Dr Mark Charny answers some frequently asked questions on clinical audit in general practice



How small can an audit be?

Q We don't really have time for the audits we are asked to do. They always seem to ask for so much information on so many people. How small can an audit be and still be useful?

A The answer depends very much on what type of audit is being carried out, and that in turn depends on the subject. Some topics require the inclusion of large numbers of patients, for example, a review of all patients with negative test results following the discovery of false negative errors in results from a radiology department. Some require only one patient, for example in a review of the care given in the case of a maternal death. Other audits may need to include all relevant patients.

In general, it is best to err on the side of collecting fewer data on fewer people. Fewer data means that the audit takes less time, costs less, is generally less prone to error because you can concentrate on data quality rather than quantity, and is easier to analyse or review.You can always re-audit a larger number of patients, or collect more information about each patient, if the initial – simpler – audit reveals cause for concern.

For example, if you wanted to review the care of patients with chronic heart failure, you could probably get a good picture of the situation in a practice by asking a few key questions on the last (or next) 10 patients:

  • Are these patients being seen regularly?
  • Have appropriate investigations been carried out?
  • How well are the patients' symptoms controlled?
  • Have all appropriate therapeutic options been considered and, where appropriate, tried?

A limited audit such as this will indicate with sufficient accuracy whether there are any causes for concern. If there are, those issues can be the subject of a more in-depth audit in a larger number of patients. If there is no cause for concern, a larger audit is likely to be unproductive.

The general rule is not to try to audit every aspect of an issue, to use an initial audit as a 'scan', and to be content with results which may not be precise but are good enough to be a guide to what further action is appropriate, if any. This is one of the ways in which audit is quite different from research.

What is the best software to use for audit?

Q What software would you recommend for a clinical audit in general practice?

A There is a wide range of specialist software available, and some general packages can also be used to carry out audits. It is difficult to make a recommendation in the abstract. The answer depends on what software you already have in the practice, the form in which the data you need is already available, your chosen approach, and the experience of those participating in the audit.

In general, I favour pencil and paper. If an audit is small, involving limited data on small numbers of patients, frequencies and simple cross-tabulations can be generated manually. This saves set-up time and costs, and makes the information more accessible.

What responsibilities do NICE and CHI have for audit?

Q How do NICE and the Commission for Health Improvement (CHI) fit together as far as clinical audit is concerned?

A In March this year, NICE and CHI jointly launched a book called Principles for Best Practice in Clinical Audit.1 The book aims to support staff leading clinical audit and clinical governance projects in the NHS.

The book's publication marks a renewed interest in audit nationally, and follows the Government's strong support for clinical audit in its response to the Bristol Royal Infirmary Inquiry (Kennedy) report.2 As part of its response the Government agreed with Kennedy's recommendation that 'clinical audit should be compulsory for all healthcare professionals providing clinical care'.

At present, NICE is the only national organisation explicitly charged with promoting clinical audit. However, CHI and the Modernisation Agency also have roles in helping to sustain quality improvements in NHS organisations.

CHI provides feedback on the implementation of clinical management strategies within these organisations in which a wide range of performance indicators have been considered. The Modernisation Agency facilitates the development of an environment within the NHS that is supportive of clinical audit.

The NHS Reform and Health Care Professions Bill currently before Parliament includes proposals for an independent Office for Information on Health Care Performance within CHI.

The Office will collect, analyse and publish reports on clinical and other NHS data. It will also develop a clinical audit programme incorporating audits that currently fall within the NICE work programme. The handover from NICE to CHI is under discussion – details are not yet known.

  1. National Institute for Clinical Excellence, Commission for Health Improvement, Royal College of Nursing, University of Leicester. Principles for Best Practice in Clinical Audit. Oxford: Radcliffe Medical Press Ltd, 2002.
  2. Bristol Royal Infirmary Inquiry. The Report of the Public Inquiry into Children's Heart Surgery at the Bristol Royal Infirmary 1984–1995. Final Report. 2001. Summary at www.bristol-inquiry.org.uk

Guidelines in Practice, June 2002, Volume 5(6)
© 2002 MGP Ltd
further information | subscribe