Objective: Domestic violence is the most common form of violence against women and a major health problem worldwide. The aim of this study was to examine the relationship between demographic factors and domestic violence during pregnancy through meta-analysis. Method : This meta-analysis study was conducted in Iran. All the articles published during 2001 up to Jun 2018 were extracted independently by 2 trained investigators from domestic and foreign databases including, Science Medlib, SID, Web of Science, PubMed, Science Direct, Irandoc, Medline, Scopus, Magiran, and Google Scholar with keywords and their compounds. The results of studies pooled using the random effects model Cochran and I2 tests were used to check heterogeneity. Data were analyzed using Stata Ver. 11.2. Results: A total of 28 articles with the sample size of 15 020 people were included in the study. The findings of the meta-analysis showed that low level of maternal education (OR:1.68) (CI 95%:1.15, 2.46), low education level of the spouse (OR:1.73) (CI 95%:1.31, 2.29), unemployment of the husband (OR:1.61) (CI 95%: 1.05, 2.48), and smoking of the husband (OR:2.51) (CI 95%: 1.64, 3.84) were important factors in the increase in domestic violence during pregnancy. Having 3 children or fewer (OR: 0.30) (CI 95%: 0.16, 0.56) and enough and regular visits to physicians to receive adequate prenatal care (OR: 0.31) (CI 95%: 0.16, 0.57) were deterrent for violence during pregnancy. Conclusion: Based on our findings, level of education, unemployment, prenatal care, smoking, and number of children are associated with violence during pregnancy. Thus, paying attention to these factors and controlling them can reduce violence during pregnancy and its adverse consequences.