8

Strengthening Preparedness for Arboviral Diseases

Albert I. Ko of Yale University, the moderator of the workshop’s final session, said that the session’s goal was to “discuss and identify the essential features of an effective, equitable, sustained multilevel response to arboviral threats—essentially, what needs to be done in the future.” To that end, the session had a panel of four internationally recognized experts on arboviral diseases who work across different disciplines and regions of the world. To begin the session, each of the four panelists were asked to describe three priorities that should be included in a global response to arboviral disease. Afterwards, the panelists engaged in a moderated discussion, responding to one another’s lists of priorities, and the last 30 minutes of the session was opened to questions from the audience.

PRIORITIES

The four panelists were Lee Ching Ng, Linda S. Lloyd, Audrey Lenhart of the U.S. Centers for Disease Control and Prevention, and Kariuki Njenga of Washington State University and Kenya Medial Research Institute. Several of the priorities they named shared common themes, such as collaboration and integration, the development and availability of useful tools, community engagement, environmental management, research, and funding.

Collaboration and Integration

The research and development of arboviral control measures tend to be done in isolated silos, and several of the panelists identified getting

outside of these silos and working across disciplines or areas of interest as a priority.

Multiple panelists identified integrating arbovirus efforts with those aimed at other vector-borne threats as a priority. Referring to Wilson’s discussion of the appearance of Anopheles stephensi in Africa, Lenhart noted this as a “really important opportunity where the arbovirus community could potentially leverage some of the investments that are being made in the malaria space by some of these multilateral initiatives.” Because Anopheles stephensi breeds in locations like those preferred by Aedes aegypti, vector-control efforts could be made that controlled both—and thus acted against both malaria and arboviruses such as dengue.

Njenga spoke of integration more broadly. His first priority was integrating key arboviruses into the World Health Organization (WHO) African region’s Integrated Disease Surveillance and Response list. “Almost all the countries follow that list,” he said, and noted that very few arboviruses are currently listed. Ultimately, he said, integrating arboviruses into that system would be easier than creating a separate system to deal with arboviruses.

Ng emphasized establishing intersectoral collaborations as a priority. Singapore has the Interagency Dengue Task Force for Vector Control, which involves different national agencies and even public schools, with vector control being incorporated into the school curriculum. Singapore also has the One Health Coordinating Committee and Working Group, with five agencies that work together to design protocols and risk assessment. “I think intersectoral collaboration goes beyond just different sectors with the typical sectors that we talk about,” she said. It should include, for instance, government agencies working with academia. Collaboration among field operations, laboratories, and clinicians is also valuable.

Tools

Several of the priorities mentioned involved the development and availability of effective tools for the surveillance, control, and treatment of arboviral diseases. Ng mentioned the use of replacement and suppression approaches, such as Wolbachia mosquitoes, as a technology that could make a major difference in Singapore, which in years past has relied on door-to-door home inspections.

Njenga offered two different types of technologies as priorities. The first was developing sustainable, affordable diagnostics for routine testing. Health departments in Africa are not testing most of their samples because the tests are inadequate, he said. Second, he called for developing and stockpiling vaccines, such as the vaccine for Rift Valley fever. There is a livestock vaccine produced in Africa, but it is just a univalent vaccine, so few farmers use it. “If we just made it multivalent so that it would be picked up by the

farmers,” he said, “we could be in a position to be better able to prevent Rift Valley fever,” he said.

Community Engagement

Lloyd centered three of her priority actions in one area: community engagement. “I believe that if we do not have true long-term community engagement, there are no sustainable arbovirus prevention solutions,” she explained, and the building of this long-term community engagement involves meeting a series of goals. The first is building trust. The second is bringing in new allies and resources. Third is creating better communication across stakeholder groups. And the fourth goal is to improve health outcomes as successful projects evolve into lasting collaborations.

Some have charged, Lloyd said, that much of what is said and written about community engagement consists mainly of vague statements and platitudes about the importance of diverse participation. She stated that the shift from such statements to true community engagement occurs by understanding more about why people do or do not do specific behaviors. Sometimes, she said, some behaviors are simply not feasible. “How do you get rid of your tires when you don’t have trash collection or your trash collection service won’t take them?” Other behaviors may be feasible, but they are not acceptable to the people expected to do them. “The standard scrub, rinse, and refill once a week for water storage containers is feasible,” she said, “but it’s not acceptable if you buy every drop of water that’s used in your household, and you can’t empty the container once a week.” Finally, some behaviors may be both feasible and acceptable, but they are not sustainable. As an example, she pointed to illegal dumpsites in public areas. Unless the municipal government supplies the appropriate services to assist, a community cannot be expected to clean up an area repeatedly.

