Implementing health communication tools at scale: mobile audio messaging and paper-based job aids for front-line workers providing community health education to mothers in Bihar, India

Introduction

The thousand days between conception and an infant’s second birthday is a highly vulnerable period requiring critical health interventions to promote optimal growth and development and to combat the major threats to morbidity and mortality. Yet despite the clear evidence for effective solutions to improve health for pregnant mothers and infants in this period—such as antenatal care, birth preparedness practices, skilled birth attendance, early and exclusive breast feeding and skin-to-skin care—reductions in morbidity and mortality have fallen short of global goals.1 Impacts of interventions may be attenuated by the challenges of maintaining their quality and coverage at scale.2 3 This is particularly true in rural and under-resourced areas where there is often a severe shortage of trained and supervised healthcare workers to implement them4 and transportation to healthcare facilities may be limited. Thus, women in these areas may have poor access to care, including key reproductive, maternal, newborn and child health and nutrition (RMNCHN) services.5

A growing body of research has suggested that a primary solution for improving access to quality healthcare in low-resource settings is to expand front-line worker (FLW) delivery of health education and services. It has also been shown that training and supportive supervision are critical to ensure successful delivery of reliable and well-communicated health messages.6 7 To provide this support and to further expand quality healthcare for marginalised populations in hard-to-reach places, mobile health (mHealth) services have been employed to empower FLWs and their beneficiaries with novel methods of education and communication. In the past decade, there has been a rapid expansion of mobile phone technology available throughout low-income and middle-income countries (LMICs), creating a new opportunity to support the delivery of educational information by providing interactive voice response (IVR) and short message service (SMS)-based training via mobile phones. Health training can be developed using engaging and scientifically vetted educational information which can be implemented at remarkable scale quickly and cost-effectively to reach diverse geographies, ethnicities and languages.

Multiple systematic reviews have assessed the impacts of mHealth interventions in low resource settings, suggesting that the use of mobile technology for clinical decision support and FLW training has led to substantial improvements in the quality of care provided.8 9 Our studies of mHealth tools supporting FLW delivery of RMNCHN services in Bihar, India, showed significant improvements in quality and frequency of service delivery, leading to improved health behaviours of beneficiaries.10 11 Furthermore, mobile-delivered education, when designed using an equity lens, can provide life-saving information in even the hardest to reach and otherwise media-dark areas.12

While the utilisation of mHealth tools is promising, particularly in low resource areas where access to information and health services may be limited, it is critical that they be designed with a human-centred approach and subsequently evaluated to assess the benefit for knowledge and health-related behaviours.13 In this study, we assess ‘facilitated communication’ tools by evaluating the impact of a package of mHealth and paper-based FLW job aids to determine if their use was associated with improved key RMNCHN knowledge, greater interpersonal discussion, and improved behaviours among childbearing women in Bihar, India.

Methods

Results

Discussion

In the studies of both the GSP and IPC interventions implemented by BBC Media Action during VHSNDs, significant differences were observed in the knowledge and health-related behaviours of women who were exposed to the tools compared with those were not. Women who were exposed to the mHealth GSP content were significantly more likely to recall health messages and were also more likely to report knowledge and healthy behaviours associated with that information. They were also more likely to report appropriate IFA consumption and current use of contraception, as well as improved knowledge about birth preparedness, TT vaccine administration, growth monitoring, complementary feeding and pneumonia care. They were less likely, however, to have had their children receive immunisations, likely in part because most women primarily attended VHSND events for the purpose of obtaining vaccines and therefore, their intention to practice was already high. The content related to preventive benefits of immunisations was also thought to be too technical, and the messages about disease avoidance not immediate enough to motivate behaviour change. Women exposed to the IPC tools during VHSND sessions were more likely to report appropriate knowledge and health practices for IFA consumption and contraception usage, as well as improved knowledge of diarrhoea management. There was no difference, however, in the use of ORS for diarrhoea. For both tools, exposed women were at least twice as likely to subsequently discuss most health topics with another person, such as their husband, relative or friend.

