Ethical considerations and consent

The study protocol was approved by the U.S. Naval Medical Research Unit No. 6 (Protocol #NAMRU6.2011.0002) Institutional Review Board, which includes Peruvian representation and complies with US Federal and Peruvian regulations governing the protection of human subjects. IRB authorization agreements were established between U.S. Naval Medical Research Unit No. 6, University of California, Davis, and Institute Pasteur. The protocol was reviewed and approved by the Loreto Regional Health Department (LRHD), which oversees health research in Iquitos. Consent without written documentation (verbal) was approved by the NAMRU-6 IRB so that no names were recorded or stored by the investigator.

Study setting

This study took place in Iquitos, Peru, located in the northeastern Amazon basin, where the human population is approaching 400,000 inhabitants [14] and divided into four districts: Maynas, Punchana, Belen, and San Juan Bautista. Detailed descriptions of the city, its Aedes aegypti population, and local DENV transmission have been published elsewhere [15–23]. This region is geographically isolated; it can only be reached by boat or airplane. The main industries in this region are small commercial entreprises, and of extractive (logging, mining) or agricultural nature. Iquitos has experienced rapid urbanization in the past three decades [14], from the neighborhoods around the city center/commercial zones in the districts of Maynas and Punchana to areas on the river (Belen and parts of Punchana) and to the South (San Juan Bautista).

The recruitment neighborhoods described in the next section were located in the more developed and central neighborhoods of Maynas and Punchana, which are relatively homogenous, with a patchwork of households ranging from wood structures with dirt floors to brick and concrete and ceramic floors. The Peruvian Statistics and Information Institute states that about 30% of the urban jungle population of Peru has at least one unmet basic need, 18.2% live in poverty, and 3% of the population in extreme poverty [24]. Evidence of extreme wealth is not observed in Iquitos, and luxury items such as air conditioners are rarely observed. Other indicators for the Maynas and Punchana districts include that 93% of structures are individual houses (row houses with shared walls), 82–38% have corregated metal roofs, 90–97% have electricity, and literacy rates are 88–92% [25].

Study subjects

Purposive sampling, a non-probabilistic sampling method commonly used in qualitative research [26], was used to recruit focus group participants from neighborhoods in the Amazonian city of Iquitos, Peru. We carried out 12 FGDs with a total of 94 residents over a single week during January 2014. To facilitate recruitment and transportation of individuals to the NAMRU-6 conference room, 2–3 residents per block within 5–6 contiguous blocks were recruited in person, door-to-door, 2–3 hours before initiation of each FGD (for a total of 7 FGDs and 49 participants) from three neighborhoods where prospective cohort studies have been ongoing since 2007 [19, 22, 27], and two neighborhoods (for a total of 4 FGDs and 40 participants) where no dengue studies have been conducted by our group since 2007. All five neighborhoods were located in the districts of Maynas and Punchana in the center of the city, representing more developed and economically stable neighborhoods than observed in neighborhoods located in the far north and south and river edges of the city. The neighborhoods without ongoing research studies were either next to or within a few kilometers of the neighborhoods with ongoing research. For context, neighborhoods with ongoing prospective studies were visited by our staff approximately three times per week to identify people with febrile illness. Our experience in Iquitos is that women usually make the health-related decisions for the family, including children, hence we intended to have a larger representation of women when requesting household participation in FGDs. One FGD was conducted with previous mosquito feeding participants (n = 3) and their mothers if the participants were minors (n = 2). At this time, the Asian-American genotype of DENV serotype 2 was circulating and Zika virus had not been detected in the city [28].

Focus group methodology

A Peruvian social scientist with over a decade of work experience conducting FGDs in Iquitos (VAPS) facilitated the FGDs in Spanish. A Spanish-speaking expert in the mosquito feeding procedures (ACM) was present in all of the FGDs to answer technical questions. Two research team members took detailed notes of the discussion and two more assisted in recording ideas and thoughts on large sheets of paper (that all could see) and in role-play exercises. Before each FGD, an IRB-approved consent form was read and participants and parents of minor participants provided verbal consent to participate in the study and be audiotaped.

