Rapid ethnographic assessment for potential anti-malarial mass drug administration in an outbreak area of Santo Domingo, Dominican Republic

The Dominican Republic (pop. 10.6 million) and Haiti (pop. 10.8 million) share the island of Hispaniola, the only remaining malaria-endemic island in the Caribbean [1, 2]. All locally transmitted cases on Hispaniola are caused by Plasmodium falciparum. Transmission by Anopheles albimanus mosquitoes occurs year-round with minor seasonal peaks typically observed in the Dominican Republic from June-July and January-December. First-line treatment for uncomplicated malaria is chloroquine-primaquine. Both countries have committed to malaria elimination [3].

In the Dominican Republic, recent outbreaks in the capital, Santo Domingo, have signaled a major shift from rural to urban transmission [4, 5]. Historically, malaria was most abundant along the Haiti border and in agricultural areas populated by migrant workers [6]. However a recent study indicates the absence of malaria from such rural areas [5]. The Santo Domingo metropolitan area (pop. 2.9 million) is comprised of the National District and Santo Domingo East, Santo Domingo West, and Santo Domingo North municipalities. Since 2014, this area has accounted for the majority of reported malaria cases nationwide. In 2019, 85% of the 1301 autochthonous cases occurred in the capital area. In comparison, no more than 65 cases annually were reported in 2011–2013 (< 7% of all cases nationally). Concurrently, imported cases from Haiti fell from 565 in 2011 to only two in 2019. Reasons for the increase in cases in Santo Domingo are unclear, but likely involve domestic rural-to-urban migration and impoverished conditions in affected areas.

An initial focus of transmission in the capital was the Los Tres Brazos neighbourhood, an area of 93 km² with an estimated population of more than 200,000 people. Situated at the confluence of the Ozama and Isabela rivers, the focus includes parts of the Santo Domingo East and North municipalities and the Distrito Nacional (Fig. 1). Most of the people in the area live in precarious socioeconomic conditions. Since the start of the outbreak in late 2014/early 2015, the country’s national malaria programme (housed within the country’s vector-borne disease agency; Spanish acronym, CECOVEZ) implemented standard anti-malaria interventions in Los Tres Brazos. These included indoor residual spraying (IRS) and distribution of long-lasting insecticidal nets (LLINs), as well as community outreach and enhanced surveillance by community health volunteers [4]. This resulted in, or at least was associated with, suppressed transmission between 2016 and 2018, with only four cases reported in the foci in 2018. However, cases in Los Tres Brazos rapidly increased again in 2019. Up to the week preceding this study (through 12 October 2019), 348 confirmed cases of malaria were detected in the focus—54% of all cases nationwide.

The resurgent and expanding nature of the malaria epidemic in the capital forced CECOVEZ to consider more aggressive strategies. The World Health Organization (WHO) recommends mass drug administration (MDA) for falciparum malaria in certain situations, such as during malaria epidemics and for transmission interruption in areas approaching elimination [7]. Successful MDA requires high population coverage, usually defined as > 80% of the target population [7]. The need for high population coverage and adherence to potentially multiple rounds of MDA underscore the importance of community engagement (CE) [8, 9]. Community engagement strategies typically involve recruiting and training community members to assist with MDA; educating the population about malaria, MDA, and its purpose; countering rumours; providing incentives; and building institutional trust [7, 9]. Rather than another “magic bullet” in the arsenal of anti-malarial interventions [10], community engagement is increasingly understood as a dynamic social process among all stakeholders involved in health interventions [11].

With these issues in mind, this rapid ethnographic assessment (REA) was conducted in Los Tres Brazos to assess malaria knowledge among community members, document attitudes toward potential MDA for malaria, and to generate a formative understanding of health and health care delivery from the community’s perspective. REA is one of several similar research strategies that include rapid assessments, rapid evaluation methods, rapid feedback evaluations, quick ethnographies, focused ethnographies, and real-time evaluations [12, 13]. These methods attempt to quickly generate in-depth, actionable information about sociocultural dimensions of health or disease in order for health programmes to improve delivery and care [13]. Here, the term rapid ethnographic assessment is preferred because it clearly references both the temporal and social science aspects of this methodology. In general, the approach relies on the usual suite of anthropological techniques, such as key informant interviewing, focus group discussions, and observations to characterize local beliefs, perceptions, or practices regarding certain diseases or health issues and then translate that knowledge into appropriate planning and delivery of health interventions [14]. These methods may also be combined with quantitative surveys [15]. Study participants may be sampled theoretically or purposively (in which recruitment is driven by a priori hypotheses). Unlike quantitative survey sampling, which seeks representative samples to make generalizable claims, sampling in qualitative studies seeks to ensure that the data are “saturated,” or that all conceptual categories to explain the data have been identified and exhausted [16]. To do so, a breadth of key informants with personal knowledge, insights, and experiences about the phenomena of interest are recruited until no further theoretical insights emerge.

