Abstract: | Selected teeth have been used to represent the entire dentition in many epidemiological and clinical investigations. The present study sought to assess the relationship between the six selected teeth described by Ramfjord and the entire dentition for the Plaque Index, Gingival Index, Calculus Index and Loss of Attachment. The computations were performed on measurements obtained in investigations of the natural history of periodontal disease in Norway and Sri Lanka. A consistently strong correlation was observed between the tooth subset and whole mouth for all indices in both populations. This relationship persisted even though single members or pairs of the subset of teeth were removed from the calculations. Some bias, however, was observed with all indices. Plaque and gingival indices obtained from the six teeth underestimated whole mouth scores in the low range (less than 1.0) and overestimated scores in the high range (greater than 2.0). Use of the tooth subset for calculus and loss of attachment consistently overestimated scores for the entire dentition. |