Importance of interpersonal communication
The study showed that in the community, IPC was very important. IPC facilitated the exchange information, importantly in ‘our words’, expression of feelings and the receipt of immediate feedback. This meant that doubts and misconceptions could be immediately dealt with and ongoing mutual support developed between health workers and community members and between community members themselves so that new knowledge and behaviours were reinforced. Furthermore, in case of emergency, community members were confident to contact BRAC staff directly by phone. This finding is consistent with another study on changing community attitudes [
35,
36]. Although our study did not compare the effectiveness of IPC with EE, Hussein et al. reported that IPC was more effective than other media when the messages were targeted to lay people [
37,
38]. Two way interactions are deemed essential for identifying the level of readiness for change and to support and convince individuals to adopt health-prompting behaviour [
39]. While messages were being communicated, we found no evidence from community members or CHWS that CHWs are specifically assessing individual’s level of readiness to change.
BRAC has chosen the multidimensional approach to BCC. The context in Bangladesh is one of rapid
social and economic development predicated on community mobilization and empowerment, especially of women. In this context, it is likely that the theoretical need to assess
individual’s level of readiness to change is less important than working with communities, both men and women and across generations for change as the BRAC IMNCS BCC programme already does [
40,
41]. The CHWs are key in communicating and chosen as peer educators because they belong to the same community and have similar socio-cultural backgrounds and face challenges akin to those of the programme participants. The female respondents also share feelings of fellowship with them based on a gender perspective. BCC theories, mostly social learning and diffusion of innovations theory, assert that credible peers can influence health behaviour change [
42,
43].
A recent study on peer educators in HIV/AIDS prevention programmes revealed their positive contribution to prevent HIV/AIDS among adolescents [
44]. However, CHWs are often illiterate or less educated, especially in our study area. Thus while on one hand, CHWs are easily understood and have a growing credibility, on the other hand their lack of formal education, an important status marker in Bangladesh, was considered as a disadvantage in communication. At programme level, CHWs’ technical and communication performance has an impact on their growing credibility therefore the programme should replace refresher training with regular in-service training that is rooted in an experience-based problem solving approach. In this way CHWs ability to assess, manage and take right decisions in saving mothers’, newborns’ and children’s lives will be incrementally increased. In addition, because time spent during IPC is important for increased rapport-building and negotiation about actions to be taken by the family, the technical medical skills in the CHW training programme should be strongly buttressed by further regular training in assertive two way communication [
45]. In the interim, CHWs’ credibility can be enhanced by having their managers and senior officers formally introduce them and by the presence of higher officials in certain group meetings. In the longer term, another way might be for BRAC and other NGO CHWs training and practice to be formally accredited through government and related technical body affiliations. However, since the initial development of Primary Health Care in 1978, (WHO) formal accreditation of CHWs has always been strongly contested because of the need for both sustained political will and long term additional resource [
46‐
48].
Printed materials
We have reported on the research participants’ functional literacy. We also reported community members’ concern to memorise messages. This derives from the importance of recall in a rural community that, barely three generations ago, depended almost entirely on oral communication with a literacy rate of just 16.8 % in 1971 [
49]. Also, school education in Bangladesh, both religious and secular, traditionally includes a lot of repetition and memorization. It is not unexpected therefore, that community members wish to memorise the messages. However, it is important that memorization is not the aim of BCC but rather that memorization prompts timely, appropriate action. Our printed materials included both Bengali text and illustrations and were produced with the objective of being both an aid to IPC and a stand-alone communication medium with this rural population.
