6

Confronting “The Blind People and the Elephant” Metaphor to Bridge the Silos

During the second day of the workshop, the focus shifted from a discussion on understanding the complex dynamics of microbes, noncommunicable diseases (NCDs), and human functioning to examining how to translate that knowledge into convergent actions. Session two featured a panel discussion that explored how to confront “the blind people and the elephant” metaphor to bridge the silos between infectious diseases and NCDs in the move toward convergent action.¹ The panelists spoke about the lens through which they are approaching the convergence and remarked on potential priorities, concerns, and opportunities for addressing the convergence. They discussed how to link and leverage disparate information produced by each field to address the convergence and offered suggestions about potential strategies to accelerate progress that the respective fields have already achieved. This session was moderated by Bridget Kelly, principal consultant of Burke Kelly Consulting. The panelists were Gene Bukhman, director of the Program in Global Noncommunicable Disease and Social Change at Harvard Medical School; Rachel Nugent, vice president of the Chronic Noncommunicable Diseases Global Initiative at RTI International; Nelson Sewankambo, president of the African Medical Schools Association; and Dennis Carroll, director of the Emerging Threats Division at the U.S. Agency for International Development (USAID).

___________________

¹ The metaphor of the blind people and the elephant depicts a group of people who are blind and encounter an elephant for the first time. Each person, touching the elephant from a different vantage point, has a unique interpretation of what the elephant resembles and what its purpose seems to be, thus leading to disagreement on the true nature of the elephant.

EXAMINING THE CHALLENGES AND OPPORTUNITIES OF THE CONVERGENCE

To open the panel discussion, moderator Bridget Kelly, principal consultant of Burke Kelly Consulting, asked each panelist to introduce themselves by sharing which lens they apply to the topic of convergence between infectious diseases and NCDs, in addition to key opportunities to address this intersection.

Extricating the Complexities of Infectious Diseases and Noncommunicable Diseases and Upholding the Equity Perspective

Gene Bukhman, director of the Program in Global Noncommunicable Disease and Social Change at Harvard University, began by describing his background working on tuberculosis (TB) in the former Soviet Union. He cited professional differentiation as one of the reasons that practitioners tend to view themselves as either in the infectious disease camp or in the NCD camp. As both a medical anthropologist as well as a cardiologist, Bukhman was drawn to working on drug-resistant TB because at the time, there had been less medical anthropological work on the issue and he was interested in exploring new stories from an ideological perspective. He has since transitioned to a broader focus of working on NCDs and injuries in populations living in extreme poverty.

The tendency to search for simplicity in public health problems has contributed to the separation between the infectious disease and NCD spaces, suggested Bukhman. He explained that historically, the discourse about health priorities among the world’s poorest populations has found simplicity by identifying a number of largely infectious threats that are the major causes of death among children and young adults. This effort is underpinned by a strong equity imperative (i.e., serving the most vulnerable populations), which has the benefit of operational clarity, Bukhman noted. At the same time, there has been a sense that NCDs are too complicated to address at a global level, where large-scale organizations tend to focus on the top five or six causes of death. He suggested that this approach has shaped the evolution of the NCD space into the inverse of the infectious disease space of simplicity and operational clarity from the risk factor standpoint, particularly in high-income countries.

A separate conversation thus emerged about older populations and the nature of the epidemiology and risk factors associated with ischemic heart disease and stroke. This approach allows for two separate spaces that are defined by their own simplicities, he said, but it does not fully address the complexities within and between the two spaces. For instance, drug-resistant TB is an example of complexity within the major infectious diseases, he

added. The NCD side is complex as well. For example, cardiovascular diseases include much more than ischemic heart disease and stroke; they also include rheumatic heart disease, cardiomyopathies, and congenital heart disease. There is also enormous complexity within cancers. He added that complexity also pervades the other disease categories that constitute NCDs, from congenital conditions to musculoskeletal conditions to neurological conditions.