Based on the analysis sessions, the findings are organized into the following core domains: (1) malaria knowledge, perceptions of healthcare, and other health concerns; and (2) perceptions of anti-malarial MDA, institutional trust, and health messaging.

Overall, most interview participants (64%) affirmed knowledge of malaria (answered “yes” to the interview question, “Do you know what malaria is?”; Additional file 1). In follow-up questions, participants gave a generally accurate description of the disease as transmitted by mosquitoes; causing fever, headache, or body aches; and preventable through the use of mosquito nets, repellent, fumigation, and/or the elimination of breeding sites (criaderos). In analysis discussions, research team members remarked that although participants connected malaria with mosquitoes, some confused malaria with dengue, another mosquito-transmitted febrile illness common in the Dominican Republic. For example, when asked how malaria is contracted, a 60 year-old woman in Guaricano said: “From mosquitoes that reproduce in clean water,” likely recalling health messages about dengue (transmitted by Aedes mosquitoes that favours small containers, tyres, or natural holes that fill with rain water for breeding).

In fact, participants appeared more familiar with dengue than they did malaria; in analysis discussions, this was said to result from greater exposure to health messaging about dengue than malaria and because of heightened fear of dengue. One assistant said that there are likely political reasons underlying the lack of awareness about malaria. The Dominican Republic is a popular tourist destination, and the country’s political leadership and tourism industry have a vested interest in preventing news about malaria from disseminating beyond transmission areas. Simply put, malaria frightens tourists. There was urgent need to control outbreaks, “before people [in the community] start talking to the press,” one assistant half-joked. All of this was said to contribute to poor dissemination of information about malaria.

Based on interview responses, knowledge about malaria was generally shaped by (1) personal experience or word-of-mouth in past outbreaks, in which a participant and/or his or her family member or neighbour had contracted the disease and dealt with symptoms, care-seeking, and diagnosis first-hand; and (2) health activities led by community health workers (promotores). In general, there was a positive perception of promotores sponsored by CECOVEZ and the Ministry of Health. Promotores are individuals who live within the community and were sometimes said to be the only source of health-related information or initial point-of-care.

Doctors, too, were trusted, although negative experiences in past malaria outbreaks had sown mistrust. For example, some participants recounted how a sick family member or neighbour who had fallen ill was misdiagnosed by a doctor, sent home, and diagnosed later by a health promotora after the patient’s condition had worsened. When malaria was suspected (as opposed to other diseases), people often trusted the CECOVEZ field team and community health workers more than doctors, usually based on past experience or overheard anecdotes in which patients were finally diagnosed by a promotora. One 38-year-old man in Los Tres Brazos said in an interview that, “hospitals don’t cure malaria.” Still, many people advised that in general, they would still seek care at hospitals or clinics, likely because they had few options anyway and because medical doctors were still respected for their knowledge: El que sabe es el medico—“he who knows is the doctor,” said an older man in an interview. Decisions for seeking healthcare tended to rest with female heads-of-households (amas de casa), such as a mother, wife, or grandmother.

Malaria was linked to trash, contamination, and inadequate sanitation. People cited the nearby canals (cañadas) that were clogged with foul water, emitting a bad odour also said to cause disease (akin to miasma). A few people thought that malaria could be transmitted through water itself. Community residents valued cleanliness, mentioning in interviews that prevention of not just malaria but disease in general could be brought about by attention to hygiene (limpieza) of the home and surrounding area. This was in turn linked to environmental contamination in their communities, which surfaced in explanations of disease. “What worries us most is what they will do about this trash here,” said a resident of Las Lila, a particularly marginalized sub-barrio within Los Tres Brazos (Fig. 2). “Sometimes, the trash truck won’t come for [up to] two weeks,” a young man in Sabana Perdida was noted to say in an interview.

