This study evaluated the effect of automated interactive voice messages promoting contraceptive use among women in Bangladesh who had undergone menstrual regulation. The intervention did not increase contraceptive use, but the main focus of this article was to explore the effects of the intervention on intimate partner violence (IPV). The researchers took precautions to minimize risk of harm, such as letting the women listen to a sample message at enrollment and asking them if receiving similar messages on their phone was acceptable to them. Still, the intervention appeared to have an unintended consequence of increasing self-reported IPV. When women were asked about such violence using a direct, closed-ended question that named specific acts of violence, 11% of participants in the intervention group reported physical violence during the 4-month study period compared with 7% of those in the usual care group. When asked using an open-ended question—”Did anything happen to you as a result of you being in this study? Good or bad?”—there were no differences between groups.
The study highlights the need to take into account possible negative impacts when designing and evaluating mHealth interventions. It also suggests that direct, closed-ended questions should be used to measure IPV instead of open-ended questions.
Authors: Reiss, Andersen, Pearson, et al.
September 2019 issue | Original Article