree of overlap and are increasingly prevalent in the western world as the population ages.57 Lower HbA1c values are both a risk factor for developing frailty and in those with frailty, and are associated with an increased risk of stroke, dementia and mortality.36,28 In patients aged 60–64 years with disability and newly diagnosed type 2 diabetes, the benefits of intensive glycaemic control in those with only low levels of functional impairment are marginal at best (106 quality-adjusted days).58 People with diabetes have a two-to-three-fold increased odds of disability irrespective of glycaemic control, cardiovascular disease and obesity are seen as the main contributors.29,31 Sub-optimal glycaemic control alone is not a significant predictor for disability and should not be the main consideration when agreeing appropriate HbA1c targets with patients.34 Alongside frailty and disability, presence of cognitive impairment in people with type 2 diabetes significantly increases the risk of severe hypoglycaemia, major CV events, CV death and all-cause death.30,34 Comorbid patients with diabetes would be expected to have a shorter length of life with a subsequent reduction in time for the development of diabetes complications.59 In studies evaluating older people with diabetes with cardiovascular comorbidities, intensive glycaemic control has shown no mortality benefit and has not resulted in a reduction of further CV endpoints.6,11,12,59
The complexity of accounting for these variables has resulted in a conflicting evidence base on the association between HbA1c and mortality.26,27,32,33,35 As such, agreeing appropriate HbA1c targets with patients is highly nuanced. In general, for otherwise healthy older adults, a target of <58 mmol/mol (<7.5%) probably reflects the best compromise between risk and benefit. An individualised glycaemic target between 58 and 69 mmol/mol (7.5%–8.5%) to avoid hypoglycaemia, symptomatic hyperglycaemia and medication burden should be considered in adults with co-existing frailty, disability, comorbidity or cognitive impairment.59,60
The use of individualised HbA1c targets and problematic hypoglycaemia
Severe hypoglycaemia has been suggested as one of the reasons why the ‘intensive glycaemic control’ arm of the ACCORD trial was noted to have excess mortality, though no direct causal relationship has been established.59 Retrospective analysis of the ADVANCE dataset by Zoungas et al showed that severe hypoglycaemia in patients with type 2 diabetes was associated with a statistically significant increase in the risk of major macrovascular events (hazard ratio (HR) 2.88; 95% confidence interval (CI) 2.01–4.12), major microvascular events (HR 1.81; 95% CI 1.19–2.74), CV death (HR 2.68; 95% CI 1.72–4.19) and death from any cause (HR 2.69; 95% CI 1.97–3.67; all p<0.001).37 It is possible that severe hypoglycaemia is contributory to these outcomes but it is more likely that severe hypoglycaemia is a general marker of clinical vulnerability in these individuals.
The risk of hypoglycaemia increases independently with advancing age, duration of diabetes and presence of CKD.41,42 Symptomatic hypoglycaemia in diabetes, whether mild or severe, is a significant source of hospitalisation (HR 2.09; 95% CI 1.63–2.67) and death (HR 2.48; 95% CI 1.41–4.38) and is associated with increased morbidity, all-site cancer, disability, medical visits, diabetes-related medical costs, medication costs, healthcare resource utilisation, and reduced quality of life, well-being and self-management.18,20,38,40,42,44,61 Independent of glycaemic control, comorbid status and diabetes treatment, hypoglycaemia is associated with a greater risk of dementia (2.39% per year; 95% CI 1.72–3.01).45 In older adults with Alzheimer's dementia, Chen et al showed a reduced progression of dementia, reduced rate of hypoglycaemia, reduced medication burden and reduced rate of diabetes complications over a 3-year follow-up period in patients following a moderate rather than intensive glucose control strategy.39 Risk of severe hypoglycaemia in type 2 diabetes is highest in those achieving near-normal glycaemia (HbA1c <42 mmol/mol (<6.0%)) or with very poor glycaemic control (HbA1c ≥75 mmol/mol (≥9.0%)).43
Since the data presented by ACCORD and ORIGIN trials, it is recognised that glycaemic targets for patients with hypoglycaemia unawareness or preceding severe hypoglycaemia should be individualised to avoid hypoglycaemia at the expense of a relaxed HbA1c target. Special care should be taken in the management of comorbid patients and patients with longer diabetes durations, such as the demographics of the patients in the ACCORD study. The unique clinical course of each patient reinforces the need to individualise glycaemic targets in response to hypoglycaemia risk.41 A reasonable suggestion is to assign an individualised glycaemic target that avoids severe hypoglycaemia and preserves hypoglycaemia awareness. This may mean that in younger, healthier patients whose diabetes is controlled with dietary and lifestyle interventions alone, a non-diabetic glycaemic target (<48 mmol/mol; <6.5%) may be appropriate but in patients with limited life expectancy, a higher HbA1c target (<69 mmol/mol; <8.5%) sufficient to prevent the symptoms of hyperglycaemia would be acceptable. Those in the intermediary zone who would not necessarily be in ‘good’ health, but whose life expectancy is not limited may thus benefit from a glycaemic target between 58 and 69 mmol/mol (7.5%–8.5%), depending on individual circumstances.
