Factors Likely to Affect Community Acceptance of a Malaria Vaccine in Two Districts of Ghana: A Qualitative Study

As part of the effort to inform local and national decision-making in preparation for possible malaria vaccine introduction, this qualitative study explored community-level factors that could affect vaccine acceptance in Ghana and provides recommendations for a health communication strategy. The study was guided by the following research questions:

• "How is malaria prevention (and treatment) understood and undertaken in the communities?
• What is the knowledge and acceptance of existing vaccines in the study communities? How do they affect vaccination practices?
• What are the characteristics of the discourse in the communities about the possibility of a malaria vaccine?
• What is the decision-making process with regard to child health within families in the study?
• What audiences and communication channels have been used in previous health information campaigns? Which were the most successful campaigns?"

In general, malaria in Ghana is hyperendemic, with year-round transmission. A section on immunisation situation provides further context for the present study, such as the fact that 2 immunisation campaigns were held in 2012, prior to the start of data collection for this study: the national immunisation days (NIDs) for polio in March and May and emergency vaccination in the Upper East Region during the same period following a cerebrospinal meningitis (CSM) outbreak that caused 23 deaths in the first 2 months of the year.

The study was conducted in 2 purposively selected districts: the Ashanti and Upper East Regions. In the district selected for the Ashanti Region, the majority of the population is Asante (Akan), the country's main ethnic group. They are also Christian and year-round subsistence farmers. In the district selected for the Upper East Region, the predominant ethnic group is Frafra (Mole Dagbon), the country's second main ethnic group. In contrast to the Ashanti district, this area has a more heterogeneous religious profile, and the single farming season coincides with the rainy season from March/April to September/October. (Malaria is more seasonal here as compared to the Ashanti site).

The qualitative methodology used in this study has its roots in anthropological research. Data were collected between May 2012 and mid-January 2013 by four research assistants with experience in qualitative data collection and knowledge of the local languages (Twi and Gruni). Prior to data collection, research assistants were trained, data collection tools piloted, and community stakeholders informed about the research. A total of 25 focus group discussions, 107 in-depth interviews, and 21 semi-structured observations at Child Welfare Clinics (CWCs), the primary locale for Expanded Programme on Immunization (EPI) activities, were carried out. The final sample size across both regions was 286 participants, including 107 in-depth interviewees and 179 in group discussions.

Results:

• Malaria perceptions: In sum, malaria was acknowledged to be one of the most common health problems among children. While mosquitoes were linked to the cause and bed nets were considered to be the main preventive method, participants acknowledged that no single measure prevented malaria. In most interviews, it was only when prompted (and not in all cases) that people remembered the preventive use of the drug sulfadoxine-pyrimethamine in pregnant women and the programmes in trial areas for IPTi and IPTc (intermittent preventive treatment of malaria in children).
• Vaccine perceptions: In sum, the communities highly valued vaccines and cited vaccination as the main motivation for taking children to CWCs. Nevertheless, knowledge of specific vaccines and what they do was limited. While communities accepted the idea of minor vaccine side effects, other side effects perceived to be more serious could deter families from taking children for vaccination, especially during vaccination campaigns. For example, in the Ashanti Region, many community members associated vaccines, especially the perceived failure of health professionals to properly administer vaccines, with paralysis. Rumors had been associated mostly with vaccines administered during vaccination campaigns as opposed to vaccines administered as part of established services at CWCs. More specifically, study participants spoke of rumours of deaths related to the H1N1 campaign, a school vaccination campaign for polio in the Ashanti Region, and filariasis drug distribution in the Upper East Region. The radio was one of the main sources for the rumours the participants had heard.
• CWCs: In sum, although women are expected to bring their children to the CWC monthly up to the age of 5, and the services are free of charge, attendance at CWCs after age 9 months was limited due to time and cost. Perceptions, evaluation, and acceptance of vaccines are strongly linked to experiences at CWCs. Observations at clinics revealed that while 2 different opportunities for counseling were offered (at the beginning of a the session and during vaccination), little attention was given to addressing mothers' specific concerns and to answering questions related to child immunisation. In some cases, the lack of interaction was associated with the heavy workload of nurses who were feeling the pressure to get everything done, and information was generally incomplete. During the semi-structured observations, observers noted that nurses often reprimanded mothers for a variety of reasons. At the same time, none of the mothers complained about nurses being disrespectful.
• Malaria vaccine: In sum, even as community members and health professionals agreed that it would be important to have a vaccine against malaria, some participants had the perception that a malaria vaccine already existed. This confusion was associated with limited knowledge of specific vaccines. When discussing the possibility of a malaria vaccine with partial efficacy, there was consensus among study participants that people were unlikely to stop using other preventive measures.
• Communication preferences: In sum, community members highlighted 4 main avenues for receiving health information: via radio, information vans, health talks, and trusted relevant people in the community. The communication tools least mentioned were posters and television. In urban areas, people preferred radio and information vans, whereas in rural areas, people preferred community durbars (gatherings of chiefs and people of the community), and they trusted their leaders and community health volunteers to provide the relevant information. Each community described different formal community processes for disseminating health information, as well as certain informal information networks. When deciding on vaccine use, different women stressed that they valued the opinions of other women who had previous experience with the vaccine and those involved in the health field, including community-based agents (CBAs), volunteers of previous programmes, and traditional birth attendants. Health administrators and health professionals described the actual strategy of the Ghana Health Service, which seemed to address participants' preferences: to allow the flexibility needed at the community level; to include communication both with the relevant leaders and, through them, with male heads of households, who had the ultimate responsibility to make decisions; and with mothers, who made the day-to-day decisions, especially those regarding routine trips to the CWC.

In reflecting on these findings, the authors explain that the study identified a number of factors that could facilitate the introduction of a licensed malaria vaccine. "These factors include the highly accepted and well-established routines of the CWCs, the high value placed on vaccination by communities despite their limited knowledge of vaccines, and the shared understanding that malaria prevention requires a comprehensive approach. The finding that communities appeared to be open to using a combination of methods to fight diseases suggests that a partially efficacious malaria vaccine could be part of a tool kit of malaria control strategies. The study has also confirmed the existence of an active and flexible national health communication structure and processes that allow for dissemination and local coordination of messages and activities."

They assert that "a number of challenges could be addressed through a well-planned communications strategy and well-designed messages. A key challenge to be taken into account is the perception by some community members that a malaria vaccine already exists. This perception is associated with the vague knowledge of vaccines and other preventive strategies. Another challenge is the variety of terms used to refer to malaria and their overlap with other conditions, such as yellow fever and pneumonia. This lack of clarity regarding malaria and its symptoms could have implications for the perceived efficacy of an eventual malaria vaccine. Finally, there is space for reviewing the communications aspects of CWC service delivery to allow for an increase in vaccine information and the use of a dynamic communication model that seeks to engage caregivers of vaccinated children and that community members do not perceive as being punitive. Communicating with mothers about the need to continue CWC visits past the nine-month mark would also be important to ensure administration of any additional vaccination, including the measles booster, which is administered at 18 months."