Today, the odds that a woman in Sub-Saharan Africa (SSA) will die from complications related to pregnancy and childbirth is one in 20 compared to one in 6,250 in the developed world, resulting both from high fertility and maternal mortality rates (UN, 2012). In a region where the majority of countries have experienced armed conflict since the end of the Cold War, we argue that the poor performance may in part be due to detrimental effects of armed conflicts on the access to life-saving maternal health care. This paper aims to increase our knowledge of how the exposure to armed conflict impacts individual women's use of maternal and reproductive health (MRH) services, specifically antenatal care and professional birth assistance. To examine this question, we combine disaggregated conflict data from the Uppsala Conflict Data Program's Geo-referenced Event Dataset (UCDP-GED) with geo-referenced maternal health data from 72 Demographic and Health Surveys conducted in 31 countries in Sub-Saharan Africa in 1990–2013, covering information on nearly 600,000 births by almost 400,000 mothers aged 15-49. We use a quasi-experimental approach (mother fixed-effects analysis) to estimate the impact of recent armed conflict patterns within a radius of 50 km of the home of each mother on a child's likelihood of being born in a medical facility. Hence, we are able to control for a vast amount of variables that may otherwise be spuriously correlated with both maternal care and conflict. In line with our expectation, the results indicate that conflict intensity significantly reduces the probability that a women give birth at a medical facility. Although the level of maternal health care is lower in rural areas the negative impact of conflict seems to be stronger in urban areas. Also, poor women and women with less education are also more strongly affected by conflict with regard to maternal health care. Finally, there appears to be a sudden drop in the care variables precisely in the month a conflict event happens, which takes approximately three years to restore.