[Definitive or exploratory periodontal experimentation: an overview of the literature]

Definitive and exploratory randomized controlled trials (RCTs) have different goals as well as different design and analysis characteristics. The goal of definitive studies is to provide unequivocal evidence of a treatment's tangible benefit to the patient; a pre-trial-specified hypothesis is tested by use of a pre-trial-specified method. The goal of exploratory studies is to elucidate biological treatment mechanisms, to identify promising treatments, and to generate hypotheses for definitive studies: multiple hypotheses are evaluated to extract as much information from the data as possible. The purposes of this study were: (1) to survey selected design and analysis characteristics of randomized controlled periodontal trials published between 1988 and 1992 (n = 86), and (2) to classify trials as exploratory or definitive studies. The periodontal RCTs surveyed were typical of exploratory studies whose primary goal was to elucidate biological treatment mechanisms. Trial reports indicated the testing of multiple hypotheses (> or = 6 hypothesis tests in 70 of the 86 trials) on a variety of biological markers (86 out of 86 trials). The sample size (< or = 30 subjects in 67 out of 86 trials), duration (< or = 6 months in 65 out of 86 trials), and design and analysis characteristics (e.g., an absence of masking in 57 out of 86 trials) were also typical of exploratory studies which strive to obtain quick answers (short duration) at a low cost (small sample size; accept bias for increased efficiency and a lower cost). No definitive trials were identified. Promising, biologically active, treatments identified in exploratory trials should be evaluated in definitive studies where the primary goal is the procurement of unequivocal evidence of a treatment's tangible benefit to the patient. The costs and benefits of conducting definitive periodontal RCTs to provide such evidence should be investigated.