The concept of visual literacy refers to the capacity of persons to code and decode visual signs other than words. Visual literacy was first discussed in relation to health promotion by Fuglesangin 1973 [
50]. Although pictures are considered as an effective way to communicate with the rural illiterate people, we found two types of barrier to their comprehension of BRAC MNCH printed materials. The first is the fact that although there was some pretesting of the BRAC materials, some of our illustrations were ineffective as ‘stand alone’ communication media, for example the picture of the pre-eclamptic woman with a headache was initially interpreted by our participants as being a ‘tired woman.’ In the interim, misconceptions regarding some of the existing illustrations can be corrected by reinforcing the training of CHWs to assist in materials comprehension through IPC. In the medium term, the solution to this problem is in the rigorous and repeated pretesting of illustrations with the target population prior to printing new editions. Pre-test and adaptation of materials needs to continue until the community members’ answer to the question, ‘What do you see here?’ aligns with the intention of the designer and health communicator [
51]. The cost of pretesting can be offset against that of the time used by CHWs to explain poorly designed materials. Nonetheless, the programme should not ignore the importance of IPC aided by the flipchart in creating a sympathetic connection between CHWs and community members [
52] and the ongoing usefulness of the stickers as a resource that is available to the families and wider community in the absence of CHWs.
Our rural respondents successfully used culturally acceptable ways of displaying and retaining posters and stickers keeping them in drawers or using rice or flour paste to stick communication materials to smooth mud walls and behind wooden doors. This contrasts sharply with the challenges faced by Dhaka slum dwellers in the sister Manoshi programme [
7]. Living in rented accommodation, some landlords objected to the women displaying these materials and tore them down. Nor did landlords allow nails to be used for fixing posters. Also, urban respondents deemed posters unattractive when stuck to walls made of corrugated iron and other irregular surfaced, often recycled building material. It might be that these female urban slum dwellers, whose 22 % literacy rate [
7] compares unfavourably with literacy of female respondents in Nilphamari (53 %) in 2010 [
23] and who have less access to public space than their rural sisters, are also less able to interpret illustrations and particularly those posted on uneven surfaces which requires the ability to ‘read’ perspective.
Secondly, religious and cultural barriers were found in accepting pictures like those in the BRAC stickers of maternal and newborn danger signs. This barrier derived from the participants concern about developing the same ‘bad fate’ as the complication shown in the illustrations. The ‘bad fate’ barrier however operated less in the context of cultural conservatism and more in the context of the fear of not being able to get treatment because of their poverty and the cost involved in using health services. BRAC already uses birth preparedness messages including saving during the antenatal period for possible medical emergencies. Birth preparedness messages therefore need to be reinforced and more closely aligned to the messages about danger signs and immediate telephone contact with BRAC staff, (already a source of security and assurance), once danger signs are recognised.
At a societal level, community members find themselves in what Van Gennep [
53] described as a ‘liminal’ state. It is not surprising therefore that our respondents were able to hold seemingly conflicting notions of bad fate and fear of economic cost closely together. This response resonates with Lambert’s analysis of the way Rajasthan is used ‘bad fate’ to exert agency and manage their illnesses [
54]. More recent work [
55] about client and health worker satisfaction with inpatient delivery care in northern Bangladesh found that respondents gave a similarly sophisticated and nuanced analysis of what make them, at one and the same time, satisfied and dissatisfied. Both users and providers described in some detail the limitations of the current but constantly changing and developing service. Like our respondents, they are living in a liminal state.
Although BRAC’s local MNCH committees are supposed to give voice to the concerns of local people about health services, the functioning of the committee was also hampered by complex tensions between committee members, especially where those with lower social status had higher formal education level than acknowledged community leaders. However, by using mechanisms that are similar to traditional patron-client relations, MNCH committee members have had modest success in motivating and encouraging community members to follow practices recommended for improved MNCH. There were some reports of improved timekeeping and behaviour with patients by health facility staff as a result of follow-up by MNCH committee members [
56]. The IMNCS programme therefore needs to work with acknowledged community leaders to improve their technical knowledge of MNCH. In the context of growing gender equality, this should become easier as the distance traditional high status male leaders keep from dirty (
napak) ‘women’s matters’ is reduced.
At a macro level, BRAC’s Health and Human Rights and Legal Aid Service (HRLAS) senior staff advocate health services having a legal ‘duty of care’. However, the enactment of such legislation is politically contentious, given the power of the medical establishment. HRLAS could also develop specific rights based messages about negotiating hospital admission, treatment and cost. It could also, potentially, extend it’s legal services to those who, in response to BCC, have tried to use secondary level health services and received inadequate care. A common persistent problem is that of untrained or absent ‘consultant’ obstetricians and anaesthetists [
57]. This problem is a function of increasing utilization of health services, inappropriate staffing configurations, and weak administration of vacant posts and unauthorised absences [
57,
58].