The majority (58%) of interview participants said that malaria could not present asymptomatically. “I think that it isn’t possible to not have symptoms, I think [you] must feel something,” said an older man in Los Guandules.

Conducted over seven days, this rapid ethnographic assessment generated contextual understanding of health and healthcare in the Los Tres Brazos focus of Santo Domingo, particularly community perceptions of malaria and MDA as a potential intervention to interrupt transmission and contain an outbreak currently affecting the area. Salient findings and their relevance for MDA planning, implementation and community engagement are summarized in Fig. 3 and Box 1. This work further demonstrates the value of rapid ethnographic assessments to inform community engagement [13, 19] and adds to previous research on motivations and barriers to MDA for malaria on Hispaniola [20] and elsewhere [21‐25].

Greater familiarity with dengue may be due to a few reasons. First, dengue is more widespread than malaria, affecting 29 of the country’s 32 provinces. Second, the year 2019 witnessed a dramatic rise in cases nationwide compared to 2018: up to epidemiological 41 (the week of this study), there were 542 cumulative cases nationally, compared to 35 at the same time point in 2018 [26]. Furthermore, annual deaths from dengue have consistently outnumbered deaths from malaria in the Dominican Republic: in 2019, there were 53 deaths from dengue, while there have been fewer than 10 deaths per year from malaria since 2012 [27]. Taken together, these features of dengue, along with public health sensitization campaigns, may have contributed to more awareness and fear of dengue in the general population.

No participant in the study was familiar with the term “mass drug administration” (administración masiva de medicamentos). This result is not surprising because the only recent anti-malarial MDA campaign in the country was in 2005 in the El Seibo province, located several hours’ drive from the capital. MDA campaigns against lymphatic filariasis have also taken place in several regions of the country, including annual treatment in the nearby Santo Domingo de Guzman neighbourhood of La Ciénaga from 2004 to 2006 [28]. After providing a formal definition of MDA, nearly all agreed that they would participate. Although desirability bias cannot be excluded, inspection of field notes indicated strong themes of solidarity. Given this finding, one way to ensure high participation in MDA would be to implement MDA as a part of a larger effort to improve community health [24, 25, 29, 30]. Campaign sensitization could use salient idioms and phrases used by participants in this study to build on the idea of participation in MDA as a way to care for and take pride in one’s community (see Table 3 above).

This study revealed low levels of institutional trust in current civic associations and the potential danger for political manipulation of health campaigns. In addition, the COVID-19 pandemic has highlighted the growing trend of medical misinformation. “Fake news” about the pandemic circulates on social media in the country driving demand for ineffective therapies such as hydroxychloroquine [31]. Mistrust of political authorities, counter-narratives to official explanations, and medical malpractice and poor oversight seem to be flourishing. These issues must be proactively addressed to ensure high community participation in MDA. One potential way forward to build trust in affected communities is to conduct MDA in conjunction with other services, such as free medical screening for hypertension and diabetes and donations of children’s school materials, chlorine tablets for home water storage, and iron tablets and prenatal vitamins. Such ancillary activities have been shown to improve participation in MDA in other contexts [24, 25, 30] and would help to align MDA with community health concerns. “Beyond health concerns, [people] see solutions to basic problems, like electricity, potable water, and lack of work as important”, wrote one assistant in an analysis essay; another assistant wrote simply that, “They seek to obtain a better way of life.” Because people also seem attuned to issues of sanitation and pollution, MDA could occur alongside trash clean-ups, elimination of mosquito habitats, bed net distribution, and residual spraying. Medical screenings could be staffed, for example, by local medical and nursing students supervised by resident physicians. The presence of medical doctors on site could also reassure people who may be skeptical of taking medicine without a doctor’s prescription. Ancillary activities like these would signal that MDA is part of a larger concern for community health and would address issues raised by participants in this study.

