Introduction: The primary objective of this study was to determine whether risedronate sodium (Actonel^?) therapy, in conjunction with conventional non-surgical periodontal treatment, reduces the rate of alveolar bone loss. Secondary aims were to compare the incidence of patient dropouts in the risedronate and placebo groups, and to document adverse events. Methods: This double-blind randomized placebo-controlled trial was conducted in 125 patients with moderate to severe periodontitis. At baseline, three, and nine months, standardized vertical bite-wing radiographs were taken and used to measure bone loss. Clinical periodontal examinations were taken at three-month intervals to assess whether or not the patient was experiencing rapid periodontal breakdown, in which case suitable treatment could be planned and delivered. Patients received scaling and root planing at baseline and periodontal maintenance at three-month intervals thereafter. At three months, subjects were randomly assigned to risedronate (35 mg/week by mouth) or placebo in blocks based on the severity of periodontitis (moderate, mean bone loss 2 -4 mm; or severe, mean bone loss >4 mm); smoking; and diabetes. Interval bone loss in millimeters was measured from the radiographs, and analysis of variance (ANOVA) was used to test for differences in bone loss between test and placebo groups. Results: Over the nine-month study duration, the test group exhibited a significantly greater increase in bone height (0.31 ±0.09 mmgain test, 0.08 ±0.09 mmgain placebo, F = 4.94, p < 0.04). There were significantly fewer dropouts in the risedronate treated group (4.8%) than in the placebo treated group (17.5%; p < 0.04, Fisher Exact Test). No serious adverse events or cases of osteonecrosis of the jaw (ONJ) were observed. Conclusions: Periodontitis subjects treated with risedronate had significantly more bone gain compared to placebo treated subjects. Significantly more subjects from the placebo group dropped out, suggesting the possibility of a patient-noticeable effect. There was no indication of ONJ.