The use of individualised HbA1c targets and psychosocioeconomic concerns
As a chronic disease process, diabetes is increasingly recognised to have a significant impact on psychological outcomes and mental health. Studies show that more than one-third of people with diabetes have depression at a level that impairs functioning, quality of life, adherence to medical treatment, glycaemic control, and increases healthcare utilisation, healthcare cost and the risk of diabetic complications.48,49,62 Coexistent depression and diabetes increases the risk of death from all causes in excess of the summative effects of having either condition in isolation.48,49,62
Studies in older people with diabetes (aged 65 years and over) show that vulnerable adults have even greater levels of depression, have increased concern related to medication side effects, have trouble remembering to take medications, have required increased assistance with medication taking, feel overwhelmed following visits to clinicians, find taking their diabetes medications unpleasant and are less willing to take insulin.46,50 Depending on how burdensome an individual views their treatment, improvements in glycaemic control can result in net harm and reduced quality of life (despite improved HbA1c) in older adults.56 A cross-sectional study by Chin et al evaluated the preferences of older adults (aged 65 years and over) with diabetes regarding the quality of life trade-offs between aggressive glycaemic control and the avoidance of diabetes complications.47 The study found that standard glycaemic targets were acceptable for most patients but that, where treatment negatively impacted upon quality of life or where the gains in quality of life were neutral, standard glycaemic targets were problematic. Reinforcement of the importance of a dialogue between patients and physicians in a shared decision-making process to include consideration of overall lifegoals, patient preferences towards different treatment approaches and diabetic complications is paramount.
Quality of life in people with diabetes is impacted by the adverse effects of diabetes treatments as well as the route of treatment delivery (injected or oral).47 Reductions in quality of life due to diabetes treatments can be large, with wide inter-person variation.46,52 Modelling studies of the NHANES diabetes population (2011–2012, aged 25–75 years) shows that the individualisation of glycaemic targets according to risk of future complications and patient age is cost saving (mainly due to reduced medication usage) and results in gains in quality-adjusted life-years (mainly due to reduced medication burden in the over-treated) over the course of a lifetime without substantially impacting patient outcomes.63
Exploratory studies into diabetes healthcare goals that are most important for patients describe social and functional goals rather than biochemical goals targeting risks and complications.51 A shared decision-making process that takes social and functional goals into account may be an approach that is key to ensuring the successful implementation of individualised diabetes care.15
Discussion
Current diabetes literature and up-to-date evidence-based guidelines report on the importance of using individualised HbA1c targets for people with diabetes. Our review discusses the evidence on individualised HbA1c targets in response to established cardiovascular disease, advanced age, long diabetes duration, frailty, disability, cognitive impairment, presence of comorbidity, problematic hypoglycaemia and psychosocioeconomic considerations. We believe a considered approach should be taken before agreeing an individualised HbA1c target, taking into account an informed opinion from patients on the respective risks and benefits of higher and lower HbA1c targets, alongside review of the presence of relevant characteristics that may influence decision making (Fig 2). Recurrent themes in the reviewed literature demonstrated the importance of a multifactorial approach to micro- and macrovascular risk management, ensuring lipid modification and blood pressure management are optimised alongside using individualised HbA1c targets.
