Mortality experience is critical for life insurance business to determine correct premium rates, reserve requirements and sound profitability. Many techniques are based on prior mortality experience and simple one- factor prognostic patterns that do not reveal the underlying change processes. The approach suggested in this paper intends to improve the accuracy of mortality predictions through integrating time series analysis coupled with survival analysis. Insurers are in a position to accurately price policies underwritten to mortality probabilities that are more accurate due to application of sophisticated statistical methods and wider range of data sources. These are applied together with clinically collected demographic and health-related data, which are used in time-series forecasting using ARIMA models and in survival analysis by using Cox proportional hazards models. From these results, it can be concluded that the hybrid model performs much better than traditional models for the prognosis of mortality rates, providing a sound approach to the establishment of life insurance premiums. Besides and more importantly, this approach is also used in helping the refinement of premium, the enhancement of strategic planning, as well as meeting the required regulations for the sustainable development and growth of the life insurance service providers.