Background: Partial or complete edentulism can have a huge impact on an individual’s life due to functional and/ or esthetic consequences. From ancient records dating to 2500 BC and up to the eighteenth century, missing teeth were replaced with animal teeth, human teeth, or “artificial” teeth made from melted wax, gold, or metal alloys. Although considered “state of the art” at the time, these approaches were fraught with problems. A dramatic change occurred with the introduction of root form osseointegrated dental implants into the clinical armamentarium. Today, such implants are used routinely to replace missing teeth in a highly successful and predictable manner. However, in a small percentage of patients, clinical failures do occur. Research has shown that the etiology of failures is multifactorial and can be due to patient-related and/or exogenous factors. Aim: The aim of this retrospective study was to evaluate the incidence of failure and the associated co-contributing factors of implants placed and restored in an academic setting. Materials and Methods: The electronic dental records of all patients who had implants placed and restored in the pre- and post-doctoral clinics or the faculty practice of the University of Pennsylvania School of Dental Medicine between June 1st, 2016 to August 31st, 2019 and explanted before February 29th, 2020, were identified using the implant placement procedure code D6010. From this group, patients that experienced an implant failure were then identified using the implant removal procedure code, D6100. Results: In total, 1609 patients received 3,180 dental implants during the timeframe of the study, 883 patients treated at PDM (total of 2162 implants) and 726 (total of 1018 implants) at PDFP. The overall percentage of patients who experienced failures was 4.9% (X/1609), 6.0% of patients treated at PDM (X/883) and 3% (X/726) treated at PDFP. At the implant level, the school-wide failure rate was 3.49%, 3.7% for implants placed at PDM and 3.0% for PDFP. Implant-related variables that were significantly over-represented in the failure group (combined data for both PDM and PDFP) showed that some variables were significantly overrepresented in the failure group such as Active Nobel BioCare (21.1%, P-value<0.001) and Straumann tissue level (27.8, P-value<0.001). In term of surfaces, both SLA and osseotite surfaces were overrepresented with P-value=0.001 and <0.001 respectively. 8-10 mm length (6.6%, P-value=0.021) and 4-4.7 mm width (6.7%, P-value=0.040), No Implant Restoration (14.6%, P-value<0.001), No Ridge Preservation (11.8%, P-value=0.006), Perio-Prosth Resident clinicians (29.2%, P-value<0.001), and Maxillary Anterior site (13.4%, P-value<0.001) were significantly overrepresented in the failure as well. Based on the multivariate conditional logistic regression results in Table 3C, estimated odds of implant failure in people with multiple implants was about 4 times than people with single implant (Adj OR=4.04, P-value=0.004); estimated odds of implant failure in people who received bone graft “before and during” the implant was about 5.6 times than people with no bone graft (Adj OR=5.57, P-value=0.012); estimated odds of implant failure in people with no restored implant was about 33 times than people with restored implant (Adj OR=0.03, P-value<0.001); Other variables did not show any significant association with implant failure (with P-values>0.05). Conclusion: Implant therapy is viable option for replacing missing teeth. This retrospective study showed that incidence of implant failure in academic setting was not significant. Multiple implants, bone grafting during implant placement in previously grafted site, and non-restored implant showed significant association with implant failure.