Overall, this study demonstrated that exposure to BBC Media Action’s mHealth-based VHSND tools was associated with a range of improvements in RMNCHN-related knowledge and practices, as well as discussions by beneficiaries with others regarding the topics. These findings align with the growing evidence for the role of mHealth solutions in improving RMNCHN programmes in LMICs. Many previous studies of mHealth tools have shown targeted benefits, particularly in ANC compliance and breastfeeding practices.20–22 However, they have less often demonstrated benefits for postnatal care practices such as immunisation compliance and contraception usage.23–25 Yet while the evidence is mounting for the benefit of digital solutions that provide direct-to-beneficiary communications, such as SMS reminders for ANC behaviours or stage-specific educational recordings,9 or those that target improved workflows for FLW healthcare delivery,10 26 27 far fewer rigorous studies have assessed the impacts of mHealth solutions for facilitated communications tools for FLW–beneficiary interactions. Our findings suggest that digital solutions for facilitated communications reflect the known benefits of using job aids for patient communications in amplifying the impacts of FLW-beneficiary interactions.11 28 Information, however, must be effectively designed and rigorously evaluated for its benefit in the target population. For instance, information that is too technical such as that regarding ORS therapy for diarrhoea or the mechanisms of disease prevention from immunisations, may be ineffective in driving behaviour change as was demonstrated by the lack of impact in these domains in our results. Finally, while it has been suggested that the most significant impacts from mHealth are seen in those programmes that combine mHealth and non-mHealth interventions,29 30 further study is required to evaluate the domains and behaviours which are most likely to be impacted by digital compared with non-digital tools, as well as potential synergism between them, with additional qualitative assessment of beneficiary experiences of these interventions.

This study was limited in that all surveys were reliant on maternal self-report, and thus exposed to both recall and social desirability biases. Further, they were collected cross-sectionally when multiple other interventions were being implemented as a part of the Ananya programme in the same geospatial areas.14 31 While the study design allows for the assumption that women would have had an equal chance of having been exposed to other initiatives, it cannot be ruled out that women who attended FLW-facilitated discussions may be more likely to also attend other educational events, such as radio programmes and street theatre. Further, it is difficult to differentiate whether the benefits for women who were exposed to the tools were due to the content of the tools or simply the confounding effect of having been exposed to an FLW, thus making it difficult to attribute impact to specific interventions. Due to the study design, we were not able to compare the digital versus the non-digital tools with regard to their benefit for knowledge or behaviour change. Additionally, while sampling areas were selected randomly, the study design was not experimental and there may have been selection bias in intervention exposure when FLWs chose those beneficiaries for whom they would use job aid tools. Importantly, the study design was such that no sampling weights were applied. Finally, because intensive support and facilitation were provided for FLWs by BBC Media Action during the implementation period, generalisability may be limited given the challenges for sustainability and scalability of interventions bolstered by such extensive support.

As the use of mHealth tools becomes an increasingly common mechanism for the delivery of health services and information across LMICs, rigorous evaluations of their impacts are critically important, particularly for implementation at scale. The Shaping Demand and Practices programme of BBC Media Action was intended to improve RMNCHN knowledge and practices across eight focus districts and approximately 28 million people,32 with subsequent scale-up statewide to more than 100 million using mobile-based job aids and health messaging tools in Bihar. While the challenges of the government-supported statewide scale-up of these interventions are described elsewhere,14 31 implementation in the eight focal districts alone required the training of over 110 000 FLWs. Thus, the scale of implementation and the rigorous evaluation across multiple datasets was unique in its contribution to the literature on mHealth.11

Conclusion

As technology advances and smart phones become more ubiquitous, even in hard-to-reach places, a rich opportunity has emerged for the delivery of quality healthcare through mHealth interventions. Our analysis has shown that implementation of digital and non-digitally facilitated communication tools for FLW support are associated with higher levels of self-reported knowledge and healthy behaviours, as well as subsequent discussions with others in the family. Further study is required, however, to understand how mHealth tools can be utilised most effectively, and in what contexts. For instance, assessments should address whether digital health tools are more effective than paper-based visualisation tools, particularly for supporting less educated or marginalised groups; or when and for what topics digitally facilitated communication is superior to direct-to-beneficiary digital tools. Only then can digital and mHealth assessments be optimised in their evidence-based use for purpose. Importantly, future evaluations of the effectiveness of mHealth interventions must also focus on health outcomes, and the long-term sustainability of health-related improvements. Technology tools continue to create opportunity for improved health impacts at scale, but their use must be evidence-based to ensure cost-effective implementation and sustained benefits for their beneficiaries.