We developed and applied an FGD guide to ensure we covered the same topics in each FGD. Each FGD began with a brief introduction and an explanation that we wanted to learn more about dengue and how DENV is transmitted from infected people to mosquitoes, followed by a brief description of the direct and indirect mosquito feeding methods. Role-playing exercises were used to simulate an invitation to participate in the project. Initially, we also intended to evaluate a mosquito feeding consent video that we had developed for the pilot project (see description below, S1 Video, Fig 1), but, after our first two FGDs in which the video was shown at the end for discussion and feedback, we decided to show this video immediately after the brief introduction because it became apparent that it was an effective tool for introducing the project and procedures to the group. This also mirrored how the study was presented to potential participants in the field. To assess people’s understanding of the direct mosquito feeding procedures, after the brief discussion and viewing of the consent video, we asked participants to describe the purpose of this study, the procedures, and their initial reactions. To ensure that these hypothetical questions felt real to the FGD participants, one member of our research team (ACM) had mosquitoes feeding on her during the FGD to show them what the mosquitoes in a container looked like, and what her legs looked like post-feeding. We noted the types of questions and discussion among the participants. We wanted to assess how well people understood the purpose and procedures, identify concerns participants had about the process, assess whether they would participate or allow their children to do so, and identify any additional concerns regarding their children’s participation. Groups were asked to state their preferences for providing a venous blood sample to feed to mosquitoes indirectly or feeding the mosquitoes directly on their arms or legs, and discuss the pros and cons of each method. Individuals were asked about how many consecutive days they would be willing to provide both venous blood samples and directly feed mosquitoes.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Key screen shots from mosquito feeding consent video.

(A) Project Title. Pilot study to evaluate the transmission of dengue virus from infected people to healthy mosquitoes. (B) Simple information on dengue. (C) Study objective. We want to know how easily people with dengue can infect mosquitoes. (D) Dengue transmission cycle. (E) Disclaimer. “Your participation is completely voluntary, you decide to participate or not. (F) Participation included a blood sample to findout if you have dengue, feeding mosquitoes with your arm or leg, and given more blood when you feed the mosquitoes. (G) Taking blood samples. (H) Risks. pain, bruise, infection, fainting. (I) Feeding healthy mosquitoes. (J-M) Feeding healthy mosquitoes, without dengue that were grown in the laboratory. (N) Providing itch cream if needed. (O) Disclaimer, (P-S) Feeding mosquitoes in the laboratory. (T-U) Storing the mosquitoes safely. (V) Study benefits. (X) Disclaimer. (Y) Thank you.

https://doi.org/10.1371/journal.pntd.0007090.g001

Regarding the mosquito feeding consent video, our primary objective was to clearly show the mosquito feeding procedures to potential participants. We wanted people to see a full cup of mosquitoes biting an arm and becoming engorged with blood. The video includes the following: (1) the title and purpose of the project, (2) a brief description of the DENV transmission cycle and how one can become infected, (3) an explanation of what is expected from a participant in the project (mosquito feeding and provision of blood samples), (4) the study’s risks and benefits, (5) a visual of mosquito feeding procedures (i.e., mosquitoes feeding on an arm and mosquitoes feeding on blood through an artificial feeder), (6) clear statements that the mosquitoes used in the experiments were laboratory reared and laboratory tested to be free of DENV, and (7) an explanation that after the mosquitoes are fed they are held in a secure laboratory environment.

Experiences of study participants with direct mosquito feeds

Between September 2012 and January 2015, we enrolled 58 DENV-positive subjects from a total of 197 people (50% female) who agreed to participate in direct mosquito feeds if they were identified as having an active DENV infection [34], representing about 70% of the febrile people invited to participate in the study. Of these, 53 subjects (35 males, 18 females) participated in direct mosquito feeds. The remaining five (4 females, 1 male) agreed to only participate in indirect feeds (drawing their blood and artificially feeding it to mosquitoes). During and directly after the procedure, we asked participants how they felt about doing the feeds, whether they wanted to continue, and if they would do it again.

