Sir, we read with interest the recent letter from Dr J. Ahearne regarding evidence-based dentistry, and the validity of clinical trials in general.1 As one of the goals of our departmental journal club is to teach critical appraisal skills, we decided to look at the two studies highlighted by Dr Ahearne,2,3 which had found opposite, and conflicting conclusions regarding the possible relationship between periodontal disease and adverse pregnancy outcome.

We used a 'debate-style' format4 for our meeting, and subsequently voted on the null hypothesis 'there is no relationship between periodontal disease in pregnancy and adverse pregnancy outcome'. Firstly, two senior house officers presented one paper each, discussing the methodology, results and their own critical appraisal of the two papers. The debate was chaired by the Trust's Professor of Surgery (PAB), with 22 dental and doubly qualified members of the department in attendance.

The paper by Moore et al2 published in the BDJ appeared to be a well-constructed large prospective cohort study, which took account of a number of identified confounding factors in the statistical analysis before concluding that there was no relationship between periodontal disease and pre-term birth or lower birth weight.

In contrast, the paper by Radnai et al3 published in the Journal of Clinical Periodontology, was a small-sample, retrospective case-control study, where the influence of cigarette smoking could not be controlled for by the statistical analysis due to the extent of difference in this factor between the cases and controls. It was pointed out during the debate that the choice of statistical test used to analyse the data may well have been incorrect in view of the non-normal distribution of the control group, so that the results as presented might well be misleading. In that respect, one of the authors of this letter (PA Brennan) who is on the editorial board of several clinical journals informed the debate that he would have rejected the manuscript for publication on that basis.

Perhaps unsurprisingly, the motion was carried unanimously in the vote.

We also considered an alternative motion: 'This house believes that there is no association between periodontal disease and adverse pregnancy outcome' (the distinction being that an association does not necessarily imply a causal relationship). Here the vote was more diverse, with several club members abstaining as 'don't knows'.

Dr Ahearne has spotted the geographic difference in patient population as a possible cause of differing study conclusions. We believe that by using a systematic process of critical appraisal, we have discovered that one of these studies is a far more reliable evidence source than the other. We advocate a similar type of regular journal club in the general practice setting, which not only contributes towards CPD, but also enables practitioners to develop critical appraisal skills. It is important not to fall into the trap of assuming that because a paper is published in a refereed journal, it must be sound. It is the responsibility of the reader to check the validity and applicability of research findings for themselves.