Guidelines from the UK National Institute of Health and Care Excellence advocate for patients with type 1 diabetes to aim for near-normoglycemia (ie, 5–7 mmol/L at fasting, 4–7 mmol/L pre-meal, and 5–9 mmol/L post-meal) to achieve a glycated hemoglobin (HbA1c) of ≤48 mmol/mol [1]. Additionally, in the context of non-pregnancy, the guidelines advise patients and clinicians to collaborate in setting personalized glycemic goals, taking into account factors including comorbidities, likelihood of complications, patient’s motivation, occupation, and history of hypoglycemia.
A gap in type 1 diabetes care planning?
The indication from the recent National Diabetes Audit is that goal-setting is not standard practice [2]. One in six patients with type 1 diabetes did not have an up-to-date HbA1c. Against a background of suboptimal control (70% had a HbA1c of >58 mmol/mol), only 40% of patients were offered increased support through structured education, and only 3.2% of patients attended. This is a missed opportunity.
Establishing individualized goals has been shown to improve healthcare outcomes in patients with multimorbidities [3] and diabetes [4]. The psychological mechanisms associated with improvement include the prioritization of effort, persistence, motivation, and the consolidation of acquired knowledge and strategies [5]. Setting goals and providing educational support leads to adherence and promotion of self-management skills in adolescent patients with type 1 diabetes [4].
Why then is setting goals not commonplace? In the UK, the quality metrics of diabetes care are fixed to hitherto unitary standards (HbA1c of 58 mmol/mol or lower), and so striving to meet glycemic goals that appear less stringent can undermine reported performance in diabetes care. More importantly, the UK consensus on this process is lacking. This is in contrast to North America, where explicit guidance has been developed and the standard of care reporting has been adapted to reflect this [6].
Writing SMART goals with patients
The mnemonic acronym SMART stands for specific, measurable, achievable, realistic, and time-bound. Its original purpose was to serve as an aide-memoire for agreeing objectives for personal development and performance management [7]. However, this approach can be adapted for clinic consultations (Table 1).
Table 1. Suggested approach for the formation of SMART goals for type 1 diabetes care
Dimension | Consideration(s) |
Specific |
|
Measurable |
|
Achievable |
|
Realistic | The balance of achieving tighter glycemic control without increased hypoglycemia is more likely in patients with:
|
Time-bound |
|
Agree a HbA1c target based on age, duration of diabetes, risk of severe hypoglycemia, presence or absence of cardiovascular disease, and life expectancy
Patients need to be informed of the health benefits and risks of pursuing a set HbA1c target. The effectiveness of intensive insulin therapy in reducing diabetic complications has been well described. In the Diabetes Control and Complications Trial (DCCT), a HbA1c of ≤53 mmol/mol represented nadir at 6 months in the control group [8]. Whilst all guidelines encourage achievement of this target for most patients, the evidence of a benefit for patients with manifested complications, established vascular disease, and of older age is lacking [9]. Given the increased morbidity of severe hypoglycemia in these groups, a less stringent HbA1c target of 64–69 mmol/mol is suggested by North American guidelines [10]. In particular, Diabetes Canada has developed a calculator for individualizing patients’ HbA1c targets [11].
Agree a complementary glucose target range throughout the day
Importantly, a glucose range that will attain target HbA1c is required to guide daily insulin adjustments. In the DCCT, the primary focus of intensive control was to lower pre-meal glucose to 3.9–6.7 mmol/L, including at bedtime [8]. Of note is the acceptance of a weekly glucose reading of ≥3.6 mmol/L at 3 am, which is below the floor of hypoglycemia requiring treatment in the UK. Instead, data from the A1c-derived Average Glucose (ADAG) study [12], which reported a correlation between blood glucose and HbA1c, is more practicable in this regard (Table 2). It can be further adapted into ranges and time blocks which take into account a patient’s glycemic variability and daily living.
Table 2. Average blood glucose (95% confidence interval) in mmol/L for specified HbA1c in patients with type 1 diabetes, adapted from [12]
HbA1c (mmol/mol) | 37–47 | 47–53 | 53–58 | 58–64 | 64–69 |
Estimated average glucose | 6.2–7.7 | 7.8–8.5 | 8.6–9.3 | 9.4–10.1 | 10.2–10.9 |
Mean fasting | 6.8 (6.3–7.3) | 8.0 (7.4–8.6) | 8.6 (7.9–9.3) | 9.4 (8.8–10.1) | 9.9 (8.9–10.8) |
Mean pre-meal | 6.6 (6.4–6.9) | 7.8 (7.4–8.2) | 8.7 (8.3–9.1) | 8.8 (8.4–9.2) | 9.7 (9.0–10.4) |
Mean post-meal | 7.7 (7.4–8.1) | 8.9 (8.6–9.3) | 9.7 (9.3–10.2) | 10.6 (10.0–11.1) | 10.9 (10.4–11.4) |
Mean bedtime | 7.8 (7.3–8.2) | 8.6 (8.0–9.1) | 10.0 (9.1–10.8) | 9.9 (9.2–10.7) | 11.9 (105.–13.3) |
*Includes fasting values |
Agree a minimum acceptable frequency of hypoglycemia
The third element to specify refers to patients’ tolerance for hypoglycemia. The minimum acceptable frequency is influenced by patients’ hypoglycemic awareness, self-management capabilities, and history of severe hypoglycemia. This discussion leads to the setting of the floor in target glucose range, and empowers patients to seek help earlier if their hypoglycemic burden intensifies.
Challenging proposed goals: Can you and will you do it?
The next four letters (“MART”) provide a framework for appraising the preliminary goals. The bar may be too relaxed, where tight control is achievable and realistic. This is seen clearly in a cohort of patients with significant residual beta-cell reserve in the DCCT, where intensive management was more readily achieved with a lower risk of severe hypoglycemia [13]. Identified using C-peptide testing in a mixed meal tolerance test, these patients can be recognized clinically by a history of honeymoon remission, an insulin requirement of <0.5 units/kg, and attenuated postprandial excursions in their glycemic profile. Conversely, tight control was associated with the preservation of beta-cell function. The corollary is that patients at diagnosis should be actively encouraged to pursue near-normoglycemia.
Patients with a fear of hypoglycemia commonly overcompensate. Increasing the frequency of glucose testing to understand the timing and triggers of hypoglycemic episodes would be the initial step; here, flash glucose monitoring has practical advantages [14]. Patients should also be introduced to advances in treatment, such as structured education [15], new basal and prandial insulins [16], and insulin pump treatment [17], all of which can lower HbA1c without the trade-off of increased hypoglycemia.
The attainability of tight glycemic goals should be reviewed against patients’ self-management capabilities and comorbidities. Often patients underestimate the discipline required for tight control, which requires regular and strategic glucose monitoring and interim review of insulin dosing [18]. And so, whilst four-times daily pre-meal testing guides titration of basal insulin and correction bolus, this practice ignores exposure to postprandial hyperglycemia, a secondary interventional target in the DCCT.
Concomitant anxiety, depression, and eating disorders, which can present in up to 10% of patients with type 1 diabetes, are barriers to patient engagement and require psychological intervention [19]. Lastly, the rate of glucose-lowering in patients with chronically poor control needs to be gradual so as to avoid aggravation of proliferative retinopathy and neuropathy [20].
Conclusion
Sadly, clinic consultations where there is no disclosure of self-monitored blood glucose are widespread. In turn, minor insulin dose adjustments based on HbA1c by clinicians are rarely effective. A structured goal-setting approach for glycemic control forms the platform for care planning and has potential to improve patient engagement.