Leveraging technology to develop and test Family Planning mass media campaigns in Niger and Côte d’Ivoire

Publication date: 2020

INTRODUCTION The use of television and radio for mass media campaigns has been recognized as a high impact practice for increasing access to accurate family planning (FP) information, building self-efficacy, promoting healthy sexual and reproductive behaviors, and promoting supportive social norms.1 Formative research to support the development of focused messaging and to inform diffusion strategies is also recognized as a best practice.2 There is emerging evidence of digital-based innovations (mobile applications, websites, online campaigns and soap opera videos) having a positive impact on the reduction of risky sexual behaviors.3 This activity, conducted under Transform/PHARE4 (hereafter referred to as PHARE), a USAID-funded project (2015-2020), used evidence-based social and behavior change communication (SBCC) strategies to develop and test innovative prototypes that address barriers to modern FP use, transform attitudes about reproductive health, and promote FP in four countries: Benin, Burkina Faso, Côte d’Ivoire, and Niger. The aim of this brief is to present an overview of two uses of technology (interactive voice response (IVR) and Facebook) for FP mass media campaigns in Niger and Côte d’Ivoire (CDI), the challenges and benefits of using technology, and considerations for future use. LEVERAGING TECHNOLOGY TO DEVELOP AND TEST FAMILY PLANNING MASS MEDIA CAMPAIGNS IN NIGER AND CÔTE D’IVOIRE WHAT IS INTERACTIVE VOICE RESPONSE (IVR)? IVR platforms are used to share information and gather feedback on specific questions through interactive SMS or pre-recorded voice calls. IVR can be used with basic phones regardless of location, language or literacy. IVR has been used to successfully provide refresher trainings for providers in Senegal. Between April 2016 and February 2017 in Niger, PHARE used IVR as a means of testing FP messages for a radio campaign which aimed to increase or sustain FP use among women of reproductive age who were proactive about their health and had previous experience or knowledge of FP, known as Healthy Proactives. Between October 2017 and August 2019 in CDI, PHARE used a human-centered design (HCD) process to design and prototype a behavior change intervention that leveraged social media. This intervention included an interactive comic book style story accessed on Facebook, with complementary radio and short message service (SMS) technology to inform young men about Didier episodes and events. The aim of the story was to engage young men in FP use and encourage support for their partner’s choice of contraception, shifting them from being inhibitors to supporters of FP use. 1Family Planning: High Impact Practices, Mass media: Reaching audiences far and wide with messages to support reproductive health behaviors 2Ibid. 3Daher J, Vijh R, Linthwaite B, et al., “Do digital innovations for HIV and sexually transmitted infections work? Results from a systematic review (1996-2017),” BMJ Open 2017;7:e017604. 4The project’s objectives are to: 1) Increase the demand for FP products and services; 2) Enhance the active support for FP among key secondary targets; and 3) Address social norms, thus creating a favorable environment for FP. https://www.fphighimpactpractices.org/briefs/mass-media/ 2 IDENTIFICATION OF THE TARGET POPULATION For both interventions in Niger and CDI, specific population segments were identified to ensure messages resonated with the intended beneficiaries. In 2013 and 2014 in Niger, Camber Collective conducted a study on supply and demand factors related to FP, of which one output was behavioral and attitudinal segmentation. Five archetypes of women with distinct FP needs, attitudes, and behaviors were identified: Modern Elites, Healthy Proactives, Sheltered Skeptics, Traditional Autonomists, and Conservative Passives.5 As information-seekers proactive about their health and who had previous experience with or knowledge of FP, Healthy Proactives were identified as the primary archetype with the greatest potential for FP behavior change. Based on this study, PHARE took the Camber Collective study to the next phase of testing FP behavior change messages with Healthy Proactives. In Cote d’Ivoire, PHARE used HCD to better understand the context in which young men in the informal sector live and work, to learn relevant ways to approach them and how best to engage them in support and use of FP, both for themselves and their partners. Special focus was put on young men in the informal sector as this population is difficult to reach through traditional channels, such as formal educational structures or workplace interventions, and is often overlooked in communication campaigns. The HCD process categorized young men from the informal sector into four archetypes: the hustler boss, the oblivious playboy, the honey-mooner and the curious virgin. To varying levels, the different archetypes were generally resistant to using FP, and their behaviors often inhibited their partners’ use of FP, highlighting the need for gender norms to be addressed. All four archetypes were selected for the intervention due to a number of commonalities across archetypes as well as the interactions between them (different archetypes talk to one another or certain ones looked up to others), with the overarching