Background: To analyze life expectancy in cities in Shaanxi Province from 2015 to 2019 and explore whether it have geographic differences. And assessing the impact factors related to these differences. Methods: Life expectancy was estimated by life table and the spatial dependence was analyzed by constructing the spatial weight matrix. We used ordinary least squares regression and dominance analysis to judge the influence of socioeconomic and health behavior factors on life expectancy, and the final contribution of influencing factors to independent variables was generalized by Meta regression. Findings: From 2015 to 2019, life expectancy in Shaanxi Province increased by an average of 2·23 years. The Moran'I showed that there was a significant spatial correlation in life expectancy from 2017 to 2019. The cities in the region of Guanzhong had a high-high clustering. In 2019, there was a low-high abnormal phenomenon in Xianyang City. The socio-demographic Index had the highest relative contribution to life expectancy in various cities although physical exercise had the highest regression coefficient (β=0·364, P<0·01). The impact of smoking has been relatively weakened in the past five years. The impact of Hypertension on life expectancy in Shaanxi Province has become greater. Interpretation: Geographic difference in life expectancy in every city in Shaanxi Province is obviously. This difference can be explained by socioeconomic factors and healthy behavior factors. Funding Information: This work was funded by National High-tech R&D Program (863 Program) for High-end Electronic Medical Record Technology Development and System Application Demonstration [grant No. 2012AA02A607]. Declaration of Interests: None reported.