The sample design consisted of the stratification of the age range of interest and the regions where the PHCU are present in the city, according to data from the 2010 demographic census and the service production of the PHCU, followed by simple random sampling (SRS) to determine the clusters, SRS of the attendance periods and systematic sampling of the users. For the analysis, the data were used of 907 PHCU users for three years or longer. The users were organized in two groups, one of which contained 686 users without a high risk factor (presence of family history of breast and/or ovarian cancer) and the other 221 users with a high risk factor. Descriptive analysis was applied with the inclusion of a confidence interval for the proportion (CI=95%) calculated by the bootstrap resampling method, and bivariate analysis by means of the chi-squared test to assess the research hypotheses. In the tests, a two-way α of 0.05 was considered, with the support of statistical software. The predominance of the mulatto color was observed in 531 (58.9%) users, 547 (61.1%) were not economically active, 415 (52.3%) belonged to economic class C, 868 (95.7%) did not have a health insurance and 327 (36.1%) had either finished secondary education or not finished higher education. Concerning the marital status, 400 (44.1%) were single and 373 (41.1%) married. The variables annual clinical breast examination (CBE) and practice of the breast self-exam (BSE) were associated; as well as participation in educative activities and accomplishment of the BSE and; opportunistic screening situation and economic class, education level and marital status. The screening actions for breast cancer were classified as "Appropriate" when they complied with the recommendations for each age range and for the presence or absence of a high risk factor, or "Inappropriate" in case of noncompliance. In the group without a high risk factor, 94 (71.1%) users in the age range from 35 to 39 years were considered with appropriate actions, 55 (21.2%) in the age range from 40 to 49 years and only six (1.7%) in the age range from 50 to 69 years, as two (33.4%) had unaltered CBE and mammograms every two years and four (66.6%) had CBE with alterations and were forwarded for mammograms. For the group of users with a high risk factor, only one (1.7%) was considered with appropriate actions in the age range from 35 to 39 years; only four (4.5%) in the age range from 40 to 49 years, and seven (7.4%) in the range from 50 to 69 years, as they had CBE and BSE each year. This study showed the predominance of inappropriate screening actions for breast cancer in the city of São Luis and is expected to contribute for managers, health professionals and teachers from higher education institutions, with a view to the adoption of measures for the practice of breast cancer screening in São Luis.