2.1 Study area

This study was conducted in two rural districts located in the department of Mono in Southern Benin, namely Bopa and Houéyogbé, where complementary feeding practices of young children were inadequate (Mitchodigni, Amoussa Hounkpatin, Ntandou-Bouzitou, Avohou, et al., 2017; Mitchodigni, Amoussa Hounkpatin, Ntandou-Bouzitou, Termote, et al., 2017) and where a high percentage of households experienced food insecurity (41% in Bopa and 34% in Houéyogbé), as compared to the national rate (10%) (INSAE & PAM, 2017). In 2015–2016, a community nutrition and health sensitization program using videos or posters was implemented in the area for a period of 6 months to improve young children's feeding practices and nutritional status (Bodjrenou et al., 2020). The present study took place in the exact same districts. However, we excluded the villages that had directly benefited from this program in the past, as we wanted to hear from people who had not previously been exposed to these tools. From the list of the remaining villages, we randomly selected two villages per district, namely Dado and Tanvé in the district of Bopa, and Hounvi-Atchago and Salahoué in the district of Houéyogbé.

2.2 Study population and selection of participants

We conducted a cross-sectional survey combining qualitative and quantitative approaches, which targeted mothers of children under 2 years old, community leaders at the village level and nutrition and health care workers (NHCWs) at the district level. NHCWs were defined as any person or professional working as an educator in an action or intervention for health, food, or nutrition. We first created an inventory of all institutions and structures intervening in the field of nutrition and health in each of the two districts by consulting the administrative registers and maps of the institutions operating at the district level and by discussing with the administrative authorities. Thirty institutions and structures were identified, including social promotion centers, health centers, NGOs, and interventional projects in nutrition. From each structure, one person actively involved in community activities was selected to participate in the study (n = 30). In parallel, we used purposive sampling to select participants at the village level. The inclusion criteria for mothers were: (i) having a child under 2 years old, and; (ii) having lived in the village for at least 2 years. A total of 42 mothers were selected with the assistance of village authorities and were deliberately chosen from diverse sociocultural backgrounds (ethnicity, religion, etc.). Community leaders were selected from each village as follows: the village chief, two of the chief's advisors and at least six other influential people or village leaders (when possible, men and women were equally represented). A total of 39 leaders were selected from all four villages.

2.3 Data collection

At the district level, we conducted semi-structured interviews in French with NHCWs, using interview guides that included both closed- and open-ended questions. Interviewees were asked to make an inventory of the tools their institutions used during nutrition and health sensitization sessions. Next, we gathered their opinions on the relevance of videos in learning sessions in the Bopa and Houéyogbé communities. We also simulated learning sessions on two different topics, using either posters or videos, then investigated NHCWs' opinions and perceptions on both tools. The topics and tools used in this test were directly drawn from the nutrition awareness-raising program conducted in 2015–2016 in the same area (Bodjrenou et al., 2020): (i) Dietary diversity: presentation of food groups and their roles, concept of a balanced diet, importance of dietary diversity, inventory of local foods within each food group, etc.; (ii) Hygiene and deworming: recommended hand washing occasions for mothers and children, importance of washing hands with soap and water, consumption and conservation of drinking water, sanitation of living area, protection of foods against flies and insects, worms and their effect on health, importance of deworming. For each theme, one poster and one video were produced and used to convey identical messages. Posters were mostly graphic with some text in French. We took care to alternate the order of presentation of both themes and both tools across structures.

In each district, the themes (dietary diversity and hygiene/deworming) were randomly assigned to villages (Figure 1). In each of the four villages, two focus group discussions (FGDs) were held separately using local language, one with mothers and the other with community leaders. Sociodemographic data of participants were collected, including age, gender, ethnicity, religion, level of formal instruction, length of residence in the village, marital status, number of children and income-generating activities, as well as previous participation to nutrition sensitization sessions and what tools were used during those sessions. As with NHCWs, learning sessions were simulated during FGDs using either posters or videos on the same theme. Video-based sessions used the following procedure: (i) small groups of four to six participants viewed the video twice on computers; (ii) the facilitator asked participants to share their understanding of the video; (iii) the facilitator paused the video to give additional explanations while soliciting participants' thoughts on the content and use of video as a tool; (iv) participants asked questions and received answers from the facilitator. Poster-based sessions used the following procedure: (i) the poster was displayed on a wall or board; (ii) the facilitator asked participants to view the poster and explain what they understood; (iii) the facilitator then gave further explanations of the poster's content before opening the floor for discussion.

Furthermore, direct non-participant observations of participants' attitudes were conducted during all the poster-based and video-based sessions. FGDs focused on evaluating the videos in comparison to the posters. Each FGD, which gathered 9–11 participants was facilitated by a moderator, aided by an assistant who was responsible for taking notes on the conduct of the FGD and reminding the moderator of any overlooked elements. Participants provided written consent for photography and audio recording of the discussions. Each FGD used the local language by default, but participants were free to speak in any language, as long as it was understood by all other participants, as checked by the facilitator.

