This study develops a model of different types of political regime changes and their effect on life integrity violations. The data cover 147 countries from 1977-93. A multivariate pooled cross-sectional time-series research design is used, employing Ordinary Least Squares regression with panel-corrected standard errors. This research focuses on different types of regime change that vary in direction and intensity. It tests the effect of political regime changes on life integrity violations at time t and at time t 2 1. The results show that regime change in general does not alter the levels of repression. However, change from autocracy toward democracy decreases repression during the transition period, whereas change from democracy to anocracy increases life integrity violations. It is also shown that the higher the magnitude of democratization, the less state violence occurs during the year of the change, once relevant control variables are considered.