One potential type of convergence is the move toward developing new ways of dealing with complexity or simplifying the complexity of diseases, said Bukhman. Old models of achieving simplicity through a focus on single diseases or single risk factors are exhausting themselves as shared risk factors become evident, although he said it is still difficult to establish exactly how risk factors are shared between congenital hydrocephalus and ischemic heart disease, for example. This underscores the need for a new science of integration, particularly in global health delivery, that will help to navigate within the convergence, he said.

Bukhman concluded by highlighting an advantage of the general approach within the infectious disease world, which is the equity imperative—priority for the worst off—that is incidentally related to the diseases in some ways. It grants priority to those who are materially disadvantaged, for example, or to those who are most vulnerable in terms of lifetime health, such as children and adolescents. The equity imperative has imbued certain infectious disease movements with clarity and strength; however, this may also have limited efforts to address TB or other diseases that have less of an emphasis on child health, noted Bukhman. These lessons should not be lost on the NCD side, he cautioned. As with infectious diseases, there is enormous heterogeneity within NCDs. Some are childhood diseases, and some kill people later in life; some are highly lethal, while others are disabling. He remarked, “We shouldn’t lose sight of those distinctions or lose sight of the focus on equity that came naturally to us in the infectious disease world.”

Understanding the Economics of the Convergence

Rachel Nugent, vice president of the Chronic Noncommunicable Diseases Global Initiative at RTI International, shared her experiences from working in the NCD space from a global policy and health economics perspective. Health economics seeks to address the questions of value for money and how to improve that value related to health and health care, but what compelled her to work in NCDs was the lack of action and investment relative to the amount of evidence available. Her work centers on analyzing the economic aspects and evaluating the economic implications of NCDs (Davies, 2018). She focuses on how resources are being spent and whether they are being spent in ways that address the problems that people are facing.

It has become clear that increasing numbers of people in low- and middle-income countries are living with NCDs, although many are not even aware of it. Many of those who have had diagnoses have no access to care. In recent years, she added, this problem has only increased in magnitude.

Progress has been made in the NCD space, said Nugent, but catalyzing further action will require engaging with decision makers. Decision makers, of course, are interested in the cost of taking action and the potential value of different avenues of investment. The past decade has seen the emergence of much better evidence about productivity effects and impoverishment effects, for example (Nugent et al., 2018b). We can now establish with greater confidence the cost of a condition, how many people it impoverishes, and the effects of lack of access to medication on disease progression and the person’s productivity (Essue et al., 2017). As efforts to draw attention and resources to NCD needs in low- and middle-income countries were largely unsuccessful, the need to think differently became apparent. A new vertical program or new global fund, she noted, is not seen as the answer. She suggested that ultimately, efforts need to focus on making care more integrated and more patient centered.

As an example, Nugent cited a review of the costs and cost-effectiveness of HIV and NCD integration in Africa (Nugent et al., 2018a), which found the evidence about the economic efficiency of integration to be deficient. It is often assumed that it will be cost saving—or at least cost-effective—to integrate screening and treatment. However, the authors were not able to find examples where it is cost saving or even cost-effective. The additional cost of integrating NCDs into the HIV platform identified in most of the studies ranged from 6 to 30 percent (Nugent et al., 2018a). The study design also included interviews with people working in global policy, donors, and people implementing in-country programs to integrate NCDs into HIV care. Almost all the interviewees thought that it would be cost-effective or cost saving to integrate NCDs in the populations with a need for NCD care (e.g., an identified burden of heart disease, stroke, or potentially cancer). They assumed that the economics would help make the case for integrating, but the evidence was not available. Nugent added that from an economic perspective, it would be helpful to demonstrate whether the assumption that integration is cost-effective is correct and, if so, under what conditions. She cautioned that continuing to operate based on unsupported assumptions is untenable and unaffordable, as has been seen in other areas of health. Many models of integration are emerging with varying results to make sense of, which she is interested in looking into. Finally, Nugent closed her remarks with the proposition that collecting a minimum economic dataset in settings that are working to integrate would be helpful in building the evidence base.

Engaging Decision Makers and Strengthening Health Systems

Nelson Sewankambo, president of the African Medical Schools Association, opened his remarks by sharing that his career began in cardiology but shifted focus to infectious disease when the HIV epidemic struck Uganda in the 1980s. As a medical school dean, a frequent concern was whether the students were being trained in the appropriate way to respond to the needs of the country. With the increasing attention on NCDs in the past 15 years—as well as members of his own family experiencing NCDs—his focus is now a mix of NCDs and infectious diseases.

He reflected on opportunities to spur progress in convergence. First, the implementation of programs at the country and local levels is shaped by how decision makers appreciate and interpret those programs. Therefore, Sewankambo asserted, policy makers and practitioners need to be brought to the table and engaged, because they have the ability to drive programs forward. They need to understand the effect of convergence, rather than feel overburdened by yet another issue to take on. He added that using evidence to inform policy and practice is taking root in many countries—producing evidence that speaks to convergence will help to gain attention and buy-in for moving in a direction that will help policy makers improve the health of their populations.

Furthermore, convergence can be facilitated by developing strong health systems that allow everyone to access quality care at affordable prices, as well as allowing for emergency detection and response capacities in the case of an emerging infectious disease event. He said the strength of Brazil’s health system allowed the Zika virus epidemic to be picked up quickly, for example. Convergence can contribute to universal health coverage as well, particularly in countries with a high risk of infectious diseases, he added. The Ebola virus outbreaks demonstrate the enormity of the threat that an infectious disease can pose to a population’s health. Sewankambo suggested finding ways to leverage this attention toward convergence across the spectrum from infections to NCDs. From the perspective of a medical educator, convergence provides an opportunity to mitigate the dichotomy between infectious diseases and NCDs by highlighting the need for medical education that is cross-cutting, rather than siloed, to more effectively address the health needs of the population.

The Implications of a Collision Between Emerging Infectious Diseases and Noncommunicable Diseases

Dennis Carroll, director of the Emerging Threats Division at USAID, provided comments that reflected the evolution of his own understanding about infectious diseases and the larger ecology in which they reside. He has

spent decades working in the infectious disease space but feels more comfortable framing his work at the nexus of infectious diseases and NCDs. In addition to addressing current threats, work on emerging infectious diseases also examines underlying drivers and attempts to project how those challenges will play out decades in the future, given the various demographic, economic, and ecological changes that are under way. In the process of trying to position this work in a more forward-leaning way, it becomes evident that the dynamics that will potentiate and exacerbate emerging infectious diseases will also contribute to another series of events, through which populations that are already challenged by emerging infectious diseases will become increasingly challenged by NCDs.

This epidemiological collision is significant in Carroll’s work, because the populations that are most vulnerable to emerging infections tend to be the populations living with preexisting NCDs. For example, during the emergence of the H1N1 influenza pandemic in Mexico in 2009, the country had a much higher mortality rate than the rest of the world (Charu et al., 2011). This was largely associated with a population with severe obesity that exacerbated the infection (Dominguez-Cherit et al., 2009). He said it is likely that by 2050, because of the economic population settlement dynamics and consumer-driven emergent middle class in countries in sub-Saharan Africa, for example, there will be increases in the underlying conditions and lifestyle factors that make them hotspots for NCDs, such as diabetes and cardiovascular conditions. Carroll suggested considering what the epidemiological collision will mean for these urbanized populations with increasing prevalence of NCDs compounded by exposure to emerging infectious diseases. In the context of this trajectory, it will be impossible to solve the problem of emerging infectious diseases without thinking about NCDs, he maintained.

Preparing for an epidemic or pandemic is not simply about managing the emerging disease, said Carroll, but about understanding the context in which it is going to emerge. He added that preparedness plans need to account for the likely population risks for both NCDs and emerging diseases. Country-by-country preparedness exercises to manage—or more aptly, to co-manage—future events would be better informed by predictions about likely hotspots for both infectious diseases and NCDs within each country. However, there is a dearth of evidence about NCD hotpots and no existing protocols about how to manage coexisting conditions, such as coronavirus in someone with diabetes. He added that to maximize insight and manage the events that will play out in decades to come, infectious disease projections and forecasting need to be situated on an evidence base about the emergence of NCD hotspots and their convergence with emerging infectious disease hotspots. Then health systems will need to be reoriented to be more adaptable and integrated. For example, one of the first steps in the

diagnostic algorithm for dealing with an infectious disease is to identify any underlying condition. Developing a playbook with clear, clinical guidelines for co-management would help to reduce the effect of infectious disease events, he suggested.

