We separately study the state of each individual during the infection process, revealing the role of the individual’s centrality properties in spreading the infection across the network. Although the continuous time SIR Markov model, based on the Markov chain stochastic process, describes the global change in the state probabilities of the network, it is limited to small networks due to the exponential divergence in the number of possible network states 3N with the growth of network size N. Instead, our approach aims to reduce the complexity of the problem to O(N) and to offer insights into the epidemic spreading mechanism. Through the new SIR approach, we study the spread of epidemics on any type of network regardless of its topological structure. We analytically derive the epidemic threshold for the new approach, which is inversely proportional to the spectral radius lmax (the supremum eigenvalue within the eigenvalue spectrum) of the network. We perform Monte Carlo simulations to validate the new SIR approach, and we compare it with the SIR heterogeneous mean field approach in the literature. We show that the individual-based approach outperforms the heterogeneous mean field approach when the effective infection rate is close to the epidemic threshold. Analytically, our study shows the role of the centrality properties of the network in the spreading of epidemics. Moreover, we analyze the deviation between the individual-based approach and the continuous time Markov chain model, and we also show that the new approach is accurate for a large range of infection strength. We summarize the contribution of the paper as follows: