and body mass index, previous prenatal depression (Center of Epidemiological Studies Depression scale), ethnicity, parity, in vitro fertilization pregnancy, smoking, level of studies and prepregnancy physical exercise.As shown in Table 1, the only factor that was predictive of compliance > 83% in both univariate and multivariate analyses was university studies (OR 3.0, 95% CI 1.0-9.2, P = .055).Our findings indicate that having university studies before pregnancy increases the likelihood of better compliance with a supervised exercise program during pregnancy.One study analyzing these factors in the general population, concluded that women with postsecondary education, no children, nonsmokers, and who were also engaged in regular recreational activity, were more likely to have a regular supervised exercise program during pregnancy. 5 All of these factors are associated with the mother's education level. Also, our study shows that maternal education is a predictive factor for compliance to the program, among highly motivated women participating in a randomized controlled trial. However, neither smoking nor practicing exercise previous to actual pregnancy predicted compliance.In conclusion, maternal education in a highly motivated population is crucial for a better compliance to an exercise program during pregnancy. Implementing strategies for improving compliance to an exercise program in a less educated pregnant population would improve exercising during pregnancy. Learning about compliance factors as well as improving healthcare staff knowledge on exercise during pregnancy, and facilitating exercise programs, could help to improve the health of the pregnant population.