At a national level, Government and NGOs need to address the financial barrier by improving health service coverage and sustaining recent interventions such as health insurance or community based financing for rural people, especially the poor. Proper advertisement, careful supervision and monitoring of any such attempt should be present to ensure such investments reach the rural poor.
Education entertainment
Although messages through EE and mass media had limited reach to the female members of the community, women who did attend EE could easily relate the situations played out with their own lives and perceived realities. BCC through drama and songs was also a very popular mode of communication reported in other studies [
13]. Research also confirmed EE and mass media as highly acceptable and effective for audiences with limited formal education [
59‐
62]. Keeping community people’s eagerness for EE in mind, low cost and low tech communications like local traditional folk songs and theatres may be effective for promoting social and behavioural change. Careful organization and advertising are also important in addition to holding these events. Also, community television with big gatherings or the use of tablets with smaller gatherings and posting videos of
Jarigan and street theatre on You Tube can be another way to broadcast appealing health related programmes. The use of tablet computers is already being piloted in by Plan International in Nilphamari [
63] and by some government community clinics and health workers [
64].
Overall, our research revealed there was less coverage of secondary audiences, particularly men. In the light of the importance of target segmentation in BCC [
65] it is important that the programme should research, design and test more effective BCC methods such as focused meetings; flash cards including the role of men in pregnancy, delivery and postpartum care; and delivery preparedness to increase men’s comprehension and participation in these issues.
This study was not designed to demonstrate behaviour change (before and after) the BRAC MNCH BCC intervention. Our findings about community perceptions of the BCC tools, reveal that a well co-ordinated, simultaneous and repeated use of different channels are likely to continue to be useful to communicate and reinforce key, carefully structured messages and support changing behaviours throughout the community, including men and older persons. In line with other communication research findings, interpersonal communication between CHWs and community members at home visits and group meetings remain immensely important. However, the BRAC IMNCS BCC programme can be further enhanced by the development of gendered storylines and attention to the seasonal cycles of conception, birth, postnatal care and childhood illness with broadcast of seasonally relevant messages. We recommend that messages and media be rigorously pretested before going to scale. While live street theatre and folk song (Jarigan) are well accepted, their usefulness could be expanded by uploading videos of these programmes to social media. Similarly print materials could be uploaded to the web. To increase synchronicity of messaging, mass texting to both CHWs and community members could be used to increase coverage and sustainability. Future research should identify how best to combine IPC, printed materials, traditional cultural forms, social media and mass media in different field situations.
Limitations and methodological considerations
One of the limitations of this qualitative study is that it did not measure the contribution of each BCC component in influencing behaviour change. This was not the intent of this study. Furthermore, health promotion and behaviour change communication researchers are cautious about their ability to disaggregate the impact of different communication media in a multi component intervention [
66,
67]. More recently, although some studies discuss the use of randomised control trials in behavioural interventions they conclude that ‘When interventions are complex, pragmatic trials may be more likely to succeed that explanatory ones’ [
68,
69]. In addition, the IMNCS was operating in the context of a long history of successful community based interventions that date back to the promotion of the use of oral rehydration solution in the late 1970s [
70] and the need to inform the programme in the light of the rapidly approaching target date for the Millenium Development Goals.
As a BRAC researcher, some bias may be present by conducting the research in the BRAC intervention area. Using convenience sampling in selecting the study area is another limitation because some research participants might be missed in remote areas. In qualitative research, there are also possibilities of misinterpretation, loss of information and biases due to translators’ interpretation and assumption [
71‐
73]. Triangulation of information from different groups such as pregnant and postpartum women; husbands of pregnant and postpartum women; mothers and fathers of newborns and under-5 children; mother-in-laws and CHWs using FGD, KII and IGD was a useful strategy for checking consistency and also disagreement within and across the groups [
34].