target population being young men in the informal sector. WHAT IS HUMAN CENTERED DESIGN (HCD)? HCD is an approach to problem-solving that places the people being served and other important stakeholders at the center of the design and implementation process. This allows their needs and expectations to inform the design decisions and therefore contribute to the intervention having a greater impact. The HCD process has three phases: Inspiration: learning and opening up to creative possibilities, grounded in the desires of the people being designed for; Ideation: idea-creation stage, testing and refining ideas, building a prototype; and Implementation: building partnerships and getting an idea out into the world. RAPID MESSAGE TESTING WITH IVR IN NIGER The team opted to use IVR to test messages to be used for a FP radio campaign. IVR allowed for rapid testing, tweaking, and tailoring of messages based on responses and provided an alternate tool to gather feedback from individuals. PHARE tested three messages targeted to Healthy Proactives. Messages addressed rumors of health providers engaging in negative behaviors and misconceptions about side effects and were translated from French into two local languages: Hausa and Zjerma. Radio announcements invited women of reproductive age to call a toll-free number to participate in a FP survey, with a chance of winning phone credit (the amount varied on levels of completion of the survey). Upon calling the number, women would select their preferred language and answer several preliminary questions that would help the IVR platform assess whether they were considered a Healthy Proactive. If they were eligible, they would proceed to answer survey questions about the three FP messages. If they were not eligible, they were screened out of the survey. Callers that participated in the survey were asked about their comprehension of the messages and whether the messages motivated the caller to use a FP method, discuss FP with their husband/partner, or speak to a health provider. PHARE analyzed responses from 5Camber Collective, FP in Niger: Using customer insights to reshape policy and programming http://www.cambercollective.com/fpniger 3 “I DON’T LISTEN TO RUMORS ABOUT CONTRACEPTIVE METHODS. I TALK TO MY HEALTH WORKER FOR THE BEST INFORMATION.” the first IVR survey and refined the messages for the second IVR survey. Radio announcements were again used to invite women to call in to the second survey, where revised messages were tested. The messages were scored based on a weighted average of actual comprehension and motivation compared to the desired comprehension and motivation, with greater weight given to motivation than comprehension. The two messages that scored the highest were recommended for the FP behavior change radio campaign:6 “IF THERE ARE ANY SIDE EFFECTS, I WILL GO BACK TO MY HEALTH CENTER FOR ADVICE.” USING A SOCIAL MEDIA-LED APPROACH TO ENGAGE YOUNG MEN IN CÔTE D’IVOIRE HCD-informed insights were used to develop a youth- friendly brand, Didier, to provide an entry point to engage with young men about FP. Two seasons, each comprising 10 episodes, of a comic book style story about a young man named Didier were posted on Facebook to generate interactive discussions around the topics of sexuality, unplanned pregnancy, condom use, and gender equality. Using this site, which is popular among young people, the Didier Facebook page provided a fun and safe space for follower interaction. Each episode was followed by a question and answer session, with discussion prompts allowing followers to interact with Didier via the Facebook (FB) wall or direct messaging, either commenting on story content or asking questions about topics addressed in an episode. A designated staff member monitored the FB page, identifying misinformation or biased comments to engage in dialogues to unpack certain ideas to maintain quality control and promote accurate information. The same staff member would delete offensive or inappropriate content. Originally designed as a weekly episode in comic book style format, following feedback from young men after Season 1 of Didier, the episodes were also made available in audio format for those who were not literate. Questionnaires were posted on Facebook to gather insights and feedback on the Didier story. To complement the information available on FB, community-based activities including live events and regular Didier club meetings were organized to engage young men from the target population and discuss Didier-related content. Live events included theatrical presentations, music, games, one-on-one and group discussions centered around Didier’s story. The latter phase of the project also included provision of voluntary FP counseling and services. Didier clubs were organized with young men working in the informal sector to meet weekly and discuss the Didier episodes, providing an opportunity to strengthen the story’s messages through interpersonal communication. A total of 20 clubs, each with 20 members, were organized during Season 2, and hosted by trained facilitators. To promote Didier more widely, radio and SMS were used. A radio campaign was launched with two local radio stations to raise FP awareness and promote new episodes of Didier. The radio campaign also served as a means of informing girls and parents about the intervention and was used as a conversation starter for discussions around FP. Young men from the informal sector were identified through neighborhood youth leaders in Abobo and Treichville. A database was created with contact information from the young men who agreed to be contacted about Didier-related activities and SMS was used to send messages to inform them of promotional events, Didier-related clubs, or to direct them to the latest episodes or FB discussions. LESSONS LEARNED The experiences from Niger and CDI demonstrate benefits of using mobile, social media, and radio technologies in behavior change interventions, and also highlight some of the challenges in achieving their intended impact. 