Objective: To compare and identify possible differences in the use of the partogram in different models of delivery and birth care. Design/Setting: This was a cross-sectional study performed in two Brazilian hospitals with different models of care: Natural Birth Center (NBC) and Traditional Obstetric Center (TOC). Data were collected from the medical records of 112 mothers with low obstetric risk who underwent an intrapartum cesarean section. Socioeconomic and demographic variables, obstetric history, partogram labor evolution, complications, and indication of cesarean section were included. A significance level of 5% and a confidence interval of 95% were considered. Data were analyzed in accordance with the Brazilian Ministry of Health and World Health Organization recommendations. Findings: In the NBC, the use of the partogram follows the Ministry of Health recommendations for filling-in variables with greater frequency and significance (p < 0.001) beginning with the active phase of labor in the correct column in relation to the lines of alert and action (p = 0.00, OR = 42.2) and continuous records (p = 0.00, OR = 53.3). However, obstetric interventions, such as use of oxytocin, amniotomy and cesarean delivery were concentrated in area 1 of the partogram in both institutions. Conclusions: In this study, the NBC used the instrument most adequately; however there was no difference from the other model (TOC) and regarding the moment of interventions, many of them were early. This finding implies that, regardless of the care model, the partogram has been used bureaucratically and not as a guiding instrument of assistance envisioning safely and timely practices. In other words, aside from a “humanized” physical structure, it is necessary that obstetric practices should be focused on best evidence, thus reducing maternal and perinatal risks.