BACKGROUND: The thesis evaluated an intervention designed to support contraceptive use delivered by mobile phone to women in Bangladesh who had had a menstrual regulation (MR) procedure. The intervention comprised 11 interactive, automated phone messages about contraception which allowed participants to connect to a reproductive health call centre. METHODS: The intervention was evaluated by a randomised controlled trial. The primary outcome was use of an intrauterine device (IUD) or implant at 4 months post MR procedure which was measured via phone interview. Secondary outcomes included intimate partner violence (IPV), a possible adverse effect. Process data was collected via the mobile phone system, the 4-month phone interviews and semi-structured interviews with trial participants. RESULTS: Four hundred and eighty-five women were allocated to the intervention group and 484 to the control group. There was no evidence for an effect on contraceptive use but more intervention participants than control participants reported experiencing physical IPV during the study period (adjusted odds ratio=1.97; 95%CI=1.12 to 3.46). Process data suggested that access to the intervention content and use of the call centre were low. Semi-structured interviews with 30 trial participants found women often concealed the study from family members for fear of revealing their MR and sometimes contraceptive use. Some women felt less able to explain their participation in the study without revealing their MR. MR, contraceptive use and phone calls with men were reported to have been triggers to IPV and familial violence in the past. CONCLUSION: The transient nature of automated voice messages may make them unsuitable for delivering detailed contraceptive information. The increase in IPV appears to be driven by attempts to control women’s reproductive behaviours and their interaction with individuals outside of their household. Care needs to be taken when developing reproductive health interventions.