Community sensitization and decision‐making for trial participation: A mixed‐methods study from The Gambia

Abstract Background Ensuring individual free and informed decision‐making for research participation is challenging. It is thought that preliminarily informing communities through ‘community sensitization’ procedures may improve individual decision‐making. This study set out to assess the relevance of community sensitization for individual decision‐making in research participation in rural Gambia. Methods This anthropological mixed‐methods study triangulated qualitative methods and quantitative survey methods in the context of an observational study and a clinical trial on malaria carried out by the Medical Research Council Unit Gambia. Results/discussion Although 38.7% of the respondents were present during sensitization sessions, 91.1% of the respondents were inclined to participate in the trial when surveyed after the sensitization and prior to the informed consent process. This difference can be explained by the informal transmission of information within the community after the community sensitization, expectations such as the benefits of participation based on previous research experiences, and the positive reputation of the research institute. Commonly mentioned barriers to participation were blood sampling and the potential disapproval of the household head. Conclusion Community sensitization is effective in providing first‐hand, reliable information to communities as the information is cascaded to those who could not attend the sessions. However, further research is needed to assess how the informal spread of information further shapes people's expectations, how the process engages with existing social relations and hierarchies (e.g. local political power structures; permissions of heads of households) and how this influences or changes individual consent.


| BACKGROUND
Individual informed consent is an essential ethical requirement and standard procedure to all research involving human participants.
'Respect for persons', one of the internationally agreed ethical principles, stipulates that potential participants should have (i) access to sufficient information; (ii) competence, i.e. having the capacity to understand that information; (iii) free choice, decision making in the absence of coercion or deception; and, (iv) comprehension of the subject. 1 Thus, participation in research requires individuals to be in a position to freely give informed consent after the individual consent interview, in which all relevant information is provided, discussed and understood by the potential participant.
However, research shows that translating the principle of 'respect for persons' into practice is often challenging. 2 Though some challenges are ubiquitous, the ethical pitfalls are more salient in resourcelimited settings. 3 In some of these settings, participation in health research may be the only way to obtain (free or better quality) health services; furthermore, high levels of illiteracy may hamper the comprehension of informed consent. 4 In order to improve potential study participants' comprehension of research, community level meetings prior to the start of the research have been recommended. 5 Community level consultations have been implemented by the Medical Research Council Unit Gambia (MRCG) in the form of community sensitizations, which is a process whereby research staff arrange meetings to make information on the research available in the villages from which potential research participants may be recruited. All villagers are invited to a central meeting space where the study is explained and questions can be addressed. At the end of this process, permission to carry out the research in the village may be given by the Alkalo (village leader) or a representative in case of his absence. Although such permission cannot, in any way, replace individuals' consent, it represents the preliminary agreement of the community.
This study aimed to assess the relevance of community sensitization for individual decision-making and to identify additional factors influencing people's participation in research in rural Gambia.

| Study site and population
This mixed-methods study was part of the anthropological research conducted within the framework of two different malaria studies carried out by the MRCG in rural Gambia (Table 1) The Gambia is one of the smallest and poorest African countries. 6 In this resource-poor context, most families' livelihood is based on the farming of groundnuts and maize, small-scale informal trade and remittances from relatives that have migrated to urban areas of the country and abroad. 7 Treatment seeking itineraries are characterized by medical pluralism. Access to health care varies greatly between study villages, but is overall more challenging on the North Bank due to structural factors such as the distance to the health center or hospital, transport options and road infrastructure. Despite the common perception that the health center is the best place to receive treatment, adults and elderly people often delay visits or rely on home treatment.

| The Malaria Transmission Dynamics study and Prinogam malaria treatment trial
This mixed-method study was ancillary to two malaria studies carried out by the MRCG in rural Gambia (Table 1). These two studies were selected because they (i) involved community sensitization meetings preceding the start of recruitment; (ii) were similar in terms of organization, profile of fieldworkers and target population; and, (iii) they contained anthropological work packages that included long-term ethnographic research carried out by the same researchers. 8

| Malaria Transmission Dynamics study
The first study was an observational study to determine the dynamics of malaria transmission in different sites across the country (Malaria transmission dynamics in The Gambia: Defining the spatial and temporal spread of malaria at micro-level (village); referred to hereafter as the "Malaria Transmission Dynamics" study) (Table 1). 9 Community sensitization meetings were organized in all 12 study villages. The study targeted all community members older than 6 months residing in the study villages.

