Battambang

In Battambang Province, the MDA was conducted in Samlout District, part of an area targeted for the elimination of multi-drug-resistant P. falciparum [5]. With a decline in clinical falciparum malaria over the past 15 years, malaria transmission in Battambang is currently unstable [6]. In this border region, clinical and asymptomatic P. falciparum infections are associated with travel to local forests and a large proportion of transmission may occur outside villages [7, 8]. In Samlout District a programme of activities aimed at containing artemisinin-resistance was already ongoing including the recruitment and training of village malaria workers (VMWs); the distribution of effective anti-malarials, rapid diagnostic tests, long-lasting insecticide-treated bed nets (LLINs) and hammock nets to stable and mobile/migrant populations; and engagement with private drug sellers to combat counterfeit and substandard anti-malarials and to enforce the ban on the sale of anti-malarial monotherapies [5].

The study site comprised four villages selected on the basis of prevalence of asymptomatic P. falciparum infections detected in earlier prevalence surveys, their size, accessibility and recent clinical P. falciparum cases detected by village malaria workers in 2013–14 [9]. The selected villages recorded populations of between 239 and 1112 residents and total incidence of clinical malaria (all species) between 27 and 109 cases per 10,000 (2015). The target population was 2366 people. Local activities by the MDA team included, prior to MDA, and then at 3-monthly intervals for 12 months, cross-sectional surveys with venous blood sampling to detect asymptomatic Plasmodium prevalence and record health status. During the MDA, MDA staff also undertook weekly supervision of village malaria workers, collection of treatment records and identification of patients.

Preah Vihear

Perceptions of the MDA and implementing organization

At both sites, villagers were familiar with the aim of the MDA “to eliminate” malaria. They were however often less aware of the intervention rationale—to treat asymptomatic infections. During the community engagement was the first time that most community members had heard messages about ‘drug resistant malaria’ and ‘asymptomatic carriers’. In Battambang, some members of the field team were confused about the science that underpinned the intervention. At both sites, previous treatment guidelines that emphasised diagnosis of malaria before giving anti-malarials prompted some confusion.

During the MDA, MORU became a well-known term. Community members also described generally positive opinions of staff’s efforts at eliminating malaria. Nonetheless, staff interpreted the drop in coverage during round 2 in 2015 as an indication that attitudes towards MORU had changed and that they had begun to see the team as outsiders. In response, representatives from the local health authorities took up more prominent roles in the third round of MDA.

In Preah Vihear, before the intervention, the initial description and notion of MDA was well-received by members of the target community, though it was viewed as a last-minute introduction. In addition to MDA prompting confusion regarding previous messages about overtreatment, MSF staff gained the impression that the non-governmental organization (NGO) was primarily associated with the treatment of symptomatic malaria. Few respondents made the link between MDA and the survey conducted the year before. Although the organization was well known, and their efforts to eliminate malaria positively received, it is possible that the change in approach and the new activities raised questions among villagers.

Financial incentives

Incentives were provided in cash for participants in the Battambang MDA in all three rounds in 2015 and the first round in 2016. Intended to compensate participate for a day’s labour, the incentives were however not readily cited by villagers as a major reason for participation. The absence of financial incentives for two rounds of MDA in 2016 did not negatively impact coverage. No financial incentives were provided in Preah Vihear.

Written informed consent

Amongst villagers, the process of informed consent for MDA was sometimes viewed with suspicion and this raised concerns about the MDA. In Preah Vihear, a few villagers misunderstood the Khmer terms used for surveys and MDA, interpreting it as having connotations of experimentation and that participants would ‘discharge liability’. Learning from the experience of staff working in Preah Vihear, in the Battambang MDA, a more neutral term, “project” was used. Both MDAs required signed consent. In Preah Vihear, for a few villagers, this stoked fears that the signature could be used for political reasons or land grabbing. Such fears were not mentioned in Battambang.

The information sheet that accompanied the informed consent in Preah Vihear provided extensive details of the potential (albeit unlikely) adverse reactions of the anti-malarials. Staff suggested that this contributed to the range of health complaints that villagers reported after MDA (see below). Staff in Battambang were informed of this and adapted their information sheet accordingly to reflect the low likelihood of adverse reactions.

Perceived adverse reactions

In Battambang, MDA was conducted during the rainy season and coincided with seasonal bouts of mild cold and influenza-like illnesses. Some villagers attributed these illnesses to the anti-malarial and this had an impact on coverage in the second round, with an estimated 10% decrease in participation. Although study staff viewed the decline as a challenge, they were able to give an understandable explanation for the minor health complaints that villagers were experiencing at the time of MDA.

