Purpose: This study was undertaken to assess the predictive usefulness of preoperative bone density, as measured by computed tomography (CT), and the intraoperative implant stability measures of insertion torque (IT), Periotest values (PTV), and resonance frequency analysis (ie, implant stability quotient [ISQ]) toward developing an algorithm for successful immediate loading, one-step exposure, or submergence of dental implants. Materials and Methods: Consecutively presenting patients requesting immediate loading in areas other than the anterior mandible were analyzed retrospectively. The implants were either immediately loaded, left exposed, or submerged on the basis of preoperative CT bone density and intraoperative primary stability measures. All implants surviving the traditional healing period were verified for osseointegration. Results: Eighteen patients were analyzed retrospectively, and they received 58 implants. Seven implants failed, for a survival rate of 88%. Primary stability measurements at insertion were correlated with one another and with preoperative CT bone density. Preoperative mean bone density for surviving implants was greatest for immediately loaded implants (983 ± 83), lower for exposed implants (803 ± 29), and lowest (480 ± 23) for submerged implants. Bone density was significantly different between submerged implants that failed and those that survived. Mean IT for successful implants was higher than for those that failed. Mean PTVs were lower (ie, better) for successful versus failed implants, although this difference was significant only for submerged implants. Conclusions: In this group of patients, objective measures of bone density by CT, IT, PTV, and ISQ correlated with each other and therefore may provide a useful algorithm for making clinical implant loading decisions. Of the technologies applied in this group of patients, PTV was the most reliable predictor at implant placement of failure to osseointegrate. Keywords: algorithm, bone density, immediate loading, partial edentulism, primary stability, retrospective analysis