Identify risk factors associated with implant failure by cox proportional hazard model

In recent years, dental implants have received special attention in dentistry. Due to the remarkable success of predictable dental implants, there is growing interests in the scientific community from descriptions of implant success toward identify factors associated with implant failure. The purpose of this study was to identify the risk factors associated with implant failure in an objective and statistically valid manner.
To address the research purposes, we used a cohort study in which a total of 2368 implants were placed on 1011 patient in the Clinic of Tehran University of Medical Sciences. The predictor variables included demographic variables, the health status and variables are related to implants. The outcome variable of interest was implant failure. The overall implant survival was estimated using the Kaplan-Meier analysis. First, risk factors were identified using Chi-square test and Fisher's exact test and then the Cox proportional hazard regression models were used for more analysis.
The overall 39 month implant survival using the Kaplan-Meier function was 0.99±0.002. The implant location (jaws) (P=0.049) and type of teeth (P=0.011) were found to have significant relation with failure times in Cox proportional hazard model.
According to result of this study, both implant location and type of teeth had effection the failure times so that, the survival in mandible was higher than maxilla. The failure rate was higher in anterior teeth than that of others (molar, premolar, canine and primary).