Current scoring systems for dental caries do not consider the dynamic nature of the disease. The aims of the present study were to describe a new set of clinical caries diagnostic criteria which differentiate between active and inactive caries lesions at both the cavitated and non-cavitated levels and to evaluate the reliability of this criteria system in a population with high caries experience. Ten diagnostic codes were defined: 0 = sound; 1 = active (intact); 2 = active (surface discontinuity); 3 = active (cavity); 4 = inactive (intact); 5 = inactive (surface discontinuity); 6 = inactive (cavity); 7 = filling; 8 = filling with active caries; 9 = filling with inactive caries. Distinction between active and inactive caries lesions was made on the basis of a combination of visual and tactile criteria. The inter- and intra-examiner reliability was assessed through repeated examinations of 50 children by 2 recorders over a period of 3 years. The percentage agreement of caries diagnoses varied between 94.2 and 96.2%. The kappa values ranged between 0.74 and 0.85 for intra-examiner examinations and between 0.78 and 0.80 for inter-examiner examinations; 81.6% of all misclassifications involved non-cavitated caries lesions. Disagreement between sound surfaces and non-cavitated active or non-cavitated inactive lesions (31.3 and 31.2%, respectively) was more common than disagreement between non-cavitated active and non-cavitated inactive lesions (10. 6%). The probability of reconfirming a sound, non-cavitated active or non-cavitated inactive caries lesion - given that the surface was diagnosed as either sound, non-cavitated active or non-cavitated inactive at the first examination - was 98.0, 68.7 and 72.5%, respectively. The results show that the use of a new set of clinical caries diagnostic criteria based on activity assessment can be performed with a high reliability, even when non-cavitated diagnoses are included in the criteria system.