Dr Mark Charny answers questions on the importance of audit and on audit standards and targets


Q What is the relationship between clinical audit and standards?

A Clinical audit monitors interventions and/or care against agreed standards. Standards may be generated in a number of ways, for example from Royal College guidelines, best practice recommendations, results published in peer-reviewed journals, the top quartile (or decile or other range measure) of performance, the average, or the participantsÍ consensus.

Clinical audit is not a way of generating standards. This would be circular logic: we observe what we do, we then set the standard as what we do, therefore we meet the standard. However, standards may be adjusted in the light of audit findings.

The purpose of audit is to identify any variations between the findings and the standards on which the audit is based, with a view to examining the reasons for the variations and putting them right. Current practice may be so far removed from ideal or desired practice that interim standards have to be put in place to bridge the gap between the present and the future. In other words, it may be necessary to move towards the standards in steps, and audit each step to ensure that the process is on track.

For example, meeting the standard of a maximum wait to see a GP of 48 hours in a practice in which it is currently 2 weeks may require a plan of action that reduces the wait by 12 days over 24 months or 1 day every 2 months. To try to move in one jump from 2 weeks to 2 days may not be possible.

In circumstances such as these, it is important not to use the practical difficulties of achieving dramatic change as an excuse for continuing to provide substandard care. However, while challenging standards are important to improve care, standards that are too challenging are simply demotivating and may hinder any improvement.

Q Why is clinical audit important?


A Clinical audit is important in a number of interrelated ways. First, and most importantly, it helps to ensure that patients receive the best possible care that can be achieved in the circumstances, given the limitations of facilities, staff and resources. To do this, those giving care must set standards which can be justified to other health professionals. It is also essential that everyone involved in providing care is committed to achieving those standards and to reviewing honestly any situations in which the standards have not been met.

Second, if audits are carried out and effective action taken to remedy any problems identified, the public and politicians can be assured that standards are being achieved and maintained.

Third, audit supports health professionals in making sure that patients receive the best possible care. It also gives them a sense of confidence that this is the case. It is also helpful if - as is increasingly the case ¿ professional practice is challenged in the courts, through the performance management framework, or by professional bodies such as the GMC.

Finally, audit is useful in informing health service managers about the existence of problems and the reasons for them, so that the organisation can take steps to improve the situation. This is true whether the problem is one of individual training, facilities, procedures or any other aspect of the complex infrastructure which underpins modern healthcare. Most problems are systems problems rather than those of individuals, and require an organisational rather than a personal solution.

In general, audit allows problems to be spotted at an early stage and managed proactively rather than reactively. This is best for everyone.

Q Where do the percentages in targets come from?

A It is important that targets are expressed quantitatively, even though it may seem slightly artificial at times; if this is not done, no one knows whether or not the targets have been achieved.

Values for targets can, and should, be chosen on the basis of a structured approach. There are broadly two ways of arriving at sound figures:

  • Setting a target for an event that does not occur all the time. For example, that 75% of the female population aged 18-60 years in a practice should be screened for cervical cancer within a defined period.
  • Setting a target for an event in a more closely defined population, which should occur all the time (or, in the case of negative outcomes, should never occur). For example, that 100% of the female population in a practice aged 18-60 years who are, or have been, sexually active and have not refused a smear, should be screened for cervical cancer within a defined period.

By excluding from the audit population the patients who will confuse the issue, as in the second example, it is possible to arrive at a reasonable goal that something always/never happens. Data are collected to weed out the exceptions (e.g. those who are not sexually active, those who refuse to have a smear, those aged under 18 or over 60 years), and the results are then unambiguous. In the first example, the percentage quoted in the target is inevitably more subjective, because it allows for the uncertainties of a more mixed population and these may be difficult to capture with precision.

On balance, I think the process of defining exceptions and removing them from the population whose care is being audited brings more clarity and is easier to manage both at the stage of data collection and when the results are peer reviewed.

Another approach is to involve a wider population by concentrating on a process, for example that all women over the age of 18 years are invited every so many years for cervical cancer screening and that whenever screening has not been carried out, reasons for this are established and recorded.

Guidelines in Practice, November 2002, Volume 5(11)
© 2002 MGP Ltd
further information | subscribe