To examine the long-term impact of large-scale training targeting midwives in a setting where they are the main female genital mutilation (FGM) practitioners. We hypothesised that trained midwives would have significantly higher knowledge, greater opposition to midwives’ involvement in this practice, and improved clinical practice in FGM prevention and care compared with non-trained midwives.

We conducted an exposure based cross-sectional study, using closed-ended and open-ended questions during phone interviews.

Khartoum State in Sudan has a high prevalence of FGM (88%) mainly performed by midwives.

Midwives who received (n=127) and did not receive FGM training (n=55).

We developed primary outcomes aligned to the three levels (reaction, learning and behaviour) of Kirkpatrick’s training evaluation model for descriptive and multivariable analyses in Stata.

All the midwives interviewed were female, mostly village midwives (92%) and worked in health centres (89%). The mean age and midwifery experience was 51 years (SD=10) and 23 years (SD=12), respectively. Overall, most midwives (>90%) reported being supportive of FGM discontinuation. Midwives who had FGM training were more aware that performing FGM violates code of conduct (p=0.001) and reported to always counsel patients to abandon FGM (p

Though past trainings were associated with higher knowledge and greater opposition to midwives’ involvement in FGM, this was not translated into appropriate corrective clinical procedures among affected women during labour. The Sudan Ministry of Health invested heavily in training midwives and it would be important to investigate why trained midwives do not implement recommended FGM-related clinical management.