Next, she said, it is important to create effective strategies for building the necessary multisectoral participation to identify and implement long-term solutions that encourage and prioritize the ideas of the stakeholders. Community participation also needs to be made easier through new technologies and the selection of appropriate physical locations.

In summary, she said, community engagement approaches are needed that reflect complex urban environments and meet not just program needs but also community and individual needs. To accomplish that, she identified three priority actions:

1. Integrating community engagement for arbovirus prevention into existing initiatives such as Resilient Cities and the Healthy Cities movements;

2. Ensuring that skilled career staff lead community engagement processes that cut across ministries, organizations, and programs to sustain high levels of trust and partnership among stakeholder groups for solutions to achieve shared outcomes;
3. Funding long-term community engagement processes that focus on problem solving and achieving outcomes valued by stakeholder communities—not just those of vector-control programs.

Environmental Management

Ng described the focus on environmental management in Singapore and her belief that it should be a priority in the control of arboviruses elsewhere. Since its independence, there has been a campaign to keep Singapore clean with a focus on sanitation and hygiene. Today, people who are caught littering are fined, although she said it seems that the littering situation has gotten somewhat worse in recent decades because of the increased use of disposable items.

Environmental management also includes design modification, she said. For instance, people in Singapore high rises used to hang their clothes out to dry on a bamboo pole, but it was discovered that mosquitoes were breeding in the poles and a new design involving drying racks had to be introduced. Similarly, Singapore does not allow gutters on new buildings because they could host mosquito larvae.

Research

One of Lenhart’s priorities was strengthening the evidence base for vector control. While there are some promising new tools available, there are others in development that do not yet have enough evidence accumulated to say for certain what their public health impact is and how best to use them. She called for improving the design of clinical trials of vector control tools for arboviruses. She expressed her encouragement to hear about experiences with various vaccines, particularly the chikungunya vaccine. Although there are no direct efficacy data coming out of these trials, she said, they have surrogates that track efficacy closely enough “to give us the confidence that that vaccine is a tool that can elicit a high degree of efficacy.” Learning from those trials could inform how other vector-control trials are designed so that they shed light on those endpoints.

Perhaps most important in the testing of arbovirus interventions is understanding the impact of combinations of interventions, she said. “None of these things [are] ever going to be deployed in isolation,” she continued, stating how preventive measures like vaccines, reactive interventions like diagnostics and therapeutics, and vector control need to be assessed collec-

tively. Ultimately the goal should be to inform the design of context-driven, locally adapted applications of interventions.

Funding

Lenhart listed funding as another priority area that underlies all others. “We’ve heard about some of these promising new tools that are available, from vector control to vaccines,” Lenhart said. “But then how do we bring those new tools to scale in the places that most urgently need them?” The places where interventions are most urgently needed often have significant local resource constraints. There are opportunities for multilateral initiatives. While there are several initiatives for malaria, such as the Global Fund and the U.S. President’s Malaria Initiative, “there is not anything analogous in the arbovirus space,” she said. There is an opportunity to convince stakeholders about the need, and in addition to the funding of programs, further research funding is also necessary. There are important gaps in needed research on vaccines, diagnostics, and therapeutics, all of which require significant research investment. Funding is needed for clinical trials for the vector-control tools, and funding research will also be important to understand the public health impacts of vector-control efforts. Finally, she said, funding decisions should consider the existing regional inequities. An example of this is the emerging situation of Aedes-borne arboviruses in Africa, where there is significantly less infrastructure for Aedes surveillance and control than elsewhere in the world.

Panel Discussion

Ko began the discussion period by asking each panelist what priorities they found particularly compelling and important to include in future responses to arboviral diseases. Ng identified Lloyd’s emphasis on shaping community behavior as particularly important. She also seconded Lenhart’s observation that “whatever you do, whether it’s vector control or behavior change, it must be context driven; there is no ‘one size fits all.’”

Lloyd highlighted Ng’s point about the importance of multisectoral collaboration, including academia, clinicians, and field operations staff who are often forgotten in designing and implementing health programs. Lloyd also agreed with Lenhart about the importance of programs being context driven, adding the need for community engagement in the funding for context-driven interventions. Lenhart said that Lloyd’s recommendation on community engagement is “absolutely critical” for the implementation of tools and strategies devised for arboviruses. Even with the best intervention or most efficacious tool in the world, she said, it will not matter if the community does not find it feasible, acceptable, and reasonable. Njenga

identified Lenhart’s point about integrating arbovirus programs into other vector-control strategies as particularly important. He also agreed with the others that community engagement will be key to ensuring the success of programs designed to prevent or control arboviral diseases.

Intersectoral Collaborations

Noting the importance of intersectoral collaborations was a theme identified by several panelists. Ko asked Lloyd whether the necessary structures exist for such partnerships and whether One Health is a viable model for such structures. Lloyd responded that it is most efficient to work within existing structures. If a country has One Health or resilient cities programs already in place, it is ideal to begin with these and build upon them.

