The purpose of this study was to evaluate the use of a chair-side test involving a count of salivary mutans streptococci (the Strip mutans test) in the assessment of caries risk in a group of preschool children living in an area with a low caries prevalence. A group of 528 4-year-old children were randomly allocated to a study or a control group. In the study group, the baseline microbial data, together with clinical findings of past caries experience, were used for caries risk assessment and for planning subsequent preventive treatment. All children were examined at baseline and after 2 years. Caries experience was assessed according to WHO criteria. There was no difference in caries experience between the study group and the control group at baseline. Within the study group, caries increment was positively correlated (P or = 2 and/or > or = 1 dmfs) developed more new lesions than those considered as 'low risk' (mean dmfs 2.6 v 0.9; P