Is diabetes all about target-driven care?

In my dictionary, ‘target’ is defined either as ‘a desired goal’ or ‘an object of criticism or attack’. This is very apt, for while health targets are perceived by the Government as a desired goal, their implementation can lead to criticism. 

Take, for example, the target whereby acute trusts are obliged to treat every patient attending accident and emergency (A&E) departments within 4 hours. This is undoubtedly desirable for the patient who presents to A&E, as well as for trusts who, in the past year, stood to gain several hundred thousand pounds if they met this target.

However, often no additional resources were made available to meet this target, and so reports abound of ways in which hospitals have complied with the letter, but not the spirit, of this target. Hence we hear of medical staff diverted from seeing more seriously ill patients not subject to the target, of areas within A&E being redesignated, and of patients being moved to other areas of the hospital even if they have not been fully assessed, just to meet the target (Gulland, 2003).

National Service Framework targets
Diabetes now has targets of its own. The first came with the publication of the National Service Framework for Diabetes: Delivery Strategy in 2003 (Department of Health [DoH], 2003a). These targets included the following.

• By 2006, a minimum of 80% of people with diabetes will be offered screening for the early detection (and treatment if needed) of diabetic retinopathy as part of a systematic programme that meets national standards.
• In primary care, practice-based registers will be updated so that patients with coronary heart disease (CHD) and diabetes continue to receive appropriate advice and treatment in line with National Service Framework (NSF) standards.

Funding has been made available to support the implementation of retinal screening, but otherwise the diabetes NSF did not come with any ring-fenced resources. Hence, other targets are less prescriptive. For example, the NSF states that a local approach to addressing the needs of the newly diagnosed could offer the following.

• Information and appropriate psychological support and the opportunity to participate in structured (usually group) education to people diagnosed with diabetes after April 2003.

The National Institute for Health and Clinical Excellence (NICE; formerly known as the National Institute for Clinical Excellence) subsequently issued guidance on structured education in 2003 (NICE, 2003), but with no date by which it had to be implemented and no resources to support its implementation. It is to the credit of diabetes specialist teams and primary care trusts (PCTs) across the UK that so many have made real progress in implementing structured education programmes by making better use of their existing resources.

This shows that a target which is perceived by healthcare professionals as a priority does not need the carrot and stick approach so favoured by the Government. The stick has now been applied, namely an implementation date set for January 2006 (DoH, 2005), and this may be a welcome prompt into action for those areas that saw structured education as a lesser priority.

General Medical Services contract targets
In 2004, the new General Medical Services (nGMS) contract was implemented (DoH, 2003b). This introduced targets for a number of diabetes-related processes and outcomes. This time, financial incentives are attached to the attainment of these targets, some of which are listed in Table 1.

These targets also have desired goals, namely that people with diabetes are protected from developing micro- or macrovascular complications of diabetes. They do this by encouraging practices to screen patients for complications and to treat to achieve ‘safe’ levels of HbA_1c, blood pressure and total cholesterol. These are indeed laudable aims, but attaining them is not always without risk.

Achieving an HbA_1c of 7.4 % in a person with longstanding diabetes is likely to require the use of multiple oral antidiabetic agents and/or insulin. It is well established that patients on insulin with an HbA_1c of 7 % are at significantly increased risk of hypoglycaemia (Diabetes Control and Complications Trial [DCCT] Research Group, 1993), to which older people (who comprise a significant proportion of the diabetes population, yet are rarely included in trials on whose evidence such recommendations are based) may be particularly vulnerable. 

A particular criticism of the current target is that a fall in HbA_1c from 9 % to 8 % is associated with a significant reduction in risk of complications (UK Prospective Diabetes Study [UKPDS] Group, 1998a), yet this achievement is not recognised by the nGMS contract. 

It may be argued that an HbA_1c of 8 % may be more appropriate in some patients, such as an older person on insulin. Similar considerations might apply to an overweight person with suboptimal glycaemic control, in whom the next step would be insulin therapy. Starting insulin in this group may not lead to a significant fall in HbA_1c, but is often associated with hypoglycaemia and also weight gain, which itself is associated with an increase in vascular risk (Nicholls et al, 2005).

Achieving the blood pressure target can also cause problems. Antihypertensive drugs are commonly associated with side effects, and achieving a blood pressure of 145/85 mmHg often requires the use of three or more agents (UKPDS Group, 1998b). Again, older people are at particular risk from such treatment, which may cause postural hypotension and increase the risk of falls. 

For men, antihypertensive treatment may be associated with an increased frequency of erectile dysfunction, which may not be a price they feel is worth paying for very tight blood pressure control. As with HbA_1c, a systolic blood pressure reduction from 180 mmHg to 160 mmHg is associated with a significant reduction in vascular risk (UKPDS, 1998b), and an individual patient may forgo the additional benefit of a further reduction to 140 mmHg if it means that he can maintain his erectile function.

In contrast, achieving a target lipid level is often much easier. The use of statins, which inhibit cholesterol formation, has led to a significant reduction in cholesterol levels in many people with diabetes. Such reductions can occur with little effort on the part of the patient, other than taking a tablet every day, and usually with no side effects. 