Lenhart agreed, saying this approach is particularly important if one desires sustainability. “The bottom-up approach is the foundation for any type of sustainable intervention.” Njenga added that over the past decade several changes have taken place in public health that make it easier to integrate with and build on existing systems rather than to start with entire new systems with a top-down approach. Integrating into existing systems that have been established by WHO, U.S. Centers for Disease Control and Prevention (CDC), or others and building in arbovirus surveillance and prevention should be viable in many cases, he said. Ng cautioned that the intersectoral collaboration that has been built up in Singapore may not be so easy to establish in other, larger countries. The conversations that are easy to have in Singapore may be much more difficult in other places.

Vector Control

Noting “broad enthusiasm” among the panelists for vector control being a key component of controlling arboviral diseases, Ko asked how to balance the need for evidence with the desire to move forward quickly and save lives. “We’re really in an unprecedented time where there is a lot of innovation around vector control, especially for the Aedes-borne arboviruses,” Lenhart said. “But the piece that is missing to a certain extent is understanding what works best under what circumstances and where.” A particular technique may be a major success story in one setting but yield much less impressive results somewhere else, and understanding why that is the case is important so policy makers in different countries have the best evidence at their disposal when they are determining where to spend their limited funds for vector control. This is where implementation science is valuable, she continued, and understanding the types of settings where people will be receptive to certain interventions.

Ko followed up by asking about the value of randomized controlled trials (RCT) versus studies with quasi-experimental designs. “RCTs are an incredibly powerful tool for helping us to understand what the public health impact of these tools are,” Lenhard replied, but noted that other trial designs are important and there is much to learn in terms of possibilities with modeling and how this can inform vector-control tools.

Noting that RCTs can be expensive, Ng suggested making it a requirement to prove a method through an RCT could hold new innovations back. She noted the importance of more innovative evaluations for vector-control tools, such as data analytics and synthetic controls.

Lloyd added that RCTs are not amenable to looking at how communities interact with a technology and then what happens over time. “We don’t have variables for how people interact with these technologies,” she said. “There are no variables. We have to create them and validate them, and I think that there are other frameworks that we can use.”

Surveillance

Responding to a question from Ko about surveillance, Ng said it is important to recognize that different types of surveillance are needed in different countries. In the case of dengue, the disease is endemic in some countries and emerging in others, while others are trying to keep dengue out. Each of them will need a different surveillance strategy. Some countries will require comprehensive surveillance, for instance, while others may only need sentinel surveillance in hospitals and health care clinics. Whatever the requirements, those in charge of surveillance should be familiar with the patterns of arbovirus outbreaks in their countries and in nearby areas.

Njenga said that there is a movement in Africa to go from indicator-based surveillance, which WHO has relied on for many years, to event-based surveillance, a shift that is being driven by the U.S. Agency for International Development and CDC. Event-based surveillance, which is being pilot tested in several countries, is community-driven, relying on individuals in the community to report suspected cases. It is intended to spot likely cases earlier than is possible when relying on hospitals to detect and report cases, and it has potential to vastly improve detection in resource-limited areas that cannot afford expensive surveillance systems, Njenga said. What needs improvement is the availability of tests that can detect the presence of arboviruses in suspected cases. Lloyd said that data presentation is critical regardless of the surveillance system. Data must be presented differently depending on the purpose, such as justifying a surveillance program to managers or communicating a risk to a community.

Equity

Inequity can be minimized by working with the existing systems, Njenga said. As an example, he pointed to the Africa CDC, which he said has “completely transformed” outbreak response and advocacy for Africa. It has been so successful, he said, because it was integrated under the preexisting African Union, which facilitated rapid buy-in from African countries and led to reduced inequity. Countries like Singapore are likely to always be better than most in their control of arboviruses, Njenga said, but it is possible to make sure that low- and middle-income countries will not be too far behind in the case of a major epidemic. “Maybe they are one step behind, but not five steps behind, as has happened in the past,” he said.

Lloyd said that one factor in ensuring equity in arbovirus control and prevention will be creating equity in access to water and sanitation. “It’s really hard to ask people to do things when they don’t have clean water, when they don’t have sanitation, when they have dirt roads that collect water.” Lenhart continued along the same lines. “I think inequity is embedded in the fabric of all of this,” she said, “because at the end of the day, these diseases disproportionately affect the poorest in the world.” Context-driven, locally adapted solutions will account for various drivers behind the disease outbreaks and the role that poverty plays. She added that one of the challenges in addressing arboviral diseases is that there are always competing health priorities, and arboviruses do not kill people on the same scale as diseases like malaria. “It’s not to minimize the impact that the arboviruses have,” she said. “It’s just, in realistic terms, that’s what’s being faced.”