Data management and analysis

After the FGDs, the entire team present during FGDs compiled all notes to produce a detailed report of each session and discussions that took place. Audiotapes were not transcribed, but were used to fill in gaps in notes and to obtain exact quotes. Data was segregated by themes in the focus group guide: (1) comprehension of project, based on misconceptions expressed by participants, (2) questions and concerns about direct and indirect mosquito feeds, (3) willingness to allow children to participate, (4) number of times individuals would be willing to participate in daily indirect and direct mosquito feeds, and (5) preference for participation in their home versus in the laboratory. Codes were developed based on these themes, as well as subthemes, by the two lead investigators (VAPS, ACM). Data was further stratified by participant—those from surveillance and non-surveillance areas. All notes were coded by one member of our research team who was not present during the FGDs, using Dedoose qualitative analysis software, version D.7.5.16. Additional sub-themes that emerged during coding were discussed and added if appropriate, and the person coding returned to previous transcripts to ensure all sub-themes were included. Any questions in the coding process were resolved by discussions between the two lead investigators present at FGDs, along with the person coding. Results are presented based on these main themes.

Participant concerns

Safety.

In all FGDs, except those conducted with previous mosquito feeding participants, a principal concern was the safety of the study (Table 2). This concern was primarily about the infectious status of the mosquitoes used in the experiments. Participants asked if the mosquitoes were “healthy”, “infectious”, “sick”, or “infected”. Some expressed fears of contracting dengue through the mosquitoes and wanted to know how researchers could verify that the mosquitoes used in the experiments were safe or “clean” and, in one case, how researchers could “guarantee that the mosquitoes that fed on infected people would not go on to infect more people.” These were important concerns expressed in 9 of 11 FGDs. In early FGDs, the number of people expressing this concern was higher than in later FGDs, likely because the consent video was played at the end of earlier FGDs. There was at least one participant in 9 of 11 FGD who asked if the mosquitoes were infected. In one FGD, participants suggested they would be more convinced if the research team fed the mosquitoes on themselves. One team member (ACM) demonstrated the feeding on herself during each FGD, and this led to a discussion about how we could know that the mosquitoes were clean. Other participants, especially those who understood the message that mosquitoes were clean and came from the laboratory expressed their faith in the research team, stating “we trust you,” “take you at your word,” “it would be crazy to think you were doing something bad,” and similar sentiments. Overall, the safety concern dissipated with more information and better comprehension of the study design.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. How key misconceptions identified in focus group discussions translated to changes in research protocol for direct mosquito feeding experiments carried out on DENV-infected people.

https://doi.org/10.1371/journal.pntd.0007090.t002

A key idea that needed reinforcement multiple times was that mosquito feeds would only be carried out on people who had laboratory confirmation of a DENV infection; i.e., only on people who were already infected with DENV. Participants asked questions like “so the person would already have dengue?” or “can it be anybody or [only] someone who has dengue?” Once participants understood that we were studying DENV in people with an existing infection and not trying to infect people with DENV in order to study the disease, most individuals expressed relief that we were not conducting feeds on uninfected individuals. This helped us understand points that needed to be very clearly stated to potential participants. Another important point that required clarification was the use of different mosquitoes (a new cup of mosquitoes) for each participant; i.e., the same cup of mosquitoes would not be used again on another person. Participants in one FGD suggested that we add a statement in our consent process that mosquitoes are used only one time for feeding experiments and then are confined to the laboratory until they are killed and tested for DENV.

One group asked to see where we held the mosquitoes and wanted to know more details on how we knew they were clean. We showed them our insectary facility. ACM explained the 3-step process used to rear mosquitoes used in feeding experiments and how we determined they were not infected with DENV. First, eggs were collected from houses in Iquitos, brought to the laboratory, hatched, and raised to adulthood; these are the grandparents. Eggs from the grandparents were collected, hatched, and raised to adulthood; these are parents of the mosquitoes that feed on people. Third, eggs from the parents were collected, hatched, and raised to adulthood; these are the mosquitoes used in feeding experiments. We explained that all “mothers” were all tested for DENV to make sure they were uninfected or clean. Because the experimental mosquitoes could not have been exposed to the virus outside of the laboratory and could not become infected from their uninfected mothers, the experimental mosquitoes were free of DENV infection. The FG participants that we reviewed this process with recommended that we make a separate video and/or pamphlet explaining details of rearing clean mosquitoes for experimental feeds.