WMA. Declaration of Helsinki -Ethical Principles for Medical Research Involving Human Subjects.
Minors between 10-18 years old were asked to provide assent in addition to the parents' consent; both agreements were necessary for participation. For children <10 years of age, only the parents' informed consent was needed. The informed consent interview was conducted in the local language, based on an English written consent form and information sheet. The study participants were actively followed up during the rainy season by monthly blood sampling carried out by field nurses and passive case detection at local health facilities. In addition, a monthly survey was carried out by the fieldworkers. As part of this study, a mass drug administration campaign was carried out in a later phase.

| Prinogam
The second study was a randomized controlled trial carried out in the Central and Upper River Regions of The Gambia on the efficacy of different doses of primaquine in clearing gametocytes in asymptomatic malaria infected individuals (Primaquine's gametocytocidal efficacy in malaria asymptomatic carriers treated with dihydroartemisinin-piperaquine; ClinicalTrials.gov: NCT01838902; referred to hereafter as "Prinogam") (Table 1). 10 As part of Prinogam, community sensitization meetings were held in each village. A couple of days after meeting the community, the field staff moved from compound to compound and verbal consent was requested for a rapid clinical assessment and a Rapid Diagnostic Test (RDT). At this stage all people above the age of one, and weighing more then 10 kg could be included. Blood from a single finger prick from all non-febrile people was used to prepare a blood slide and perform an RDT. If positive, the blood slide was taken to the laboratory to confirm the infection and estimate the parasite density. Patients were further recruited in the study if they consented to participation after undergoing the full informed consent interview. The interview was conducted in the local languages, based on the English written consent form and information sheet of the study, or with the support of an ad hoc multimedia tool in the local languages as part of a nested trial. 11

| Study design
The anthropological study used a sequential mixed-methods design comprising qualitative and quantitative methods. 12 Qualitative research was carried out in both studies in order to get an in-depth understanding of the research context. Qualitative research was carried out before, during and after the community sensitization in the context of the The field staff made individual visits to compounds directly after the community meetings to discuss the study in more detail and to carry out the individual informed consent. Children between 10-18 years were asked to provide an assent, in addition to the parents' consent; for children <10 years old, there was only the parents, informed consent. The study participants were actively followed up during the rainy season by monthly bleedings carried out by field nurses and passive case detection at local health facilities. The field nurses collected a blood sample for the detection of malaria infections by PCR (polymerase chain reaction), and in case of symptomatic individuals an RDT for the immediate detection of a malaria infection was performed. In addition, a monthly survey was carried out by the fieldworkers. Each year, the community sensitizations were repeated in order to discuss the study with villagers and inform about new procedures. In 2014-2015, a mass drug administration was carried out in the beginning of the rainy season.
Community sensitization meeting in each village. The field staff moved from compound to compound and verbal consent was taken for a rapid clinical assessment and an RDT. Blood from a single finger prick from all non-febrile people was used to prepare a blood slide and perform an RDT. If the RDT was positive, the blood slide was taken to the laboratory to determine the parasite density. Participants with a Plasmodium falciparum mono-infection with a density of at least 20 parasites/μL were informed of the result and invited to the clinic the following day. After obtaining the written informed consent, a finger prick blood sample was collected for haemoglobin (Hb) measurement and G6PD (Glucose-6phosphatase dehydrogenase) screening. If eligible, participants were referred to the trial clinician for randomisation and enrolment. Each participant was followed up on days 1, 2, 3, 7, 10, 14, 21, 28, 35, and 42 or on any other day outside scheduled visits if they felt sick. At each visit, a blood sample (about 0.5ml by fingerprick) was collected for the determination of gametocytaemia, parasite clearance and Hb. In addition, a venous blood sample (3ml) was collected from a subset of 100 participants that consented for a direct membrane feeding assay to determine infectiousness to mosquitoes on day 7.

| Data collection
Data were collected using individual interviews, group discussions and participant observation.
In-depth interviews were based on question guides focusing on a broad range of possible factors potentially influencing people's decision-making regarding trial participation such as the informed consent procedures, social relationships, the meaning of blood and perceptions regarding the MRCG. Interviews were carried out with community members to assess their attendance during the community sensitization, their intention to participate and their perceptions regarding the MRCG. Interviews were recorded and fully transcribed. When it was not possible and/or inappropriate to record the interviews, conversations with informants were written down during or immediately after the interview. Interviews were mostly carried out by the researchers and their research assistants in English or in one of the local languages.
Group discussion were carried out when informants agreed to be interviewed together, for example, women who were looking after children in the same compound, men chatting in the village bantaba (central meeting place).
Participant observation. The researchers and fieldworkers repeatedly visited and stayed in the study villages for a few days at a time.
During these stays, people's everyday routines were observed and informal conversations about relevant themes were held with adult community members, which were written down after the interview as soon as appropriate. The long-term engagement with the study villages facilitated trust between the social science research team and the study participants enabling to discuss more sensitive topics such as, for example, fears of trial participation and other barriers such as conceptions of blood relevant to blood sampling.