Vaccines

Ko asked the panelists about the role of arbovirus vaccines in the future. Lloyd responded that vaccines are another tool that needs to be reviewed on a country level and assessed based on acceptability and feasibility at a local level. Ng said she believes that once a safe, effective vaccine is developed and tested, there will be countries that use it. However, there are questions related to the use of vaccines against arboviruses, such as the role of cross-reactivity. In general, a vaccine designed for a particular virus, such as dengue, will not protect against others, such as Zika and chikungunya. Thus, arbovirus protection will likely require multiple vaccines. “That’s why I think vector control is very important,” she said. “It solves three problems in one go.”

Njenga said that vaccines do have an important role to play and pointed to Rift Valley fever as an example. It is seen mostly in Africa and affects domesticated animals such as cattle but can also jump into humans. Livestock vaccines exist, and if it were possible to get to the point where

livestock were routinely vaccinated, he said, it would reduce the risk of a major epidemic of Rift Valley fever to near zero. The uptake of the vaccine by farmers has so far been relatively low, but there has been an effort to create a multivalent vaccine that would protect livestock against Rift Valley fever along with other diseases that are more concerning to farmers. If that effort succeeds, Njenga said, it will instantly make Rift Valley fever a limited problem. Lenhart said that vaccines are critically important and should be part of an integrated strategy for arboviral disease control, which would also include vector control. Thus, further research and development of vaccines should take place to improve this aspect of the toolbox.

QUESTION-AND-ANSWER PERIOD

Peter Daszak, president of EcoHealth Alliance, opened the question- and-answer period with a question about detecting emerging diseases. He asked why diagnosing febrile illness is so difficult in various countries, as Njenga had described.

Njenga responded that there are many accurate diagnostic tests for arboviruses that are simply not available in low- and middle-income countries. “[This is] an embarrassing statement to make for viruses that we have had for centuries. . . . Why can’t we work to make diagnostics a bit more sustainably and affordably available?” However, he added that some of the most advanced technologies, such as genomics tests and rapid sequencing, are beginning to appear in Africa, thanks to the work done in response to the COVID-19 pandemic. These advancements will make it possible to come up with diagnoses that could not be made before.

An audience member, Randy Nett of the CDC commented on a point that had been made the day before by Erin Staples and Peter Hotez about the lack of uniformity of vector control and vector surveillance in the United States. He noted it will be important to have vector control in every community and make sure that it is a systematic operation with a basic set of abilities and practices. In this respect, Ko commented, the United States could learn from countries in the Global South, particularly Brazil, “where they have zoonotic control centers of the highest quality that can react to rabies, yellow fever, recurrent leptospirosis, everything, in addition to dengue, Zika, and so forth.”

Next, Matthew Zahn, who works in local public health in Orange County, Southern California, made a comment about the difficulties of vector control and prevention in the United States. In his county of 3 million people, which he described as the sixth largest county in the United States, there is a vector-control program that conducts surveillance, but it is up to the public health entity to respond when a vector of potential risk is detected. Unfortunately, he continued, the decision about what to do is

often a political one, and the politicians are likely to fall back on responses that individuals can do—or choose not to do—on their own. He noted public acceptance is a challenge for interventions such as Wolbachia-infected mosquitoes, even if only from a small minority.

A virtual attendee suggested that socioeconomic issues need to be considered as part of any arbovirus strategy, saying that the disease risk can be stratified using these considerations to better determine whether to take preventive and reactive interventions in different locations or contexts. Lenhart agreed with a virtual comment that socioeconomic issues need to be considered as part of an arbovirus strategy and can be used to stratify disease risks. She added that there are close associations between certain sociodemographic factors and risk of arboviral diseases. Traditionally, she added, analyses of disease hotspots have looked predominantly at historic patterns of disease transmission, but incorporating some socioeconomic factors could “perhaps further refine those types of predictive models.”

Concerning prediction, Anne Wilson, a presenter in the previous session, asked, “what are some of the things that we can do to be prepared for anything?” Valerie Paz-Soldán of Tulane University suggested that community engagement could be important as long as effective systems existed for collecting and disseminating data. These systems should not only be user-friendly in terms of collecting data so that everyone can use them, but also should have some sort of dashboard function that easily allows community members to see the status of their community. Too often, she said, data from low- and middle-income countries go to the country’s health ministry or the Pan American Health Organization (PAHO) or a similar organization, “and unless somebody is really curious and looks at the report, they just won’t know what happened with all that information.” Thais dos Santos of PAHO noted that the most effective way to engage with communities is by using their own evidence. As PAHO has seen with some of its own programs, people in communities get motivated by seeing “the actual impact of their own actions.”