In Preah Vihear, MDA was conducted towards the end of the dry season and staff had no such easy explanation for the complaints that villagers experienced at the time of the MDA. In fact, staff, particularly the locally recruited village guides struggled to respond to villagers’ questions about perceived adverse reactions. News of adverse events spread through the villages and staff described them as a major reason why coverage declined in target villages. Based on history and physical examinations, MDA health staff confirmed that almost all of these complaints were unlikely to be related to the MDA.

At both sites, concerns about the adverse events that villagers attributed to the anti-malarial were compounded by several factors. First, many adverse events were seen as requiring medical attention—and in some cases in Battambang, because they did not seek help from the MDA health staff, this was viewed as an additional cost. Second, the adverse events were seen as potentially limiting one’s capability to work and to earn needed income.

In Preah Vihear, the adverse events prompted changes in attitudes towards the MDA. These were linked to rumours suggesting that the anti-malarial was different from that used by the VMWs. Linked to the misunderstanding about the nature of the MDA as experimentation, some villagers viewed it is a dangerous and potentially targeted at the Khmer people. Private practitioners were described as participating in spreading these stories.

The number of villagers who sought assistance at health facilities in Preah Vihear was so large that they were unable to deal with all requests for treatment. This may have been related to the presence of extra MDA nurses and an expatriate physician. Staff explained that participants could report and seek assistance for any health problem until 1 month after the last MDA dose and some villagers associated all symptoms/medical complaints during this period with the MDA. The availability of free healthcare and the presence of an ambulance and expatriate physician may have heightened concerns about the anti-malarial. There were also complaints about the treatment that was offered to palliate adverse events and about the costs of travelling to health facilities for this assistance, even those were reimbursed in both sites at a later stage.

In Battambang, people were concerned about having to purchase intravenous (IV) drug “kits” from private health providers to manage perceived adverse reactions to the study drugs. Although expensive, these kits were popular because of their perceived “powerful energising” effects and because they could be administered by a friend or neighbour (former soldiers who had received some medical training).

At both sites, villagers were also worried about the study because of the widely reported 2014 case of negligent needle practice by a health provider in Battambang Province, which led to multiple HIV infections [13]. Staff in Battambang described this as a challenge during community engagement but it was not described as a reason for non-participation amongst villagers. In Preah Vihear, however, it was mentioned by villagers as a reason not to participate.

Community engagement

Local community engagement activities began around 6 weeks before the start of the MDA in Battambang and 3–4 weeks before in Preah Vihear. For the Battambang team, delays in the ethical approval procedure gave additional time for planning the community engagement (and the opportunity to incorporated lessons from the Preah Vihear). In Preah Vihear, staff—and villagers—were critical of the lack of time available during the community engagement prior to the MDA to provide adequate information. Field staff in Battambang also described being rushed to provide adequate community engagement.

Differences in the number of personnel who were responsible for community engagement explain part of this: in Battambang a team of five (later six) field staff were engaged specifically in the community engagement, and senior members of the study staff also took active roles in community engagement activities in the four villages; in Preah Vihear, two health promotion officers (daily workers, not fixed MSF staff, but trained for this specific task) were charged with delivering community engagement across eight villages (nurses were present in villages only during 3 days). The number of intervention villages also played a role.

Staff involved in the Battambang MDA adapted the community engagement to the local context, according to formative research prior to and during the MDA. This include initial interviews with local community leaders and semi-structured interviews with villagers 1 week after each round of MDA. The semi-independent staff who conducted interviews were able to feedback information to the team who implemented MDA. The presence of the study team in the villagers during 7 days after each round of MDA meant that they were able to respond to complaints and concerns in small gatherings or house-to-house visits.

This capacity to adapt community engagement in Battambang led staff to alter their approach after the decline in participation in round 2 during the 2015 MDA. Prior to round 3, MORU staff took less prominent roles in meetings and rather local health staff from the government-run facilities led the events. Coverage subsequently increased in round 3 and this model based on local health staff was used in the 2016 MDA, during which coverage remained high (despite the removal of incentives in rounds 2 and 3).

In Preah Vihear social science research was conducted retrospectively to understand the reasons for the rapid decrease in participation in the MDA. MDA health staff were present in local facilities during and after the first round of MDA who were able to listen and respond to villagers’ concerns. Because the MDA was halted, community engagement was not adapted. The MSF team had initially assumed that gaining the approval of the village leaders, would be sufficient to ensure that all villagers would agree to participate. The leadership dynamics in each village were however more complex. Other formal and informal leaders (such as teachers, policemen, monks, and representatives of political parties) also influenced other people’s decisions about whether to participate in MDA.

Malaria as a health concern

In Battambang, in spite of the recent declines in clinical cases, and a limited understanding of the disease, malaria was a concern to villagers. This concern was often based on past experiences, for example, malaria was a major cause of ill health when some of the study villages were settled over 20 years ago. People also recognized that a bout of malaria has financial implications, particularly in terms of opportunity costs when a sick person is unable to work. Villagers were therefore generally receptive to malaria prevention interventions and appreciated the effort to ease the burden of disease in their communities.

