Background and purpose: This study aimed to manage medical errors before and after the implementation of accreditation in public, private, and social security hospitals of Mazandaran, Iran. Materials and Methods: This descriptive study has been done in 38 hospitals. Data were collected through documents reviewed relating to 2013 and 2014. The paired t-test and Friedman test were used by statistical software SPSS. Results: Results showed that the most and the least percent of reported errors, before accreditation, in sequence, were related to public clinical unit (55.9%) and operating rooms (0.6%), and after accreditation in public clinical unit (46.6%) and operating rooms (2.3%) in teaching centers. The most errors (before accreditation) occurred in the morning (62%) and the least, in the evening (8.3%) in teaching centers. Furthermore, after accreditation, the most errors occurred in the morning (64.8%) and the least, in the night (17.3%) in therapeutic hospitals. Paired t-test showed that there is no significant difference between medical errors before and after accreditation. Friedman test showed that structural/systemic errors reported were the most important medical errors in teaching centers after accreditation and therapeutic hospitals before accreditation (P < 0.05). Conclusion: There is no significant difference between the rate of reported errors before and after the implementation of accreditation. This illustrates that the role of management in controlling of medical errors has been poor, and stronger management should be applied in providing health care services.