Objectives: To investigate whether the interval between diagnosis and initiation of treatment above 60 days in women who underwent outpatient treatment between 2000 and 2015 by the SUS in Minas Gerais, Brazil, is associated with the patients' macroregions of residence. Methods: Non-concurrent longitudinal study of a cohort of 8872 women from Minas Gerais who underwent outpatient cancer treatment (chemotherapy or radiotherapy) under expenses of SUS in the same state, from 2000 to 2015. The response variable was the interval between diagnosis and initiation of treatment of women with cervical cancer (up to 60 days or more than 60 days) and the explanatory variable was the patients' macroregion of residence (13 macroregions of residence). The covariates were age, cancer stage at diagnosis, first treatment received, number of comorbidities, and distance between municipality of residence and treatment. Logistic regression was used to assess associations. Results: The average number of days to start treatment was 50.7 days, and most started within 60 days (61.9%). About 83% of women had their treatment in the same macroregion of residence; however, only 40% of women had their treatment in the municipality where they live. It was found after all adjustments that living in any of the 12 non-A health macroregions increases the chance to start treatment after 60 days from diagnosis. The highest magnitude of association were found in the F macroregion, with OR equal to 4.27 (95% CI; 3.32-5.50). Conclusions: There is a disparity in the timely treatment of women among health macroregions of the state of Minas Gerais. Investigations of the possible inequalities and barriers in the itinerary of women with cervical cancer that influence timely treatment are needed.