Decision aid in patient–physician encounters when mutually agreeing an individualised glycated haemoglobin target. People with diabetes should be fully informed wherever possible to reach a shared decision with their physician on a target appropriate for them based on their characteristics. DCCT = Diabetes Control and Complications Trial; HbA1c = glycated haemoglobin; IFCC = International Federation of Clinical Chemistry.
In the UK, national diabetes quality performance indicators target a specific HbA1c at a population level.64 While this may be useful at reducing population-level risk, population-level HbA1c performance indicators do not adequately consider the additional costs and adverse effects of a uniform glycaemic target to a heterogeneous diabetes population.43 Furthermore, these indicators are not consistent with widely published evidence-based guidelines encouraging individualised approaches.54 We suggest implementation of additional quality indicators; for example, flagging low HbA1c values in patients with frailty as a marker of over-treatment to encourage appropriate glycaemic target individualisation.
Strengths and weaknesses
The majority of the studies (26) included in this review were rated as having a low risk of bias. Due to the diverse nature of the topics included in this review, often including patient populations that are difficult to recruit, we decided against excluding studies where the risk of bias was moderate or high, instead accepting this as a limitation.
Individualised care and glycaemic targets are equally as important in type 1 diabetes but can often differ with care plans and targets used in people with type 2 diabetes. We excluded studies referring to type 1 diabetes only as this area of diabetes care remains largely in the realm of specialists.
Gaps in the literature remain on the evaluation of the impact of using individualised glycaemic targets on healthcare outcomes for people with type 2 diabetes. Few studies between 2012 and 2018 have evaluated this, with some useful insights: individualised glycaemic targets are cost-effective; improve quality-adjusted life-years; reduce rates of severe hypoglycaemia, medication burden and healthcare utilisation; and increase glycaemic target-achievement.17,63,65–68
Conclusion
The management of people with diabetes is complex. In clinical practice, many patients will have a variety of factors (Fig 3) that should be considered when personalising their care and assigning individualised glycaemic targets. Our findings suggest that a significant body of evidence exists for adjusting glycaemic targets in response to individual patient factors. Approaches to personalised care and glycaemic treatment target setting need to be undertaken as part of a shared decision-making process between physician and patient. Further efforts are needed to improve practice and to adjust national performance measures that incentivise the pursuit of uniform tight glycaemic targets. Future work evaluating the impact of using individualised glycaemic targets in people with diabetes and on the use of electronic records as a tool to aid this process could enable increased efficiency and more widespread use of personalised care plans in diabetes.67
Patient factors that should prompt a re-evaluation of glycaemic goals in people with diabetes. HbA1c = glycated haemoglobin; GLP-1RA = glucagon-like peptide-1 receptor agonist; SGLT2i = sodium-glucose co-transporter-2 inhibitor.
Key points
Use of individualised glycaemic targets in people with type 2 diabetes is endorsed by national guidelines.
Current guidelines are non-specific regarding the decision-making process for adjusting glycaemic targets.
Individualising glycaemic targets should be considered as part of a shared decision-making process between physician and patient.
A variety of patient characteristics should prompt a re-evaluation of appropriate glycated haemoglobin targets by physicians.
Agreeing on glycated haemoglobin targets with patients is highly nuanced. Factors such as established cardiovascular disease, diabetes duration, life expectancy, episodes of severe hypoglycaemia, hypoglycaemia unawareness, presence of significant comorbidity, and presence of psychological or social concerns should be considered.
Supplementary material
Additional supplementary material may be found in the online version of this article at www.rcpjournals.org/clinmedicine:
S1 – Search strategy and study quality.
- © Royal College of Physicians 2022. All rights reserved.
References
Jump to section
Related Articles
Cited By...
- No citing articles found.