Abstract
BACKGROUND: Text or voice messages have been used as a popular method for improving women's knowledge on birth preparedness and newborn health care practices worldwide. The Aponjon service in Bangladesh provides twice-weekly messages to female subscribers about their pregnancy and newborn care on mobile phones that they own or share with family members. It is important to understand whether women's singular access to a phone affects their service satisfaction and the adoption of health messages before deploying such interventions in resource-limited settings. OBJECTIVE: This study aims to evaluate the effect of women's singular and shared access to mobile phone messages on their service utilization and perceived behavioral change around birth preparedness and pregnancy care. METHODS: In 2014, Aponjon conducted a retrospective cross-sectional survey of 459 female subscribers who received text or voice messages during their pregnancy by themselves (n=253) or with family members (n=206). We performed multivariable regression analyses to investigate the association between pregnant women's differential access to messages and other socioeconomic factors and outcomes of service satisfaction, ability to recall service short code, ability to identify danger signs of pregnancy, preference for skilled delivery, arrangement of a blood donor for delivery and pregnancy complications, maternal nutrition, use of potable drinking water, and washing hands with soap for hygiene. RESULTS: In the multivariable analysis, women who had singular access to messages had higher odds of reporting high satisfaction (odds ratio [OR] 1.72, 95% CI 1.12-2.63; P=.01), recalling the service short code (OR 2.88, 95% CI 1.90-4.36; P<.001), consuming nutritious food 5 times a day (OR 1.58, 95% CI 1.04-2.40; P=.03), and following the instructions of Aponjon on drinking potable water (OR 1.90, 95% CI 1.17-3.09; P=.01) than women who shared access with family members. Women's differential access to messages did not affect their knowledge of danger signs and preparedness around delivery. Adolescent women and women aged 20-24 years had lower odds of planning safe deliveries than older women (aged≥25 years). Secondary education was statistically significantly associated with women's ability to recall the short code and pregnancy danger signs, plan safe delivery, and select blood donors for emergencies. Higher family income was associated with women's satisfaction, recognition of danger signs, and arrangement of blood donors and nutritious diet. Women who received more than 4 antenatal care visits had higher odds of liking the service, preferring skilled delivery, recalling danger signs, and consuming nutritious food. CONCLUSIONS: The capacity of women to independently access mobile phone messages can improve their adoption of mobile health services and some pregnancy health care practices. A holistic approach and equitable support are required to improve access to resources and knowledge of delivery preparedness among low-literate and younger women in low-income households.