Abstract: Predicting the future condition of existing bridges is important for bridge management. Current methods for condition prediction of existing bridges have high requirements for completeness of bridge inspection data. In view of the current situation of unsystematic and incomplete storage of bridge inspection data in our country, this paper combines the concept of Time-in-Condition (TC) and proposes the condition deterioration model updating method for existing bridges using historic inspection data. By assuming TCs follow independent normal distributions, the likelihood function is given to Bayesian update the probability distributions of TC models. Combined with Markov chain model, the formulas for the calculation of condition transition probabilities are derived based on current condition, time in current condition and service time in future. The accuracy of the proposed method is verified through numerical examples, and the effects of data amount, data completeness and data accuracy are explored. Finally, the proposed method is applied to the condition prediction of the superstructures of reinforced concrete bridges of a city using the real inspection data of 185 bridges, the TC models are updated for different conditions, and the future deterioration risk is evaluated.