Comprehension of study purpose and mosquito feeding procedures

We determined that the most effective way to communicate the purpose and procedures of the mosquito feeding experiments was to early in the FGD present the video (Fig 1, S1 Video) used in the informed consent process. We decided that the video should be preceded by a brief explanation of its content and why it was important. Based on the types of questions answered after the video, it was clear that many participants required further explanation, interaction, and reinforcement of key points by the research team.

To test comprehension, we asked participants “why are we doing the study?” Initial responses varied widely, but the most common responses indicated that participants captured two key concepts: (1) we wanted to know more about dengue, and (2) we were trying to understand what would happen to the mosquitoes after they fed on an infected person. Participants used phrases like “you want to see what kind of reaction the mosquitoes will have” or “[you want to see] if we can give the mosquitoes dengue.” Although it was clear that participants in many cases tried to repeat messages from the consent video, their focus on transmission from humans to mosquitoes demonstrated an understanding of why feeding mosquitoes blood, either directly or indirectly, was an essential part of the project. Comprehension of this key message was incomplete initially, with some participants expressing interesting misconceptions (see below and Table 3). In most groups, however, individuals who grasped the concept that we were interested in what happened to mosquitoes, rather than what happened to human participants, provided explanations to others in the group who did not understand this concept. This method of peer-to-peer explanation helped researchers assess how well some participants had understood, as well as allowed us to document how the concept was expressed among the participants themselves so we could use their language in information materials developed in the future.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Assessment of comprehension of participants in 11 focus group discussions about the reason the mosquito feeding studies are being carried out and to verify that the objectives stated in the video were clear.

https://doi.org/10.1371/journal.pntd.0007090.t003

Motivation to participate in direct mosquito feeding experiments

When asked what would motivate them to participate in a study like ours, participants felt it was important to better understand dengue and this would in turn help their community. This seemed to be a satisfactory reason for most to participate. One participant seemed taken aback when probed about her willingness, she replied “why wouldn’t I participate!” As described previously, once the FGD participants had their initial concerns and questions addressed and understood the scientific objectives of the study, enthusiasm to participate increased.

Clinical attention that would be received during the study was seen as valuable. FGD participants liked the idea that they would be monitored by a doctor and our research team until they were dengue free. There was interest in knowing how their dengue was progressing and knowing if they were recovering. For example, some expressed interest in knowing if they had anemia, and those who had heard about platelet counts wanted to know if that kind of information would be provided. We clarified that clinical monitoring was not dependent on participating in mosquito feeds or intensive blood draws, because all participants in our studies get medical attention. Most FGD members continued to express willingness to participate. A related motivation was obtaining a dengue diagnosis.

Before we clarified that for our future experiments we would need to directly feed mosquitoes and take tubes of blood, more participants thought feeding mosquitoes directly would more desirable than providing blood samples. It was seen as less painful and would not require as much blood. Some participants also expressed curiosity, stating, “I want to know how it feels to feed the mosquitoes”, or “I would want to see how many mosquitoes got infected if I were sick”.

Preference for feeding mosquitoes or providing blood samples

We implemented a variety of exercises to discern people’s attitudes about providing blood samples and feeding mosquitoes, and their perceived advantages and disadvantages for each. These questions were posed after the FG facilitator (VAPS) felt that the participants understood the procedures would only be conducted on DENV-infected people and that the mosquitoes used in the experiments were clean. In the initial FGDs, we asked people which of the two options, direct mosquito feeding or blood collection, they would prefer, followed by an explanation that we would need both and a question about their willingness to provide both. In later FGDs, we started by saying we needed people to feed mosquitoes and give blood samples. We asked whether they would be willing to do both, and the pros and cons of each. About half of participants said they would be willing to do both, whereas a little over a quarter said they would only be willing to feed mosquitoes directly and a little under a quarter said they would only be willing to provide venous blood samples for indirect feeding.