LANGUAGE AND CONCEPTUAL BARRIERS

Following the panelists’ remarks, Kelly observed that the panelists used language in particular ways—by drawing analogies and thinking about how to explain things to people with different perspectives and priorities. She asked the panelists to remark on any language barriers they have encountered and how language may influence people to understand the importance of this convergence.

Bukhman noted a difficulty with the term noncommunicable disease. In common parlance, NCDs are fairly narrowly defined as lifestyle diseases related to factors such as being overweight and lack of physical activity and their associated cardiometabolic problems. The infectious disease world has tended to focus on the world’s poorest billion to frame infectious diseases as a priority. From an equity standpoint, this has created a disconnect in focusing on NCDs because when it is narrowly defined, the burden is underestimated. A broader definition of NCDs—like the definition used by the Global Burden of Disease, the definition used by the World Health Organization in global health estimates, and most of the International Classifications of Diseases (ICD) codes—captures a more diverse, heterogeneous set of diseases. Partners in Health, a nongovernmental organization largely focused on the poor in Africa and Haiti, has benefited from being more specific about setting NCD priorities at a disease-specific level and building on analogies with priorities in the infectious disease world, he said. This can create more commonalities among work focusing on conditions that do not fit neatly within conventional framing of NCDs—for example, congenital diseases, surgical conditions (such as appendicitis or unstrangulated hernias), rheumatic heart disease, or type 1 diabetes—but may be more consistent with the implicit ethical paradigm within which the global health infectious disease world has traditionally operated.

Defining NCDs too narrowly may lead to missed opportunities in national conversations that strive to forge links between infectious diseases and NCDs as conventionally construed, continued Bukhman. The co-occurrence of TB and overweight, or HIV and hypertension and diabetes, are issues that receive a large share of attention, but they are a relatively narrow subset of the larger set of issues at the interface of HIV and a broader definition of NCDs. He suggested that either the NCD category needs to embrace

the full diversity of its constituent conditions or there should be more specific acknowledgment when only a subset of NCDs is being discussed.²

Nugent commented that the concept of convergence has yet to be well defined or agreed upon. For instance, the 2013 report by the Lancet Commission on Investing in Health described the convergence of poorer countries toward the achievements of wealthier countries in terms of maternal and child health and infectious diseases, contending that it is possible for poorer countries to converge on the levels of mortality and burdens that have already been achieved in higher-income countries (Jamison et al., 2013). This is a different type of convergence than the workshop’s focus; therefore, considering both explicit and implicit uses of language in these contexts is helpful, she added. The term NCD originated from a need to identify a term for a community to coalesce ideas, actions, and advocacy around. Nugent clarified that the term NCDs achieved that goal, albeit at some cost. Language can identify and categorize people, thus contributing to the sense of division between the NCD and infectious disease spaces. Though not often openly acknowledged, Nugent noted that the concept of NCDs has come to be associated with wealthy and privileged lifestyles among many in the global health community. These cognitive barriers can influence how these problems are understood, she said, which can hamper efforts to deal with them effectively.

Sewankambo said that the terms infectious diseases and noncommunicable diseases are so deeply engrained in the common vernacular of health professionals and in the medical education system that they are substantial barriers. However, he noted that recent events have provided opportunities to demonstrate that the two entities are related. The human papillomavirus (HPV) vaccine to reduce the risk of cervical cancer and the hepatitis B virus vaccine to reduce the risk of liver cancer both illustrate how an infectious agent can be linked to an NCD in the longer term (Chang, 2011; McClung et al., 2019; Song et al., 2019). Similarly, there is a link between malaria and Burkitt lymphoma, which was prevalent in the children in Uganda and a number of other countries (Legason et al., 2017). Demonstrating these types of potential linkages between infectious diseases and NCDs could lay the groundwork for approaching those two sets of conditions within a single system, he said.