| Sampling
Sampling was theoretical (i.e. in accordance with emerging results from the data collection in order to contribute to theory building). 13 Informants were selected from both study sites. People were interviewed regardless of their attendance to the community meeting and their individual decision to participate in the malaria studies.
Informants were categorized according to relevant criteria (e.g. gender, profession, village) in order to obtain maximum variation. In addition, snowball sampling techniques (i.e. where participants identify additional respondents) were used. 14 This technique was particularly useful when it concerned (i) more sensitive topics such as perceptions regarding blood whereby the referral by somebody within their social network increased the new respondent's trust in the research team, and (ii) made it possible to contact people who were hard to reach, such as marabouts (traditional healers). Infants or children below the age of 14 were not included in our sample since different informed consent procedures were used (assent from parents or guardian in addition to informed consent).

| Analysis
Data analysis was concurrent to data collection. Preliminary data from the interviews, group discussions and observations were analyzed to inform the interview and observation guide; additional data was consequently collected to confirm or refute temporary results until saturation was reached. Interviews were systemized and analyzed with NVivo Qualitative Analysis software (QSR International Pty Ltd. Cardigan UK). Quotes were selected for this manuscript to illustrate main themes.  together with their wives and children. Authority figures (compound heads) within each compound were then selected to respond to the questionnaire as the exploratory qualitative research had indicated their principal role in decision-making processes affecting the compound.

| Sampling
When compound heads were not available during the visit of the interviewer, the next person in the internal hierarchy was selected, which in most cases was a household head or spouse.

| Analysis
Quantitative data was double entered in EpiData 3.1. (CDC, Atlanta; WHO Geneva, 1996) and cleaned and analyzed in SPSS (IBM SPSS Statistics 22). Frequency tables and cross-tabulations for the main outcome variables were produced. Ethics of the American Anthropological Association (AAA). All interviewees were informed before the start of the interview about project goals, the topic and type of questions as well as their right to decline participation or to interrupt the conversation at any time. Verbal instead of written consent was preferred as requesting the subject's signature could have been a potential reason for mistrust. All original recordings, transcriptions and notes were anonymized and access to the social science database was restricted to members of the research team.

| Qualitative strand
In-depth interviews (n=238) and informal conversations (n=17) were carried out with community members and field staff ( Table 2-3). Group discussions (n=9) were conducted when informants agreed to be interviewed together, for example, women who were looking after children in the same compound, men chatting in the village bantaba (central meeting place) (Table 4). In addition, on several occasions observations were done by the research team and recorded in field notes. The research team observed (i) community sensitization meetings in eight different villages; (ii) reactions to the Prinogam trial at community level and at the health center; and, (iii) daily events in the study villages.

| Quantitative strand
In total 124 questionnaires were completed; there were no refusals.

| Communication during sensitization meetings
The qualitative research demonstrated that both research staff and community members considered community sensitization procedures important to the success of the study ( According to the people of village X, the Alkalo of village Y is very harsh to his own people. He is a divide and rule form of ruler. He is always in trouble with his own people in the village. Even the program that MRC brought to the village, the Alkalo wants to create a problem, to boycott the program due to his personal feelings without considering the interest of the village as a responsible leader. He even convinced his own relatives not to join the MRC program because he told them the MRC is always involved in blood taking which has a big impact on our health by weakening our body. But thanks to the good people of the village, most especially the youths and some village elders, who have played a vital role in convincing the people to participate in the MRC program with or without the Alkalo. This was based on the sensitization made by the MRC people, which has been fully appreciated by the community.

Informal conversation
Farmer Malaria Transmission Dynamics

Quotes regarding benefits of research
There is a high degree of hope that when you take that medicine, from the time that you have taken it to the end of the rainy season, you won't encounter the problem of malaria, but this cannot be proven now. How can we put this to prove? It's only through testing that we can prove this. That is if we test it over and over.

Fieldworker Malaria Transmission Dynamics
You see this Coartem, the new medicine that you are seeing here, we were the first people who made a research on it in the Gambia. Now the whole country is using it. So, when we know the medication that is introduced now, would be beneficial, then we can add it onto our treatment facilities. This is what would be the toubabs (white people) pride. If there is any benefit for the toubabs, then that is the benefit.