6English translation from original French messages 4 Quick application: After initial set-up, IVR can be quickly implemented and can be used for iterative testing at scale, without requirements of holding in-person events. Survey control: The PHARE team had control over the survey at all times; they could modify messages as needed and monitor response rates in real-time. Learning and adapting: Improvements to survey response options could easily be made as needed. For example, binary response options were modified to multiple choice responses to overcome too many “yes” responses, and to allow for a better assessment of true understanding and motivation of the FP messages. Rapid data analysis: Responses collected in the first IVR survey were used to adapt messaging and quickly re-test in the second survey. The use of technology for collecting data allowed for quicker analysis of data when compared to more traditional research methods, and therefore was especially useful for quickly doing repeat surveys. Appropriate accessibility: The format for IVR message delivery can be adapted depending on the context. In Niger, voice messaging was chosen due to high levels of illiteracy. Messages were also made available in three languages. Responses remain confidential and the system can be accessed when and where the respondent desires, as opposed to having an in-person interview or focus group. Lengthy set-up time: Initial set-up of the IVR survey was timely, taking approximately five months for the development, translation, and recording of each survey question to be uploaded onto the IVR platform. In Niger, the survey was made available in three languages which increased the time it took to set up, as well as tailoring response options to women who were not literate. Low survey response rate: The project team anticipated that 600 Healthy Proactives would complete each survey round, based on estimates of the population and proportion of Healthy Proactives, access to phones, and expectation of an estimated 50% survey completion rate based on Camber Collective’s experience of implementing similar activities. However, there were far fewer call-ins than expected (365 in the first survey and 162 in the second). The variance in number of callers between the two surveys may have been influenced by the length of time the radio advertisements were aired for; for the first survey, they were aired over three months but only over one month for the second survey. High caller drop-out rates: Among the low caller numbers, not all were eligible for the survey or did not complete it. Only 31 Healthy Proactives completed the first survey, and 21 the second survey (5% and 3.5% respectively of the anticipated sample of 600 per survey). In the first survey, caller drop-out was mainly due to non-eligibility of callers but also to many women (n=104) not answering a question about their age. This was attributed to lack of familiarity with using IVR. To ascertain if women were of reproductive age and thus eligible to BENEFITS OF USING IVR FOR FP SURVEYS IN NIGER CHALLENGES IN USING IVR FOR FP SURVEYS IN NIGER 5 participate, callers were asked to input their age using digits on their phone dial pad. In the second survey, the question was changed to selecting your age from an age bracket, and non-response rates dropped significantly (down from 26% to 10%). In the second survey, the main reason for drop-out was non-eligibility of women as Healthy Proactives. Costs: It cost $9,531 to conduct the first IVR survey, and $3,156 to conduct the second (with higher set up costs for the first survey). A local firm estimated that a similar survey using face-to-face methods would have cost approximately $5,000 for the first round and $1,000 for the second round.7 The low numbers of callers and Healthy Proactives who completed the IVR survey made this a costly approach estimated at $307 per survey completed by a Healthy Proactive in the first round and $150 in the second round. Yet, it is difficult to do an exact cost comparison as the IVR survey allows for rapid and consistent analysis and adaptations compared to in-person focus groups as well as availability of multiple languages and anonymity of participants. BENEFITS OF USING SOCIAL MEDIA TO REACH YOUNG MEN IN THE INFORMAL SECTOR IN CÔTE D’IVOIRE Wide reach: Social network platforms such as FB provide an optimal entry point for reaching a large number of young men (and women) at any time of the day. The number of followers grew from 25,000 during Season 1 to over 57,000 at the end of Season 2. Sharing relatable content: FB provides a social, fun space to share a range of relatable content and an interactive platform for open discussions between followers and with Didier. By creating a FB page