Carroll maintained that the current global health agenda is defined by the epidemiological priorities of the 1980s and 1990s, which have not evolved to keep pace with changing epidemiology. This is reflected in the lack of flexibility to adapt to evolving epidemiology that restricts congres-

___________________

² Jay Varma, senior advisor at the Africa Centres for Disease Control and Prevention, also took issue with the language of NCDs, which he noted takes everything that is not caused by a single microbe and lumps everything left over together.

sional funding. A consequence of the paradigm’s inflexibility, he added, is the lack of a documented evidence base about the current global burden of disease and the cost-effectiveness of combined packages for dealing with the interface between infectious diseases and NCDs. The 1993 World Development Report, for example, provided an evidence base that propelled an extraordinary increase in global health funding around the world (World Bank, 1993). A similar resource could facilitate shared understanding of the dual burden and cost–benefit discussion of strategies to address them. The transition will continue to be difficult without such a resource that maps the larger ecosystem of converging NCDs and infectious diseases, he said. Nugent replied that such an evidence base already exists in the second and third editions of the Disease Control Priorities, which provides an extensive body of evidence on cost-effectiveness, including high-priority packages of care for low-resource countries with cost estimates (Jamison et al., 2018). Carroll responded that the resource would benefit from broader dissemination and outreach to policy makers to bring it to the forefront of discussions.

DISCUSSION

Kelly opened up the floor for discussion. Workshop participants reacted to some of the remarks that the panelists had made and discussed approaches to strengthen health systems to address the convergence.

First, reflecting on the language issue of defining NCDs as lifestyle diseases, Cathryn Nagler, professor of pathology, medicine, and pediatrics at The University of Chicago, noted NCDs that are influenced by changes in the microbiome are not exclusive to high-income populations because they cut across all segments of the population. They may be referred to as lifestyle diseases, but they are not voluntary lifestyle choices; they are societal lifestyle choices. She also highlighted the fact that many of the factors that have led to changes in the composition of the microbiome over the past 30 to 50 years have been recommended by the medical and the societal advisory boards, such as the use of antibiotics and the transition toward processed foods.

Relatedly, on the topic of language and communication, Jay Varma, senior advisor at the Africa Centres for Disease Control and Prevention, emphasized the power of translating a message to policy makers to achieve action, drawing on his experience working in decision making in disease control programs. Lack of an evidence base is not the primary issue—although better understanding of the magnitude and precision of the association between the factor and the outcome is important. The issue is how to translate the scientific understanding that an intervention is likely to have an effect on the outcome in a way that is accepted by policy makers and resonates with their sense of causation. Concepts such as how the environment shapes

and affects health can become vague and abstract for policy makers, he added. The extent to which the scientific mechanism by which environmental and lifestyle factors are causing damage can be articulated clearly—that is, identifying the “enemy”—will determine acceptability of these interventions at a policy level. A powerful potential benefit of this convergence is the opportunity to create messages that are more clear, he suggested.

The discussion transitioned to the topic of strengthening health systems. Miriam Rabkin, associate professor of medicine and epidemiology at the Mailman School of Public Health, Columbia University, suggested that the real elephant in the room is the weakness of health systems. Neither infectious diseases nor NCDs are being treated with a trusted, well-staffed health system that provides high-quality and affordable services. The term health system may not be ideal, but well-trained people need to be in place to build trusted relationships with communities to engage them in care, she added.

