Audit is widely used to improve the quality of care in both medicine a
nd dentistry but has not always been subject to systematic evaluation.
The aim of this study was to assess the effect of an audit exercise o
n the diagnosis of occlusal caries. Using combined visual and radiogra
phic methods, 11 dentists assessed 60 extracted molars for occlusal ca
ries on two occasions. An audit exercise, in the form of a meeting, wa
s carried out in the intervening period. Inter- and intra-examiner rep
roducibility were estimated for each of the two examinations. Validati
on was through a gold standard set up by sectioning the teeth. Wide di
agnostic disparity was seen at both examinations, all inter- and intra
-examiner kappa values were below 0.73. Intra-examiner kappa values in
creased significantly on the second occasion. Mean sensitivity and spe
cificity were 0.65 and 0.74, respectively, at the first examination an
d 0.65 and 0.72 at the second. No changes were seen that could be attr
ibuted directly to the audit exercise but practising the diagnostic te
st appeared to be beneficial to intra-examiner reproducibility of diag
nosis.