International Journal for Quality in Health Care 14:173-174 (2002)
© 2002 International Society for Quality in Health Care
Editorial |
Changing clinical practice—and perceptions
We know that clinical guidelines are not self-executing. But the Leeds University Maternity Audit Project [1] has thrown new light on the complexity of change—light that deserves to shine not just on maternity or on the UK; the messages could apply equally in many specialities and in many settings.
The project identified four accessible markers of compliance with available evidence-based guidelines and applied them to existing records in 20 maternity units in 1988 (before the evidence was widely published) and again in 1996 to show how practice had changed. Staff interviews collected further background on attitudes and access to guidelines and information.
The good news was that, overall, there was a massive and appropriate shift in clinical practice; the bad news was that there were notable pockets of resistance. This resistance was not associated with aversion to the published guidelines (most respondents thought they were a ‘good thing’), nor did it prevent maternity units from developing their own consistent policies (many had adopted them locally), nor was non-compliance with one guideline correlated with non-compliance with others in the same unit. The composite scores of individual units were not provided, nor were all other non-correlations given. The improvements were ‘despite the fact that few units have access to the Cochrane database, have prepared or disseminated guidelines, or taken any action to implement recommendations’; there was no reference to internal audit on these issues. The authors conclude that systematic uptake of research findings was ‘rudimentary’ in 1996 but, if the new statutory duty of managers in the National Health Service (NHS) to introduce ‘clinical governance’ is effective, it will lead to closer compliance with guidelines in future.
This prediction assumes that we can all learn from past experience (as well as research evidence), and that clinical governance has an active ingredient that was missing from previous NHS initiatives such as quality assurance (1980s), medical audit, and clinical effectiveness (1990s). Since April 1991, all hospital doctors in the NHS have been required to participate in systematic medical audit [2,3] (which became multidisciplinary in 1995). Hospitals set up audit committees, appointed audit assistants and organized audit meetings. Some professional associations and Royal colleges provided training and guidelines on clinical practice and audit. The British Medical Journal ran a regular section on how to do it (which grew into the journal Quality and Safety in Health Care); doctors were given the opportunity, the incentives, and the instructions to audit systematically and to improve their own performance against valid published criteria.
In retrospect, at least three essential elements were missing in the UK at that time. Firstly, doctors did not know how to compare patterns of practice against standards: the use of evidence, guidelines, audit, and measurement were not taught in medical school, in speciality training or in the working environment. Secondly, the power of systematic reviews and meta-analysis had yet to illuminate many contentious areas of clinical practice. Thirdly, the extent and mechanisms of doctors’ accountability to the public and to managers was unclear both ethically and legally. The recent public inquiry into paediatric cardiac surgery in Bristol [4] documented many of the failings of medical audit and management in the early 1990s, and pointed out that these were not unique to Bristol, or to cardiac surgery or even to that era; most of its messages could be usefully considered in any country, any town, any speciality.
One major change since 1996, attributable to Bristol, is that health service managers, regardless of their professional background, are now held statutorily accountable not only for finance but also for clinical performance [5]. Another is the development of guideline methodology and dissemination (especially electronically) in the NHS and worldwide. Maybe these, combined with the passage of time, will overcome the pockets of resistance.
In an ideal world, the findings from Leeds would sound clarion alarms and pose questions in every maternity unit: do we have local guidelines on these four topics? Are they consistent with current evidence? When did we last audit our performance against them? How would we score now, or in 1996, or in 1985? And for those who are not in the regular business of babies, what are the equivalent markers of our compliance with best practice? Routine examination of diabetic feet, hand-cleansing between patients, timely thrombolytics in myocardial infarction, prophylactic anticoagulation in surgery? The maternity study is a good example of a simple criterion-based audit, using data culled by non-clinicians from existing records to quantify current practice in common conditions against evidence-based standards; and the next step, delegated to whoever reads this, is to define and implement an action plan, and repeat the audit to demonstrate the improvement...
A further message for quality improvement appeared not in this Journal, but in the national Sunday broadsheet newspaper that published a news feature about the same paper on the Leeds study in February [6]. ‘Birth ward errors kill 200 babies: overstretched doctors and midwives are providing substandard maternity care, admits official report’. Those headlines were probably chosen not by the political journalist who researched the article, but by an editor with an antipathy to public funding of critical research and an eye to fuelling the chronic debate on resourcing the NHS; neither the Leeds paper nor the newspaper report made any suggestion that failure to implement guidelines was associated with lack of funding, or that staff were ‘overstretched’.
Perhaps one way we can build the public faith in health care and our own faith in quality improvement is to celebrate achievements, rather than bemoan the residual bad apples, tail-enders and outliers. In 1985 the median value among the 20 maternity units for giving steroids to pre-term babies was zero but by 1996 it had risen to 82%, reducing mortality for those treated from 18 to 7.2%; if all pre-term babies had been treated, a further 200 would have been saved. By my calculations, the Observer could have said, ‘Massive shift in practice saves 850 babies a year: improvements following introduction of evidence-based guidelines and clinical audit are set to continue’.
Charles D. Shaw
UK
References
- Wilson B, Thornton JG,. Hewison J et al. The Leeds University Maternity Audit Project. Int J Qual Health Care 2002; 14: 175–181.
- Department of Health. Medical Audit in the Hospital and Community Health Services. Health Care (91)2. London: Department of Health, 1991.
- Health Circular Resulting From the Government White Paper ‘Working for Patients’. London: HMSO, 1989.
- The Bristol Royal Infirmary Inquiry (proceedings, submissions and report), UK: http://www.bristol-inquiry.org.uk. Accessed 15 March 2002.
- Health Bill, 1999, London, UK: http://www.hmso.gov.uk/legislation.
- Ahmed K. Birth ward errors kill 200 babies. The Observer 2002; 24 February, p. 7.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  Minerva BMJ, August 3, 2002; 325(7358): 288 - 288. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||