This study investigated the association of patient and provider characteristics on dental implant performance and survival probabilities. The outcomes of interest in this study were implant failures and peri-implant tissue health status. Univariate and bivariate statistics were used to describe the study population and to assess association of outcomes with patient and facility characteristics. Data were gathered from the period 1985 through January 1, 1990, from the Department of Veterans Affairs Dental Implant Registry. Logistic regression models were used to relate the outcomes with independent variables. Results showed that implant survival is associated with (1) the medical status of the patient, (2) the surface coating material of the implant, and (3) implant surgical and healing complications. Peri-implant soft tissue health was associated with (1) patient's use of tobacco, (2) surface coating of implants, and (3) implant provider's experience. Palabras clave: dental implants, epidemiology, longitudinal study, regression analysis