Community sensitization
Fieldworker Malaria Transmission Dynamics

Quote regarding side effects
The new medicine is just like Paracetamol or chloroquine or any other medicine… In the sense that when you take it, it can cause dizziness or high body temperature because the body is not used to it. It can also give you a little stomach disorder, but apart from that, it brings no other problem.

Quotes regarding blood samples
It can be that you are not feeling sick during the rainy season but that you are sick. We will do a blood prick on filter paper. The MRC has a machine to test and see if you have malaria or not. We will count the malaria parasites. We want to know why some people are stronger against malaria? Why is malaria aggressive for some and not for others?

Community sensitization
Fieldworker Malaria Transmission Dynamics Some will claim that MRC is here to withdraw your blood. The ways in which blood is withdrawn is known to all… they collect your blood and put it in a plastic bag and tell you to give it to the person who needs it. But this one is just blood checking and if anybody asks, tell them that this is blood checking.
We are here to check whether you have malaria or not, we are not here to withdraw your blood.

Community sensitization
Fieldworker Malaria Transmission Dynamics R: I can now see how some people of MRC are misquoted. Other people like saying that the MRC are taking blood, and they take their blood. But a little amount of blood is not enough for a person to sell. People are misquoted, it is not right. I: Do people think that the blood sample will be used for witchcraft, the devil or bad eye? R: I don't see anything true in that. The little blood will not be enough. It is just a sample. I: it is a common belief of people? R: No, if people suspect it, the MRC will not even be able to hold a community meeting in this village.

Interview
Farmer Malaria Transmission Dynamics

Quotes regarding individual informed consent procedures
The most important thing is to say that I will participate willingly because I want to, and not because Mr. X is participating so I will also participate just for the sake of participation. Your participation should be based on something. It has to appeal to your interest and this is why we give you something that the toubabs called the consent form.

Fieldworker Malaria Transmission Dynamics
It was just about participating in the MRC study. It is not a force. Before you participate in the MRC study they will ask you questions to make things clear to you, they are not there to bother anyone but they are only there to help people. Because even the pricking, they normally only prick the finger and then the drugs will be given to us. The blood that is taken is not much, it is only a little that is collected.

| Attendance during community sensitization
The survey indicated that 38.7% (48/124) of respondents had attended the Prinogam meeting (Table 7) and observations and interviews confirm that many community members were not present during the Malaria Transmission Dynamics study meeting (Table 8).
Those members of the community who were perceived as in charge of healthcare and research participation (i.e. compound leaders and mothers) tended to go to the community meetings (Table 7). People who were not present at the community sensitization but did participate later on in the Prinogam study did not always know or could not recall what the study was about (

| Awareness about medical studies
Although only 38.7% of all respondents were present at the community sensitization meetings of Prinogam (Table 7), 92.7% of all respondents were aware that the MRCG was planning a study in their village when surveyed after the community meeting and before individual screening for study (Table 9). This to some extent expected discrepancy can be explained by the MRCG fieldworkers encouraging people to spread the news about the study and community members' active strategy to send representatives of the compound to the community sensitization meeting to inform others afterwards (Table 10). The survey showed that 9.2% of respondents who did not attend the meeting reported that they had sent somebody else to the meeting to obtain information (Table 7). People developed this strategy since (i) they thought it was important that their compound was represented during all types of community meetings, and (ii) because certain people did not have time to attend the community meetings due to livelihood activities but still wanted to be informed (Table 10).

| Intention to participate
Overall, almost all (91.1%) respondents stated they wanted to participate in the Prinogam study although recruitment and the individual informed consent procedure had not yet started (Table 9). When differentiating between levels of awareness, both qualitative and quantitative data indicated that respondents who went to the community meetings were more interested in participating (95.8% (46/48)) than those who were completely unaware of the study (55.6% (5/9) (i.e.
did not go to community sensitization meeting and did not hear information about the study) (

| Reasons to participate
The MRCG was generally described as 'good'; (40.3%) and seen as improving health (21.8%) or giving benefits (21.8%) (  (Table 14). These individual benefits were highly valued given the limited access to other health care providers. People were very well aware that these benefits were linked to participation in the studies.

| Barriers to research participation
The perceived barriers to research participation among the nonparticipants were travelling, old age, sickness, reluctance to give a blood sample and the disapproval of the household head (Table 14).
For people who were undecided about their participation, critical factors that would influence their decision were the amount of blood to be taken, the agreement of the household head and the potential benefits of participation. Qualitative research during both studies confirmed that the approval of the compound head or husband was an important factor for the decision-making of other people in the compound or household (Table 14). In rural Gambia, the male compound which is associated with depleting life-force (Table 14).