These findings provide substance to enact key recommendations for community engagement strategies highlighted in other studies [9, 25, 32]. In general, these studies emphasize early formative research about the community; strengthening bonds between communities and MDA program staff; being responsive to community needs and concerns; and fostering shared responsibility. These strategies could become operationalized in this context through, for example, conducting more formative research into themes of community solidarity, social networks, and key community figures; holding community-based educative talks (charlas) by trusted promotores to raise awareness about malaria; contextualizing MDA as part of a broader effort to improve community health; and incorporating ancillary medical care alongside MDA activities. Even though 64.0% of participants appeared to know about malaria, it will still be crucial to launch educative campaigns that use clear, non-technical language through diverse media and forums. For example, the vocabulary of malaria as being “hidden” (escondido) or “covered up” (tapado) may be more effective at communicating ideas about asymptomatic infection than biomedical language. Appropriate venues for transmitting information about malaria and MDA could include in-person meetings at community gathering spaces such as churches, youth clubs, colmados, and clubs and even through networks of gang leaders or drug dealers in marginalized areas. Involving women’s groups would also be particularly worthwhile because women had considerable decision-making power over health matters in the home. It is important to recall, too, that in this setting, forms of collective organization and mutual aid go back centuries [33]. For example, the convite is a form of cooperation among rural neighbours who join together to accomplish tasks [34], while lines of credit (fiao) are commonplace at corner store colmados for neighbourhood residents in need [35]. To have greater local resonance and generate trust, MDA activities could be modelled on these longstanding cultural practices of mutual aid and reciprocity.

Other important media to consider for transmitting health messages include digital platforms such as YouTube, Facebook, Instagram, and WhatsApp. Word-of-mouth will be important in spreading health information and countering unsubstantiated rumours. Precarious economic circumstances are widespread throughout the transmission focus; a large number of people may lack radios, televisions, and smartphones (or electricity in general). In these cases, it will be important to develop tailored messaging strategies that circulate through reliable intermediaries and other trusted but less formal social networks. It will be necessary to identify, recruit, and supervise these intermediaries, who could spread messages about MDA and counter rumours. Another potentially promising way to spread information about MDA would be through clustered friendship connections [36], which deserve more research in this setting.

This study was limited by its primary focus on general health and anti-malarial MDA, whereas a traditional malaria knowledge, attitudes, and practices (KAP) survey would shed more light on the spectrum of malaria knowledge, prevention, and treatment in the focus. Therefore, the next step for the research team is to conduct a KAP survey based on a representative sample across a wider segment of the focus. An additional limitation to this study is selection bias. Only individuals who were willing to speak at length about malaria and their communities were purposefully enrolled; those with less interest in MDA, less knowledge about malaria, or harboring mistrust towards health authorities may be under-represented. The definition of MDA did not mention the possibility of side-effects and was given by CECOVEZ-affiliated interviewers, which may have also introduced response bias. The question “Do you know what malaria is?” is also fraught with issues of desirability bias (saying “yes” to a question in biomedical language posed by public health personnel). Furthermore, all three-field assistants were men, which may have influenced rapport with female participants. The shortened timeframe prevented more in-depth training of team members, sampling in other neighbourhoods, obtaining a larger sample, and more time for analysis and reflexivity. Longer-term ethnographic work would provide richer contextual data.

Nonetheless, certain strengths helped address these limitations. First, the team was composed of locally hired assistants, two of which had years of experience in the malaria programme, were familiar with the study site, and had already established rapport with community members. The other assistant was also Dominican and had extensive prior qualitative research experience. Second, the questionnaire was developed at length in team discussions; the team decided that the question “Do you know what malaria is?” would be the most comprehensible to people living in the focus and could be verified by the follow-up questions regarding symptoms and transmission. Third, the built-in analysis sessions allowed the team members to reflect upon their own experiences and observations and interpret emerging themes through shared discussions, which helped identify bias. Fourth, the study included multiple methods to help triangulate findings (structured interviews, informal conversations, personal observations, and team discussions) to confirm or offer alternative explanations. Lastly, one study coordinator (HK) took daily notes as reflexive practice to track sources for bias and how key findings arose in team discussions. Longstanding relationships among all team members and with community members further facilitated team cohesion, lively analysis sessions, and rapport with study participants.

This rapid ethnographic assessment yielded rich contextual information crucial to evaluating the feasibility of potential anti-malarial MDA. The findings reveal pathways to ensure that MDA be conducted in a socially responsible way.