Uganda is one of the countries with highest fertility rate in the world, 6.7 children per women. It is estimated that 56 percent of all pregnancies are unintended and the contraceptive prevalence rate in Uganda is 23 percent. Unwanted pregnancy is common and induced abortion is illegal. Unsafe abortion is responsible for significant morbidity and mortality among women in Uganda. Almost 40% of admissions to emergency obstetric care units in Uganda due to unsafe abortion is reported and considered high in international comparison. Studies have revealed that trained midlevel providers can deliver safe post abortion care for incomplete abortion and use manual vacuum aspiration. The prostaglandin E1 analogue misoprostol has been shown to be an effective tool in the treatment of incomplete abortions. This option is so far under-used in developing countries, especially outside the larger hospitals and private clinics. One significant limiting factor in providing safe post abortion care is the lack of providers. So far technical training has been mainly limited to physicians. Training of midlevel providers in misoprostol treatment of incomplete abortion will support task shifting in places where doctors are costly and scarce. By evaluating the effectiveness of mid-level providers (midwives); conducting MVA and administering misoprostol treatment of incomplete abortion the project is attempting to contribute to the reduction of maternal mortality and morbidity and safeguard high quality of post-abortion care.

Women with incomplete abortion will be randomly allocated to undergo a clinical assessment and treatment (MVA or misoprostol) either by physician or midwife with safety and efficacy as main outcomes in a RCT carried out in hospital setting in Uganda. Our hypothesis is that there are no significant differences in effectiveness and safety between manual vacuum aspiration and misoprostol treatment of incomplete abortion provided by physicians and midwife.

The involvement of midlevel providers in treatment of incomplete abortion has previously not been systematically evaluated in African health care setting.

The project concerns a randomised controlled trial (RCT) aiming to determine whether midwives and physicians can perform medical treatment for incomplete abortion equally safely at primary care level. All patients included are women admitted due to incomplete abortion and will undergo clinical examinations and treatment in accordance with standard procedures (Bluhm et al., 2007, WHO, 2010). A total of 880 women seeking with symptoms of incomplete abortion will be included after giving their informed consent and randomised to diagnosis and care provided by either midwives or physicians. The study will be conducted at the National referral hospital (Mulago in Kampala). A coordinating centre will be established at the Mulago Hospital in Kampala in order to monitor the study. Included in the coordinating team physicians will be involved and have the ultimate responsibility for the medical care provided within the study. The coordinating physicians will evaluate and monitor the midwives and physicians performance to safeguard that the care provided is safe and thus do not put women at risk or the health care provider in any legal inconvenience. All health care providers at the study site included will be provided training that follows the structure of Ipas who have updated standardized training modules within abortion and post abortion care which is used internationally. Dr Charles Kiggundu, one of the principal investigators is currently involved in training for providers at national level in Uganda and thus has adequate competence and experience. The health care provider's attitude and communications skill is one core component in the training. The importance of the study and the study procedure will be another important aspect of the training in order to motivate health care providers involved in the care at the selected study sites.

Inclusion Criteria:

• Bleeding
• Contractions during pregnancy

Exclusion Criteria:

• Known allergy to misoprostol
• Uterine size more than 12 weeks of gestation
• Suspected ectopic pregnancy
• Unstable hemodynamic status and chock
• Signs of pelvic infection and/or sepsis