Summary Background Spousal violence is a global issue, with ramifications for the reproductive health of women. We aimed to investigate the effect of physical, sexual, and emotional violence on potentially preventable single and recurrent spontaneous fetal loss. Methods We analysed data from the Cameroon Demographic Health Survey. In the violence module of this survey, women were questioned about their experience of physical, emotional, and sexual violence inflicted by their spouses. Respondents were also asked about any stillbirths and spontaneous abortions. We measured risk for single and recurrent fetal loss with odds ratios, with adjustment for intracluster correlations as appropriate. We also estimated the proportion of preventable excess fetal loss at various levels of violence reduction. Findings 2562 women responded to the violence module. Those exposed to spousal violence (n=1307) were 50% more likely to experience at least one episode of fetal loss compared with women not exposed to abuse (odds ratio 1·5; 95% CI 1·3–1·8). Recurrent fetal mortality was associated with all forms of spousal violence, but emotional violence had the strongest association (1·7; 1·2–2·3). If the prevalence of spousal abuse could be reduced to 50%, 25%, or entirely eliminated, preventable excess recurrent fetal demise would be 17%, 25%, and 33%, respectively. Interpretation Spousal violence increases the likelihood of single and repeated fetal loss. A large proportion of risk for recurrent fetal mortality is attributable to spousal violence and, therefore, is potentially preventable. Our findings support the idea of routine prenatal screening for spousal violence in the African setting, a region with the highest rate of fetal death in the world. Funding None.