| Perceptions of MRCG
Qualitative research before and after the community meetings of both studies showed that the perceived benefits and disadvantages of participating were related to (i) the reputation of the MRCG in the  (Table 7), and the overall willingness to participate prior to individual consent (91.1%) ( Table 9).

| DISCUSSION
Community sensitization is often proposed as a necessary strategy to ensure the success of medical interventions and research programs, by both research staff and community members and it is often part of the procedures aiming to improve the community engagement and T A B L E 1 0 Illustrative quotes on awareness about medical studies

Quotes on trial awareness Data source Informant Study
In our Fula community (the sensitization is organized in a predominantly Fula community), the number of people available at this point is enough for the meeting. Follow the example of the Alkalo, he gives information. Let's start the meeting. The information spread to the people available at this meeting, will later be disseminated to the other people absent in the meeting.

Fieldworker Malaria Transmission Dynamics
Almost everybody from the compound was at the community meeting. Six people were absent because they were in the bush. These were his son and his junior brothers (…) They went farming. The reason I was present during the meeting was so that I could explain everything and whatever I accept, they will follow.

Interview
Compound head/ farmer

Malaria Transmission Dynamics
People will talk about the MRC study. It is the new agenda. By now it will be on the 'word radio'.

Informal conversation
Compound head/ farmer

Malaria Transmission Dynamics
I do understand the most important of information said during the community sensitization meeting, not everything. (…) I: What is the general reaction of the village on the community meeting? R: People in the village talked about it yesterday. The general reaction is positive. The Alkalo is very happy though they are not educated so they might not understand everything, but the people who came and explained it to them did a very good job. I: Would you like to participate in the study the MRC is about to start? R: Yes, I am the head. I: What do you need to do as head? R: I will help people to better understand the village. I will also explain to those who might misunderstand what the meeting was all about.

Interview
Housewife/ farmer

Malaria Transmission Dynamics
T A B L E 9 Awareness about MRCG study and desire to participate before the individual informed consent procedure (Prinogam) (N=124) the informed consent process. Little is known however about the impact community sensitization has on individual decision-making for trial participation. Our study presents three interrelated key findings and potential challenges to this process.
First, although only 38.7% of all respondents were present at the Prinogam community sensitization meetings, 92.7% of all respondents were aware of the trial when surveyed after the community meeting and before individual screening for study participation and individual informed consent. It looks like the first-hand, accurate information delivered at the community meetings was cascaded as 'lay' information to those other community members who could not attend the sessions.
Second, this information seems to influence the decision to participate in the trial as 91.1% of respondents stated their intention to participate in the study. In stark contrast, respondents who were completely unaware of the trial were less likely to be willing to participate. These findings should be further confirmed, since the study was conducted before the start of the trial's recruitment thus we do not know the actual outcome of consent interviews. However, these ob- T A B L E 1 1 Willingness to participate compared between levels of awareness before individual informed consent procedure (Prinogam) In order to support the government * 1 0.9 *Misunderstanding by respondent: participation in this project did not imply that the government was supported.
certain compounds, the approval of the male head of the compound or household may be key for women and children's research participation.
Prevailing international perspectives on informed consent emphasize the importance of individual decision-making, but in many settings community leaders or family members de facto play at least a minor role in the decision-making process for participation in research. 18

| Study limitations
A limitation of the quantitative study was the non-random sample.

| CONCLUSION
Although not substituting individual consent, community sensitization meetings represent a first step in getting access to the communities. In practice, not everyone may be present during these meetings, but key figures within the compound are more likely to be present and explain the study to compound members, acquaintances and friends afterwards. Therefore, "community meetings" may be seen as an effective tool in providing first-hand, reliable information to communities as the information is cascaded to those who could not attend the sessions.
Further research is needed to assess how the informal spread of information further shapes people's expectations, how the process engages with existing social relations and hierarchies (e.g. local political power structures; permissions of heads of households) and how this influences or changes individual consent. In addition, more research is needed to understand if mechanisms similar to those observed in The Gambia may be expected in research-naïf communities.

ACKNOWLEDGEMENTS
We would like to thank the MRCG fieldworkers Dullo Baldeh, Bintou Jarju, Maisey Bojang, Fatou Y Manneh and Hejinkey Darboe for their assistance with data collection during fieldwork as well as for transcription and translation of recorded interviews and discussions. We also thank the Prinogam and Malaria Transmission Dynamics field teams for good co-operation and assistance with fieldwork arrangements. We are most grateful to all community members for their time and efforts while participating in the study and their confidence in our research.

CONFLICT OF INTEREST
No conflicts declared.