Clinical audit has been integrated into the work of clinicians and managers in the NHS over the past 25 years. Clinical audit is a component of clinical governance and quality improvement. There are two large programmes collecting data about the care of people with diabetes running across parts of the UK at present. How do they compare?
The National Diabetes Audit (NDA) looks at four key areas and answers four key questions based on the National Service Framework for Diabetes (The Information Centre for Health and Social Care, 2008):
- Is everyone with diabetes diagnosed and recorded on a practice diabetes register?
- For people with diabetes:
- What is the annual rate of specific complications?
- What proportion of people with diabetes receive key processes of diabetes care?
- What proportion of people with diabetes achieve treatment targets?
The NDA is considered to be the largest clinical audit in the world and is funded by the Government. It covers England and Wales and collects data from both primary care and secondary care settings. There is also a paediatric audit programme. Routinely collected process and outcome data is extracted from clinical IT systems to populate the audits each year. Reports from the NDA are available at www.ic.nhs.uk/diabetesauditreports.
The NDA is being “reinvigorated” (Hillson, 2009) and is now part of the National Clinical Audit Support Programme. A new specification has been agreed and tenders to deliver the programme have been invited. The new specification aims to address the problems identified with the programme in previous years:
- There will be a focus on making the audit more than just a data collection exercise.
- The data will be presented in ways that make it more accessible and useful to clinicians, managers, commissioners and people with diabetes.
- More attention will be paid to giving feedback to the clinicians on the front-line who collect and record the data.
- Communication will be improved and there will be more engagement with clinical teams.
The income of GPs and the money available to them for investment in service development has been linked to their achievement since the inception of the new GMS Contract (Department of Health, 2004) using the Quality and Outcomes Framework (QOF). Achievement indicators have been set for a range of chronic conditions. Data is routinely extracted from GP computer systems. In due course it is published and patients can discover how their practice team’s achievement compares with that of others.
It is interesting to reflect that it is the attainment of targets by their patients with chronic conditions that earns GPs the majority of the points on which a portion of their pay depends.
The introduction of pay for performance in the UK has led to improvements in the achievement of important aspects of chronic condition management (Campbell et al, 2007). However, concerns have been expressed that not all people with diabetes have benefited (Millett et al, 2009) and that some indicators are easy to achieve without maximising benefits for patients: 50% of people with diabetes achieving an HbA1c level <7.5%, for example (Koshy and Millett, 2008), although this indicator changed in April to 50% of people with diabetes achieving an HbA1c level of <7.0%. There is evidence that a small number of practices are playing games with prevalence data and exception reporting to improve income (Gravelle et al, 2007; 2008).
How does the NDA differ from the QOF programme?
There are two important differences between the programmes.
First, the NDA collects data about everyone with diabetes, whereas the QOF process does not collect data about children or teenagers. However, while the NDA data is only reported from practices that have consented, QOF reports on almost every practice in the UK.
Second, GPs can exclude patients from the QOF process, for example if they decline procedures or medication, cannot tolerate additional medication or have a terminal illness. An individual who is excluded will not be counted against an indicator if their values are not within the indicator range. However, if their values are within the indicator range they are included.
Therefore, the practice summary achievement data cannot provide audit data about every person on the practice diabetes register. It is worth noting that if data about an individual has been changed by “game playing” for QOF purposes, the NDA data collected about that patient may also be affected. One major weakness of both of these audits is that the data cannot be correlated because the data sets are not similar enough.
An audit purist might argue that neither programme is a true audit: data collection on its own does not constitute audit. When clinical audit groups were active in general practice between 1990 and 2003, practice teams determined their criteria and standards and worked through audit cycles checking whether changes in systems had resulted in improved results for the patients. There was a concern to reflect on every person with diabetes and improve his or her clinical care.
Today, there are concerns that practices do not have the time or energy to reflect on their achievement and the achievements of their patients. Provided that the QOF indicators are attained and that the NDA audit data do not identify the practice as an outlier, then all is well.
A new approach is being piloted in Wales: the 1000 Lives Campaign (2009) will address this problem from the perspective of the person with diabetes. General practice teams will reflect on any individual who develops a complication of their condition or whose values deteriorate significantly using the “global trigger tool” (GTT). The GTT is being piloted and refined to ensure that a practice can use it both to identify individuals whose clinical care may need reviewing and to learn how to improve management and clinical systems to raise the standards of care that they offer. It will be interesting to reflect in 2030 about which GTT is more important for people with diabetes and the clinicians caring for them.