In Preah Vihear, some villagers also described how they wished to avoid illness because of its financial implications. However, accustomed to the “test and treat” approach to malaria control, most people were more comfortable treating themselves when they were sick than taking an anti-malarial when feeling healthy.

Learning from experience

The delayed start of the MDA in Battambang provided time for community engagement and for dialogue between the MDA teams. The team in Battambang had time to adapt elements of the study design and community engagement based on the experiences in Preah Vihear. Key revisions to the MDA procedures included: changing the term used to describe MDA from a word that could be interpreted as experimentation to a term more aligned with “project”; placing less emphasis on the potential adverse reactions of the anti-malarial used for MDA and no visible emergency health care interventions, such as an ambulance within the target villages. Staff at the Battambang site were also pre-warned of the potential impact of concerns about perceived adverse reactions. Hence, documenting the MDA process, analysing the challenges and communicating the lessons learned is key to maximizing the chances of success of this approach as part of future malaria elimination programmes.

Communicating the risks of adverse reactions

At the time, the anti-malarial used for MDA, DHA–PQP, was the first-line treatment for uncomplicated falciparum malaria in Cambodia. This ACT, like all pharmaceuticals, has the potential to provoke adverse reactions, some of which are very rare but severe. In light of the “experimental” nature of MDA as an approach to malaria elimination, staff administering the ACT were obliged to outline the possibility of adverse reactions. In Preah Vihear, as a result of the requirements by the MSF ethical review board, MDA staff disclosed all possible adverse reactions and this contributed to unwarranted anxieties among villagers. In light of the MSF experience MDA staff in Battambang put less emphasis on the risk of adverse reactions during meetings. Adapted explanations combined with the reduced visibility of ancillary care in Battambang, may have left the villagers less concerned about taking the anti-malarial. Paradoxically, trying to reassure populations in Preah Vihear by installing a visible and easily accessible (available 24 h per day) medical presence may have created fears.

Scale of MDA

Careful consideration for the size of the target population is needed when designing MDA. Because the overarching aim of the MDA was to look at the feasibility of scaling up, the target population was much larger in Preah Vihear: eight intervention villages with a total population of three times that of the four villages in Battambang (of which only half were targeted in MDA each year). This meant that concerns about adverse reactions and other aspects of the MDA that emerged in the early stage were difficult to manage: there were probably not enough staff members present for a long time in villages (they were in health centres) to ease adequately the worries and prevent them spreading to villages where MDA was subsequently conducted. Because staff in Battambang conducted MDA in two villages only, they were able to monitor and quickly respond to villagers’ concerns.

Responsive (and inclusive) community engagement

Formative research enabled staff in Battambang to tailor the community engagement to the local communities—such research has also been highlighted as an important component of community engagement for mass anti-malarial administration in Laos [14]. Moreover, because they were present in target villages in sufficient numbers and accessible to villagers who could voice concerns staff were able to respond to specific worries about perceived adverse reactions. They were also able to adapt the overall community engagement strategy in response to a drop in coverage in round two; staff responded by asking representatives from the local health authority to take more prominent roles in village meetings and coverage in the subsequent round increased [12].

Understanding community (health) concerns

A key element of the formative research is understanding how malaria is perceived in a target community. In Battambang members of target communities were worried about the direct (healthcare, transport) and indirect (opportunity) costs of a bout of malaria. They were therefore receptive to messages about the benefits of eliminating a troublesome illness from their villages. The research also provides the opportunity to gather data on how to adapt messages to local ideas about malaria, for example, what are the terms that resonate with people and how to pitch the benefit of the intervention. Such concerns were less pronounced in Preah Vihear: malaria incidence was low and the disease was not a pressing concern for the target population (even among those who were more at risk because they work in the forest).

Incentives

The impact of financial incentives on uptake is equivocal. On the one hand, in Battambang, their absence did not have a negative impact during two rounds of MDA in 2016. However, complaints about perceived adverse reactions were compounded by the opportunity and direct costs e.g. missing a day of work and having to purchase an “IV kit”. The incentives are likely to mitigate at least the impact of these costs and might therefore reduce the vehemence with which people complain about the MDA.

Local hierarchies and politics

In Preah Vihear, the team hoped that an approval of health authorities would ensure the participation of other community members based on the ubiquitous hierarchies in rural Cambodian life. However, villagers openly challenged the decisions of their leaders. Such questioning might also have been prompted by the position of the local opposition political party. In Myanmar, the fragmentation of a community along political lines also had implications for uptake of the MDA [15].