Clinical measurements of periodontal attachment loss, while not always representing the histological changes, are used to establish the diagnosis of adult periodontitis. Such measurements are difficult to perform accurately and are labor intensive. To counter these problems, index teeth and index sites have been employed in an attempt to estimate the severity of the periodontal condition without the need to resort to elaborate attachment measurements. Unfortunately, such indices usually tend to underestimate prevalence while often overestimating severity. The purpose of the present study was to examine the correlation of alternative clinical, radiographic and microbiological parameters, with periodontal disease using the diagnostic criteria of established periodontitis. 508 adults included in this study received thorough periodontal examination which included probing pocket depth, clinical attachment level, plaque, gingival and calculus scores, together with radiographic analysis and assays of subgingival periopathogenic microorganisms. Radiographic alveolar bone loss and probing pocket depth had the highest correlation with clinical attachment loss (phi = 0.72 and phi = 0.75 respectively). Plaque scores (phi = 0.17), like gingival scores (phi = 0.06) and calculus scores (phi = 0.42) had poor correlation with established periodontitis. Periodontopathogenic species demonstrated high specificity and negative predictive values; but low sensitivity and positive predictive values make for an overall low correlation of these species with established periodontitis. However, when used in a logistic regression model, the presence of P. gingivalis (odds ratio = 6.25) has shown to contribute significantly to the estimate of probability for established periodontitis. The use of these various alternative parameters for screening of periodontal disease is discussed in light of their sensitivity, specificity and predictive value.