2.4 Data processing and analysis

A double entry of quantitative data was carried out and validated using SPSS 22.0 software. Descriptive analyses (frequencies and means) were carried out to determine participants' sociodemographic characteristics and describe nutrition and health NHCWs' feelings about the videos and the messages conveyed therein. Data from the FGDs and interviews were translated from local language into French; data from FGDs and interviews were fully transcribed. Community leaders and mothers' perceptions, opinions, and preferences regarding the videos and posters were manually coded by two researchers and analyzed using a thematic content approach. Discussions occurred in case of coding disagreement. We focused on stakeholders' opinions on the suitability and appeal of the videos (versus posters) to rural populations, as well as the clarity of their intended messages. We evaluated two additional parameters with community leaders and mothers: the ability of videos to hold their attention (concentration) and the ease of memorization of messages. Close direct non-participant observations of participants' attitudes while watching the videos allowed assessing the concentration. We also used the data for a Strengths, Weaknesses, Opportunities, and Threats (SWOT) analysis of the use of videos and posters in nutrition sensitization programs.

3.1 Sample characteristics

Most of the institutions surveyed in this study have been operating in Bopa and Houéyogbé for more than 10 years (Table 1) and 77% of them had organized nutrition learning sessions. NHCWs were predominantly women. At the village level, the majority of community leaders were over 40 years old, had five children or more and had been living in the community for more than 10 years. All the mothers surveyed were between 15 and 49 years old, and most of them were married and had fewer than five children. The Sahouè was the dominant ethnic group in both mothers' and community leaders' groups. Moreover, 31% of mothers and 59% of leaders had received no formal instruction (Table 2).

We did not observe any differences in the perceptions of participants related to the order of presentation of the two communication media. Nor did participants' socioeconomic and demographic characteristics had any particular influence on their perceptions of the videos and posters. Thus, the presentation of our results does not take these parameters into account.

3.2 Inventory of communication media used in nutrition sensitization programs

Among the institutions that organized nutrition awareness-raising sessions (n = 23 out of 30), five (5) did not use any communication media at all. The others (n = 18) mainly used, in descending order, image/picture boxes (n = 11), posters (n = 6), radio broadcasts and theaters (n = 2 for both), informational handouts, and cooking demonstrations (n = 1 for both); videos were seldom used (n = 1). Some institutions used more than one material.

In the villages, 48 out of 81 community leaders and mothers who participated in the FGDs had already benefited from nutritional advice, mainly from health centers. In 29 cases, these nutrition sessions incorporated no communication media. Picture boxes were used in 16 of the cases and posters in 3.

3.3 Assessment of learning videos and posters

3.3.3 Clarity of messages/comprehension

Pictures presented in posters were easily misunderstood, and very often the facilitator had to explain the poster several times before participants understood the different messages. Some pictures from the hygiene and deworming poster were considered ambiguous by almost all community leaders and mothers, and participants found them difficult to understand without explanations. For instance, one mother said, ‘I don't understand this picture (Figure 2a). Why using a net to cover meal? We are not used to putting our meals under a net here in our village, but rather we ourselves sleep under the net’. The video, however, showed a woman covering her food, so participants understood the purpose immediately. Likewise, there was some confusion around the photos showing worms (Figure 2b), as most FGD participants in both districts reported seeing ‘something in the picture that looks like a dish of macaroni/spaghetti’. A community leader from Houéyogbé explained, ‘Videos show things better than the poster. For example, in the video on the topic of deworming, we can clearly see how the worms enter the body from dirty hands. On the poster, we don't see that. The facilitator was obliged to explain it to us. With the video, it's really clearer and easier to understand’. The same was true for the video on dietary diversity. ‘The way the roles of different food groups were explained in the video is engaging and facilitates understanding. In the poster, we just saw several photos of foods. We could not know what these mean’, said one mother. However, several mothers highlighted the exhaustive nature of the posters' content as compared to videos; a mother from Houéyogbé commented, ‘The poster related to food diversity shows a greater variety of foods available here in our village as opposed to the video, which shows only some of them. Thus, I prefer the poster to the video’.

Opinions of NHCWs contrasted more than those of respondents at the village level. Approximately half of the former group thought the videos addressed the topic better than the posters, while the reverse was true for 20% of the respondents. For 33% of respondents, there was no difference between the videos and the posters. Most of those who preferred videos to posters pointed out that the videos were in local languages and thus participants could understand their messages without any other explanations, while the posters required further explanations from the facilitator. Globally, topics were better understood through videos than posters, especially when dynamic processes were addressed. After the presentation of videos, participants asked fewer questions related to misunderstandings in comparison to poster-based presentations; according to interviewees, this was because the videos were clearer than the posters.

3.4 SWOT analysis

Videos offer major opportunities, including their attractiveness and the clarity of their messages (Table 3). Since the videos were in local languages, they could be self-utilized, even by people with little or no formal instruction. As videos do not require a facilitator's presence except for during a question-and-answer session after the projection, they represent a major advantage in terms of standardization of messages. On the other hand, videos offer limited time for imagination, reflection, and learning processes because of their temporality; they may therefore not be suitable for learning large amounts of messages. Video also requires non-negligible logistical features, including electricity and specific equipment; however, in settings where these amenities are not a problem, it can be used in both small and large groups. Posters offer the possibility to deliver a large amount of learning messages, but the clarity of messages and learning processes from participants very much rely on the facilitator's skills (speaking skills, interaction with participants, etc.) and on standardization across facilitators. Static images from posters are likely to facilitate memorization and can be displayed permanently so that participants can access the material at will. Since posters are more traditional and common than videos, they may be perceived as an overused medium and thus be associated with a lack of motivation and mobilization from the community.