for the Didier story and incorporating characters based on all four archetypes, men could identify with the characters and their experiences. Real-time adaptations and learning: Weekly episodes posted on FB followed by immediate discussions about the content of each episode, allowed for real-time insights on the content. They also allowed PHARE to better understand which questions or discussion topics young people engaged with or liked, informing timely adjustments to future content. Adaptations were made based on feedback received outside of FB, at a community-level. For example, an audio version of Didier was developed following insights gathered from focus groups with young men, which revealed that not all young men could read Didier in a comic book style. Complementary technologies: Radio and SMS worked well together to direct young men to the FB site and to Didier clubs or events. SMS in particular enabled the target population to be reached as PHARE established a database of phone numbers early on in the project, whilst the radio enabled a wider range of audiences to be reached, including men from the target segment not in the SMS database. CHALLENGES OF USING SOCIAL MEDIA TO REACH YOUNG MEN IN THE INFORMAL SECTOR IN CÔTE D’IVOIRE Limited targeting potential: FB as an open platform provided limited scope to target a specific segment. It also meant that Didier was only available to young men accessing FB. Age, sex and geographical location were the only data collected, making it 7PSI, IVR Pretesting https://www.psi.org/publication/ivr-pretesting/ 6 difficult to determine whether Didier was reaching its target population of men in the informal sector. To help overcome this issue, targeted SMS messages and Didier events and clubs were a way to ensure the target segment was being reached. Impact not sufficiently assessed: The impact of Didier on its FB followers – and more specifically on the target segment - was not sufficiently assessed to determine the impact on young men’s FP use or support for their partner’s FP use. Online questionnaires were posted on FB, but results were unreliable due to low response rates. In the final phase of the project, a survey was conducted with 100 members of the Didier clubs, which did reveal positive changes in attitudes to FP use, especially related to joint decision- making regarding FP use. There was a 4% increase among participants who believed both the men and women should be involved in making the decision to avoid pregnancy and in the decision to use FP.8 CONSIDERATIONS FOR FUTURE USE IVR can be a fast and cost-effective means to test iterative messages confidentially. If IVR is used minimally within the target population, sufficient time must be built into programming to allow for questions to be designed in a user-friendly way. Programs must consider how to direct callers through the survey as well as time for adapting questions that cause challenges for callers. To overcome low call-in rates, effective means to invite beneficiary populations to participate is essential. For instance, a “push” survey could be used if databases of phone numbers from the target populations are already available. Didier proved to be appealing to a large number of people, as demonstrated by the number of followers, but the nature of FB meant its followers were likely beyond the intended target group, so the reach and impact of Didier among the target population was difficult to assess. Complementing broader social media interventions with specific community-based interventions can ensure that the intended audience is reached. During design, programs could consider introducing eligibility criteria using polls to monitor how their target population is voting or interacting with the content. The use of technology for FP mass media provides an opportunity to reach a large number of people with appropriate FP information, often at national scale, and allows messages to be adjusted and improved quickly based on real-time feedback. Using technology for developing and testing FP mass media campaigns also provides users with a safe and anonymous space to interact when they are free. This is a particularly important factor in contexts where FP is a taboo subject, especially among certain populations. In order to reach the intended population with the technology in question or to know that the intended population is being reached, programs need to have a strong understanding of technology use and access by the target population. For technology to be an effective means for developing and testing FP mass media campaigns, its use must be appealing and familiar to the target population and necessary measures must be in place to tracks its use and impact among those it set out to reach, including the incorporation of appropriate monitoring techniques and tailored impact evaluation designs. This brief is made possible by the generous support of the American people through the United States Agency for International Development (USAID). This brief was prepared by PSI for USAID, Contract Number AID-OAA-TO-15-00037. The contents are the sole responsibility of PSI and do not necessarily reflect the views of USAID or the United States Government. For more information, please visit: www.psi.org/project/transformphare/. 8Results collected after four months of implementation. www.psi.org/project/transformphare/

View the publication