Participation of children in direct mosquito feeding experiments

Most parents stated that they were comfortable allowing their children to participate in direct feeding if the child could decide. There was some discussion among participants about the age at which children would understand the procedures, but parents felt that, as long as the child’s choice was respected, they would allow their participation. At the same time, many parents were skeptical that their child would choose to participate or be able to sit still throughout the process, with a few FGD participants mentioning how the sight of a nurse caused their children to cry or run away. In contrast, a few parents said that their child would want to participate because they were “brave” or that it would be interesting to them. A few people said they would want to experience a feed before allowing their child to participate: “I want to know what it feels like before giving authorization for my child.” Some parents wanted to ensure that less blood would be taken from children than adults, “If they take my blood for three days, then the child should only do two days….” As with adults, some children were recognized as sensitive to mosquito bites; parents of these children would not allow them to participate.

Participation in multiple direct mosquito feeds and providing multiple blood samples

We used a variety of strategies to probe participants on their willingness to participate in direct mosquito feeding and provide blood samples multiple times. These strategies included asking people to raise their hands and playing games with individuals jumping to the “yes” or “no” side of a line on the floor based on their responses. Responses ranged from none to as many times as necessary (explained by the facilitator as every day they had a fever). Those who answered as often as necessary qualified their responses by stating they would participate “as long as I still have dengue,” “until my platelets are okay,” or “until my hemoglobin is okay.” There was a strong connection between willingness to participate and the clinical follow up that would accompany study participation. About 65% of participants who indicated a willingness to participate said they would be willing to repeat the procedures at least three times. Others were willing to have samples taken every other day or at the beginning and end of their illness.

Acceptability of direct mosquito feeds at home

We asked participants their opinions about conducting direct mosquito feeds in their own homes versus a laboratory setting to determine whether people had safety concerns about performing the experiments in their communities. Overwhelmingly, people preferred their homes. Everyone expressed a desire to participate in the fastest and most convenient procedure. One FGD participant stated a preference for home feeds “so that my family sees this.” The few individuals who expressed a preference for participating in the procedures in the laboratory gave the following reasons: “because it is safer and I would be afraid if mosquitoes escaped, my family could get sick,” “because people in my family would gossip,” “I would have more confidence, it is more credible…,” and “so I can get more information.” In one FGD, the issue of prying neighbors and their possible negative reactions to mosquito feedings was raised.

Incentives for study participation

Although we did not include questions about incentives for providing blood samples or participating in direct mosquito feeds in our original FG guide, the issue of incentives emerged during one FG discussion in a group where since 2007, participants received small thank you gifts (i.e., powdered milk, vitamins) after providing blood samples. We asked participants what they thought about incentives and whether they should be given for participation in direct mosquito feeds. One participant stated it would not motivate her, but that “some people would do the mosquito feeding for money or necessity.” We then asked the group directly whether we should provide incentives, and the group answered, definitively and in unison, “No!” Although none of these participants suggested we should stop providing milk and vitamins for our ongoing studies, there were a variety of strong responses that indicated incentives were not the reason they would participate and that they often had to set their gossiping neighbors straight about this.

Perceptions from participants in mosquito feeding experiments

Conclusions

The impact of dengue in Iquitos is large and far-reaching, making people willing collaborators in studies to reduce the local burden of the disease. This, combined with the normalcy of mosquito bites in Iquitos, made direct mosquito feeding experiments, which might seem peculiar to some, acceptable to the majority of our FGD participants, and for parents, also an acceptable experiment to be carried out among children who could assent to the process. Taken in context with the response to direct feeding experiments [34], in Iquitos, when conducted by experienced researchers using a DENV-free generation of mosquitoes raised in a secure insectary, this is a low-risk, non-invasive procedure that is experimentally superior to artificial mosquito feeding methods. The use of formative research, to identify and develop effective communication strategies for “unusual” procedures provides and effective path to opening up new research procedures to communities where they might be appropriate.

Copyright statement

Author Robert D. Hontz is a U.S. military service member. This work was prepared as part of his official duties. Title 17 U.S.C. § 105 provides that ‘Copyright protection under this title is not available for any work of the United States Government’. Title 17 U.S.C. § 101 defines a U.S. Government work as a work prepared by a military service members or employees of the U.S. Government as part of those persons’ official duties.

Disclaimer

The views expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, nor the U.S. Government.