The Quality and Outcomes Framework (QOF) gives GP practices incentives for providing specified activities for chronic conditions, diabetes included. Practices receive rewards either for processes, such as monitoring albuminuria, or for outcomes, such as reaching target levels of HbA1c. The principal assumption underlying QOF is that patients will benefit because the activities have proven effectiveness.
This principle, however, is not without challenges. The first challenge relates to the rigour with which evidence is applied. For example, the incentive to screen for depression in individuals with diabetes (later withdrawn) was included although it did not meet the criteria for screening set down by the UK National Screening Committee (Gilbody et al, 2006; Luchins, 2010). The second challenge is the relevance of trial evidence to general practice populations. For example, older people figure largely on GP diabetes registers, but most trials exclude them. The mean age of people enrolled in trials exploring the benefits of glucose-lowering techniques ranges from 50 to 66 years (Hemmingsen et al, 2011). While these challenges will be familiar to most readers, the third may well not be: this is the matter of patient involvement in the decision-making process about their treatment.
Patients and health professionals diverge in the values and preferences they have for treatment: patients usually have higher thresholds for interventions (Steel, 2000; Montgomery and Fahey, 2001; Man-Son-Hing et al, 2005). Patients are more likely to decline an intervention if they are given the full information about the extent of its benefits and harms than if they are advised to have the intervention with less than the full information. A patient’s rejection of an intervention may conflict with health professionals’ attempts to adhere to a protocol or to apply what they believe to be in the patient’s best interests (Sanders et al, 2008). Yet such a conflict would not arise if we did two things.
First, remember the ethical principle of patient autonomy. This means we must involve patients in decisions about their long-term care. Second, we must see the benefits and harms from a lay perspective. Intensive glucose lowering may reduce microvascular complications but at a cost of more frequent hypoglycaemic episodes (Hemmingsen et al, 2011). Those patients who declined such treatment made decisions that were appropriate for them. Those doctors who recognised their right to do so and excepted them from QOF were practising patient-centred care. The language chosen by QOF for patients who decline treatment is “dissent”. This term brands them as rebels against medical orthodoxy rather than individuals exercising their autonomy.
Properly involving patients in decisions requires good, honest communication. Yet a search online of “shared decisions” and “diabetes” turns up endless titles on decision aids, paper or digital, rather than communication aids. However, general research into communicating risk (Australian Government National Health and Medical Research Council, 2004) gives useful guidance, such as the following: use language that patients understand, such as frequencies (“one in 10” rather than 10%); give balanced information (for example, give absolute risks rather than relative risks when the latter exaggerates benefit); and stop and check that the patient has understood.
The management of long-term conditions, including diabetes as well as others, can be neither “doctor knows best” nor “protocol states best”. Having people with diabetes and health professionals making decisions together will require that some practitioners learn new skills in communicating risk and enabling patient involvement.