Development of effective messages to promote maternal immunization in Kenya

Objectives: This study evaluated messages and communication approaches for maternal immunization uptake in Kenya. We identified persuasive communication aspects that would inform maternal immunization attitudes, intent, and vaccine uptake. Methods: We conducted a two-phased mixed methods study with pregnant women and their male partners in three regions of Kenya. Discussions were conducted in English and Swahili languages by trained focus group moderators. Baseline measures included a survey and discussions about potential messages and accompanying visuals. Follow-up focus groups with the same participants included a survey about previously discussed messages, visuals, and communication impressions. The second round of focus groups focused on message preferences developed from the first round, along with rank order discussion for final message selection. Following transcription of focus group discussions, we conducted analyses using NVivo software. Quantitative data analyses included frequencies, factor analyses, reliability assessment, regression modeling, and comparative assessment of rank order. Results: The sample (N = 118) included pregnant women (n = 91) and their partners (n = 27) from diverse Kenyan regions (Bondo/Lwak/Siaya, Mombasa, and Nairobi). A four-factor solution resulted from factor analyses that included subscales “positive ad attitudes” (n = 5 items, α = 0.82), “negative ad attitudes” (n = 4 items, α = 0.75), “ad indifference” (n = 2 items, α = 0.52), and “ad motivation” (n = 4 items, α = 0.71). Overall, the positive ad attitudes factor (β = 0.61, p = 0.03) was the only significant component in the overall model examining message selections (χ²₍₆₎ = 262.87, p = 0.17). Among the tested concepts, we found that source and situational cues had a strong influence on women's attitude formation and intention to obtain recommended maternal vaccinations. With self-acknowledged variations in knowledge, participants were particularly attuned to images of relatable women, providers, and depictions in realistic or actual Kenyan clinical settings. Conclusions: The results indicated that positive attitudes were shaped by incorporating highly relatable factors in messages. Implications for subsequent campaigns and research directions are discussed.