It is argued that the periodontal diseases can no longer be regarded as universally prevalent conditions to which all members of the world's population are at equal risk if they fail to practise good oral hygiene. Rather, they should be regarded as a range of different diseases for each of which certain individuals, which together comprise certain minority groups, are at relatively high risk. The epidemiological evidence for the existence of high-risk groups is reviewed, from which it is concluded that world-wide the prevalence of severe destructive periodontitis is of the order of only 7-15% of the adult dentate population. A working classification of the different types of gingivitis and periodontitis is offered, as is a summary of the theoretically possible approaches to the detection of high-risk groups and individuals which are explored in detail in subsequent papers. Successful identification of such individuals will permit scientifically valid, rational and targetted prevention and treatment.

The evidence for systemic predisposition to periodontal diseases is reviewed in relation to cellular and humoral immunity, drug therapy, diet and nutrition and stress. It is concluded that, apart from defects of polymorphonuclear leukocytes (PMN) and Ehlers-Danlos Syndrome, little firm evidence exists for other diseases, though insulin-dependent diabetes and acquired immune deficiency syndrome (AIDS) may accelerate and/or potentiate the damage of existing disease. The precise rôle of drugs, diet and nutrition and stress remain to be elucidated, but recent advances in these areas offer the prospect of assessing risk using carefully controlled studies.

The fundamental concepts of measuring periodontal diseases and the interpretation of such information as an historical record of disease, rather than disease activity, emphasises the need for improved diagnostic and prognostic tests. Criteria for an indicator of disease activity were suggested and an index fulfilling these should allow sites to be categorised as "active", quiescent or healing, and enable one to predict the risk of future disease activity. The ability of current measurements and indices, routinely used during clinical assessments of periodontal diseases, to fulfill the suggested criteria was considered and rejected in all cases. It is concluded that clinical parameters are only capable of identifying disease retrospectively, indicating the need for longitudinal, rather than cross-sectional studies in the search for clinical and laboratory markers of disease activity and susceptibility.