Editorials
Health services administration Endocrinology General medicine
Volume 197 - Issue 4

Individualising type 2 diabetes management: new treatment options and models of care

Author:  N Wah Cheung

Med J Aust 2012; 197 (4): 196-197. || doi: 10.5694/mja12.11041
Published online: 20 August 2012
Better outcomes require tailored strategiesThe past 15 years has seen the advent of many new classes of antidiabetic agents. We now have biguanides, sulfonylureas, thiazolidinediones, dipeptidyl-peptidase IV inhibitors, glucagon-like peptide 1 receptor agonists, a variety of insulins and a-glucosidase inhibitors. This has resulted in a surfeit of choice, but has also increased the complexity of decision making as there is no simple algorithm for escalation of treatment ...
Download PDF (96 KB)
Share: Share on Blue Sky Share on Facebook Share on LinkedIn Share by email

Better outcomes require tailored strategies

The past 15 years has seen the advent of many new classes of antidiabetic agents. We now have biguanides, sulfonylureas, thiazolidinediones, dipeptidyl-peptidase IV inhibitors, glucagon-like peptide 1 receptor agonists, a variety of insulins and α-glucosidase inhibitors. This has resulted in a surfeit of choice, but has also increased the complexity of decision making as there is no simple algorithm for escalation of treatment that is optimal for all patients.

The latest American Diabetes Association and European Association for the Study of Diabetes recommendations for glycaemic management of type 2 diabetes recognises this complexity (see Appendix).1 Metformin is consistently recommended at all stages, but almost all combinations of the first six types of drugs listed above are given as options. Emphasis is now placed on tailoring treatment to the individual patient; clinicians should consider comorbidities, risk of hypoglycaemia, effects of treatment on weight, potential adverse effects, drug potency, dosing profile, glucose profile, pregnancy planning, Pharmaceutical Benefits Scheme restrictions (in Australia), cost, and patient acceptance.

The emphasis on individualisation also applies to glycated haemoglobin (HbA1c) targets. Three years ago, the Australian Diabetes Society embedded the concept of individualising HbA1c targets into its management guidelines.2 A major catalyst for this was the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study, which found that aiming for an HbA1c level of less than 6.0% (< 42 mmol/mol) in people with diabetes and vascular disease did not reduce macrovascular complications and was associated with an unexpected increase in mortality.3 Conversely, the United Kingdom Prospective Diabetes Study (UKPDS) demonstrated that tighter glycaemic control early in the course of the disease translated into better long-term outcomes despite failure to maintain this level of glycaemic control.4 There was also concern about aggressive treatment of people with significant comorbidities, or older people, in whom there is greater potential to do harm.2 Thus modern diabetes guidelines encourage the application of informed clinical judgement and personalisation in setting flexible HbA1c targets, rather than focusing on a uniform HbA1c level. Similarly, modelling of hyperlipidaemia management suggests that a tailored statin treatment strategy results in better outcomes than a single treat-to-target guideline.5

HbA1c is also used as a measure of quality of diabetes care. In the UK, the Quality and Outcomes Framework (QOF) uses HbA1c-based indicators to determine general practitioner pay for performance.6 Similar systems exist in the United States. These aim for generic (not individualised) HbA1c targets, but the QOF accepts that not all patients can achieve the ideal target by (i) using quality-weighted targets, so that incentive payments are triggered at multiple HbA1c levels (9.0% [75 mmol/mol], 8.0% [64 mmol/mol] and 7.5% [58 mmol/mol]) and (ii) using “exception reporting”, which allows practices to remove patients from their denominator calculations, so that they are not penalised for non-attending patients or patients for whom indicator-related management is inappropriate. These aspects of the QOF have generated concerns about the potential to “game” the system — for example, by focusing on patients with mildly elevated HbA1c levels, who can easily achieve targets, or by removing patients with complex conditions from the scheme.7

In Australia, HbA1c is also being evaluated as one of a suite of factors for determining incentive payments in the Diabetes Care Project (DCP) (previously known as the GP Coordinated Care for Diabetes Pilot).8 Other important elements of this project are increased funding for treatment of patients with complex comorbidities, training for practitioners, care coordination, and information technology support for care teams. Incentive payments are triggered for: improving the median HbA1c level in the practice population for whom baseline HbA1c levels are greater than 7.5% (> 58 mmol/mol); and maintaining the HbA1c level for those with a baseline level of 7.5% or less (≤ 58 mmol/mol). Considering the baseline HbA1c level, and rewarding even modest improvements, allows a degree of individualisation, rather than treatment to rigid targets; however, the incentive payments in the DCP still rely essentially on standardised targets. Also, as participation is voluntary, some patients may be excluded, and analysis of outcomes will need to take this into consideration.

It remains to be seen whether the new DCP model, in part designed to achieve standardisation of care and thus reduce health disparities in the population, will present barriers to appropriate individualisation of management. Increasing sophistication of computerised decision support systems and data extraction should be able to promote individualisation by introducing novel indicators such as achievement of personalised clinical targets (eg, an HbA1c level of less than 6% [< 42 mmol/mol] in a healthy young patient), clinical action measures (eg, escalate therapy if the HbA1c level is greater than 7% [> 53 mmol/mol]), care measures, patient preference and patient-reported data.9 However, while computers can provide guidance, they will never substitute for decisions based on informed clinical judgement. The latter, when done well, has been described as the “intuitive and iterative negotiation of the patient’s narrative of illness”10 — we await the development of a quality indicator for this!

Author

  • N Wah Cheung MB BS, FRACP, PhD1
  • 1. Department of Diabetes and Endocrinology, Westmead Hospital, Sydney, NSW.

Correspondence: wah.cheung@sydney.edu.au

Competing interests

I am the President of the Australian Diabetes Society and a member of the Australian Government’s Diabetes Advisory Group for the Diabetes Care Project.

References

Provenance: Commissioned; externally peer reviewed.

View this article on Wiley Online Library