Sir, I was saddened to read in J. Ahearne's letter (BDJ 2005; 199: 315) that he believes that a 'clear definition of what constitutes evidence based dentistry' is needed.

I believe that a definition already exists.1 It is 'the conscientious, explicit and judicious use of current best evidence in making decisions about individual patients ... integrating individual clinical expertise with the best available external clinical evidence from systematic research'. The traditional approach involves a five-step model.2 There is, of course, practical difficulty in applying the model, particularly since patients and presentations are rarely identical. Nevertheless, I will try briefly to work through the process, given Dr Ahearne's clinical scenario of a pregnant patient showing signs of periodontitis attending for a dental examination.

1. Asking an answerable clinical question In this scenario, the question might be: 'can active periodontal disease during pregnancy cause pre-term delivery or an underweight baby?'

2. Searching for the evidence This requires good searching skills, tailored to the focus of the question developed. It also requires interrogation of appropriate sources of evidence, eg Medline database, Cochrane library. A systematic review or a 'best-evidence synthesis',3 might be the optimal type of publication to look for. If no reasonably current systematic review is available, you may have to decide to look at the results of some observational studies, (such as the cohort study of Moore et al.4 or the case-control study of Radnai et al.5). Randomised controlled trials would not be possible, for ethical reasons. Your choice of which papers to look at would probably be dictated by the titles, abstracts, easy availability of the full-text article, and time you had available.

3. Critically appraising the evidence for its validity and relevance This requires a systematic approach. Use of a relevant 'checklist' can assist greatly. Bias can creep in to all aspects of evidence, including that existing in the reader's subconscious! In the two papers by Moore et al.4 and Radnai et al.,5 the critical appraisal process led in one forum6 to the conclusion that the results of the latter may be misleading, whereas the former appeared to be a well-constructed study of 'higher' methodological weight, appropriate to the question. It would be worth looking at other evidence if you had time, since there may be some better quality evidence which supports the 'flawed' study, and it is hard to ignore a lot of evidence favouring one particular view, even if it is not of optimum quality. (No smoke without fire!)

4. Making a decision, by integrating the evidence with your clinical expertise and the patient's values You cannot assume that a pregnant patient wishes the best outcome for her baby. I once had a pregnant lady tell me that she intended to continue to smoke during her pregnancy because she did not want a large baby! Based on the process at 1-3 above, it might be appropriate to advise that there has been some work which suggests that periodontitis during pregnancy can have an adverse effect, but it is not clear whether this is true, because studies have shown conflicting results. Explaining things in this way would form part of the information and consent process for the patient, who should be told about her periodontal condition anyway, and offered relevant treatment.

5. Evaluating your performance Reflecting on the process undergone at steps 1-4 is one method. Try describing and analysing what happened during the process, your thoughts and feelings, what was good and bad about the experience, what else you could have done and what you will now do, either to improve the way you handle a similar situation in future, or as a result of your experience.

It's not easy, but it is worthwhile and you will be able to demonstrate a considered process for your practice. Good luck.