However, since the publication of the Heart Protection Study (HPS; Collins et al, 2003), there has been a move to higher prescribed doses of statin (e.g. 40 mg simvastatin, from 10 mg), and to broaden their use to a much wider patient group (for example, including younger people) regardless of the specific vascular risk (and hence benefit from statin use) in an individual person. 

Anecdotally, this appears to have led to a higher incidence of side effects (commonly muscle pains), and even though the HPS provides an evidence base for widespread use of a higher dose of simvastatin, clinical experience suggests that a more individualised approach is preferable. 

Responding to the targets
So, how should we as diabetes specialists respond to these targets? First, they should be welcomed as they do provide a framework for structuring diabetes care and designing care pathways to help achieve better outcomes for patients. Second, as with all targets, they can provide a lever to securing resources for diabetes care.

However, when it comes to implementing them in individual people with diabetes, we need to consider very carefully not just the targets prescribed from above, but also the desired goals of the individual, and of ourselves as professionals involved in the care of that individual. 

In this context it may be helpful to consider what our goal is in providing care for people with diabetes. We can then structure the services that we provide to help achieve these goals, and assess the effect of striving towards a given target in terms of whether it will help or hinder patients in achieving their own goal(s), and us in achieving ours.

I would describe my goal as being:

‘To enable a person with diabetes to remain healthy and symptom free, with minimal disruption to their lifestyle and psychological wellbeing.’

At a service level, this means that I have to provide services, or ‘tools’, that are consistent with meeting that goal, e.g. a service in which people with diabetes:

• learn and maintain appropriate self-management skills
• have access to appropriate treatments consistent with the goal
• have access to screening for vascular risk factors and long-term complications of diabetes
• have access to treatment for acute and chronic complications.

Many process targets are entirely consistent with my overall goal. In type 1 diabetes, screening for microalbuminuria is important, as it is a marker of early diabetes-related renal disease. In type 2 diabetes, however, it is a marker of vascular risk, and previously we did not screen this group for microalbuminuria but focused on reducing overall vascular risk, while also screening for overt proteinuria (a marker of more severe renal disease). 

The nGMS contract, however, stipulates that all diabetes patients should be screened for microalbuminuria in primary care. As a result, a number of patients with type 2 diabetes have now been identified as having quite marked levels of proteinuria. This suggests that our previous recommendation of proteinuria screening had not been implemented for these patients. Whatever the merits of microalbuminuria screening, applying a target and financial incentive to it has been effective in identifying individuals with previously undiagnosed proteinuria, which, as a result of the incentive, has been appropriately treated.

My approach to biomedical targets will be dictated by whether they will help achieve my stated goal, which, because of its emphasis on psychological wellbeing,
has to take into account the personal targets of the patient. Hence, prescribing an angiotensin-converting enzyme (ACE) inhibitor to a young person with type 1 diabetes and microalbuminuria is likely to help them remain healthy. And if that can be achieved without causing side effects then they will remain symptom free. 

However, if the ACE inhibitor causes dizziness or a cough, the potential long-term gains will be offset by real and immediate symptoms. These may be disruptive to the person’s lifestyle and psychological wellbeing, especially if that person is depressed about having diabetes and the restrictions it places on him/her. In such cases, taking extra medication is unlikely to be a high priority, and the individual is likely not to take it. The person’s goals will be his/her priority, and unless those goals are known, understood and respected by the professional, and at least taken into account by the professional’s goals, then the latter are unlikely to be achieved.

Consider the case of a man in his 30s with type 1 diabetes from childhood. His HbA_1c was 8.2 % and he had frequent hypoglycaemia on a twice-daily insulin regimen. He felt very restricted by his diabetes and frustrated about the effects it had on his lifestyle, including the fact that he had never felt able to eat out at a restaurant. His goal was to have the confidence to take his wife out for a meal, knowing how much insulin to take without running the risk of having a hypoglycaemic episode.

To help achieve his goal, he switched to multiple daily injections and attended an intensive education programme from which he gained the knowledge and confidence to match his insulin dose to his carbohydrate intake. Now, he rarely experiences hypoglycaemia and, although his HbA_1c has not gone down, the restrictions on his lifestyle have been reduced, and he has achieved his goal. As a result, he has embraced some of the ‘professional’ goals by starting treatment for his hypertension, and is keen to try to improve his HbA_1c. This is clearly a preferable outcome to an approach focused upon reducing his HbA_1c to meet an arbitrary target, which would have risked increasing his hypoglycaemia and made his goal appear even less attainable. 

Figure 1 shows how a care pathway incorporating management tools, patient goals and targets as ‘a means to the end’ can achieve the desired outcome, whereas one in which the target becomes the end in itself is likely to clash with a patient goal. 

Conclusion
Targets are here to stay in diabetes care. They can be a desired goal or an object of criticism and our challenge is to use them to help improve diabetes care and not dictate it. The targets that really matter, and which are likely to be successfully achieved, are those agreed by both patient and carer. Finally, targets can and should be ignored if, in an individual situation, they would do more harm than good.