Background. Until July 2001 medical abortion was only authorized in France in public hospitals or private clinics. A new law effective in July 2001 allows private practitioners (gynaecologists or general practitioners) to provide medical abortion in their offices as long as they are linked to a hospital official agreement. Unfortunately mifepristone was not available outside hospitals before September 2004, so the study was conducted still providing the drugs in the hospital family planning clinic. Objective. To evaluate the efficacy and the acceptability of mifepristone medical abortion with home administration of misoprostol provided by private practitioners linked with the hospital. Patients and methods. Four hundred thirty three women seeking medical abortion before 7 weeks LMP were included between 2 January 2003 and 7 July 2004. All consultations before abortion and 2 weeks after abortion took place in a private provider’s office. Drugs were administrated in the hospital family planning clinic: patients were given 3 tablets of mifepristone (600 mg) orally by the midwife and received 2 tablets of misoprostol (400 μ g) that they would take at home 48 hours later. In case of any problems or complications, patients could phone or meet their private providers, phone the hospital midwife or go to the hospital emergency service. Private providers received training inmedical abortion training and could at any time reach a medical specialist in the hospital family planning clinic for information or to refer a patient. Results. Efficacy was evaluated for 339 women, because 94 patients were lost to follow- up (21.7% ). Efficacy of medical abortion was 93.8% (318/339). There were 21 surgical aspirations (6.2% ): for women’s decision in 1.5% of cases, for medical decision without complications in 3.5% of cases, and for failure of the method in 1.2% of cases (2 ongoing pregnancies and 2 heavy haemorrhages with transfusion). The family planning midwife received a phone call from 21 patients after mifepristone (4.8% ), Twenty- five patients had an emergency consultation (5.7% ), and 22 patients went back to their private providers before their appointment for follow- up (5% ). Twenty- two patients (5% ) were referred by the private provider to the hospital medical specialist. Acceptability is known for 26% of patients; 96.2% thought that the abortion procedure was acceptable. Discussion and conclusions. The failure rate of medical abortion in this study is largely due to aspirations for incomplete abortion. To improve the efficacy of medical abortion offered by private providers linked with the hospital, all the relevant professionals (private providers, residents in the emergency service, family planning providers) must be well trained in medical abortion, especially in how to interpret and react to ultrasound images obtained in the follow up visit. The procedure is very acceptable to women. Medical abortion offered via a network should be well accepted by practitioners, since only 5% of women will need more than two consultations and only 6.2% will need surgical aspiration in the hospital. This study allows us to be optimistic about the expansion of medical abortion in France outside the hospital via a provider- hospital network based on the fact that since September 2004 private providers can get mifepristone directly in the pharmacies of the city. Background. Until July 2001 medical abortion was only authorized in France in public hospitals or private clinics. A new law effective in July 2001 allowing private practitioners (gynecologists or general practitioners) to provide medical abortion in their offices as long as they are linked to a hospital official agreement. Unfortunately mifepristone was not available outside hospitals before September 2004, so the study was conducted still providing the drugs in the hospital family planning clinic. Objective. To evaluate the efficacy and the acceptability of mifepristone medical abortion with home administration of misoprostol provided by private practitioners linked with the hospital. Patients and methods. Four hundred thirty three women seeking medical abortion before 7 weeks LMP were included between 2 January 2003 and 7 July 2004. All consultations before abortion and 2 weeks after abortion took place in a private provider’s office. Drugs were administrated in the hospital family p lanning clinic: patients were given 3 tablets of mifepristone (600 mg) orally by the midwife and received 2 tablets of misoprostol (400 μg) that they would take at home 48 hours later. In case of any problems or complications, patients could phone or meet their private providers, phone the hospital midwife or go to the hospital emergency service. Efficacy was evaluated as 339 women, because 94 patients were lost to follow-up (21.7%). Efficacy of medical abortion was 93.8% (318/339). There were 21 surgical aspirations (6.2%): for women’s decision in 1.5% of cases, for medical decision without complications in 3.5% of cases, and for failure of the method in 1.2% of cases (2urrent pregnancies and 2 heavy haemorrhages with transfusion). The family planning midwife received a phone callfrom 21 patients after mifepristone (4.8%), Twenty-five patients had an emergency consultation (5.7%), and 22 patients went back to their private providers before their appointment for follow-up (5%). Twenty- two patients (5 % of patients were admitted by the private provider to the hospital medical specialist. Acceptability is known for 26% of patients; 96.2% thought that the abortion procedure was acceptable. Discussion and conclusions. The failure rate of medical abortion in this study is largely due to aspirations for incomplete abortion. To improve the efficacy of medical abortion offered by private providers linked with the hospital, all the relevant professionals (private providers, residents in the emergency service, family planning providers) must be well trained in medical abortion, especially in how to interpret and react to ultrasound images obtained in the follow up visit. The procedure is very acceptable to women. Medical abortion offered via a network should be well accepted b y practitioners, since only 5% of women will need more than two consultations and only 6.2% will need surgical aspiration in the hospital. This study allows us to be optimistic about the expansion of medical abortion in France outside the hospital via a provider- hospital network based on the fact that since September 2004 private providers can get mifepristone directly in the pharmacies of the city.