Other participants followed up on the challenges and the different ways to strengthen health systems. John Harley, founding director of the Center for Autoimmune Genomics and Etiology at Cincinnati Children’s Hospital Medical Center, suggested that the primary issue is the inadequacy of systems to deliver care—not the lack of knowledge about how to deal with infectious diseases. Similarly, Mahir Rahman, clinical associate from Eden Health, added that health systems would be able to serve patients more effectively if health literacy better empowered patients to be self-advocates and to articulate their needs. Syra Madad, senior director at NYC Health + Hospitals, commented on the blurring line between health care delivery and public health within health systems, although they are separate entities. Health care delivery focuses on individual patients, but policy and guidance are often provided by public health and not well translated at the health care delivery level; this is particularly evident in infectious disease outbreaks, she said.

Tolullah Oni, clinical senior research associate in the MRC Epidemiology Unit at the University of Cambridge, reflected on the need for a balanced system, based on experience in reorienting and reorganizing health systems. In South Africa and other countries in sub-Saharan Africa, many health care systems are designed around curative models to respond to diseases. Introducing primary prevention elements into those systems is challenging because of how those systems are currently functioning. Most health care providers are trained within a system of specialization and overspecialization, which requires vertical systems that allow those specialties to be exercised in delivering care. Although those health care systems have the capacity to treat diseases, she stated they have little capacity for health promotion or behavioral interventions within those systems. Discussions about the importance of behavior rarely extend to how to deal with problems that pertain to behavior—the tendency is to respond with a medical product, to the exclusion of interventions that relate to empowering people to manage

and improve their health. Oni suggested that a combination of responses within a balanced model allowing for both generalist and specialist care would be more effective. Discussions are ongoing in countries in sub-Saharan African to consider how to remodel primary health care such that health promotion and services can be delivered outside the health system as traditionally construed.

Oni raised the question whether the ultimate objective of convergence is (1) converging to treat co-occurring diseases, (2) converging in terms of screening and secondary prevention, or (3) converging in how primary prevention is done. She cautioned against conflating all people with patients, as is often the case in discussions about health system strengthening. It is helpful to think about convergence in terms of future proofing health and health care delivery as well as health proofing the future. “We have to somehow converge in terms of the science of primary prevention and work out how to slowly close the tap, instead of just working out how to plug the leaks in the bucket,” she said.

Dorothy Indyk, associate clinical professor at the Icahn School of Medicine at Mount Sinai, suggested eliminating the dichotomy between prevention and treatment in discussions of convergence across different diseases. Prevention and treatment of HIV/AIDs were first integrated conceptually 15 years ago, she noted. In that continuum of prevention and treatment, the locus shifts into the community and the global realm at the prevention level, then it moves to the individual level and along the continuum toward treatment and care. In that sense, HIV/AIDS work has provided a wealth of insight and knowledge about the convergence of NCDs and HIV—how HIV is linked to comorbidities and NCDs, as well as learning about cancer through HPV. Prevention of biological disasters, for example, could help to bring together work on NCDs and HIV. Indyk suggested that the discussion should be focused on integration to achieve healthy lives in every environment by adapting and optimizing what we have learned across silos in a bottom-up way, rather than limiting the discussion to diseases and NCDs.

CONVERGING TOWARD THE MIDDLE

To wrap up the session, Kelly remarked that intentional work around convergence and integration provides an opportunity to move toward the center and reframe health systems. She asked panelists to describe their optimism about how things might be different moving forward.

Carroll replied that getting to that middle will require looking comprehensively at the entirety of the ecosystems involved to understand problems, not in isolation, but within the context of systems. Doing so, he said, will require funding streams to be aligned toward investing in integrated health systems.

Sewankambo commented that although he is an internist, many people think he is a public health physician because of the breadth of his work to engage communities, families, and patients. It would be helpful to make people beyond the workshop more comfortable with talking about convergence. Public engagement also contributes to helping people understand how they can prevent health challenges at an earlier stage.

Nugent suggested finding concrete ways to bring convergence to upstream factors that drive delivery—for example, engaging with governments and donors to address siloed budget lines and considering ways to address siloed training of practitioners. This focus needs to extend beyond the health system to address broader drivers across multiple sectors. For instance, the activities of different departments and ministries within governments—such as education, sports, agriculture, and transport—can affect prevention. A useful, concrete exercise is to measure the budgetary contributions that different ministries make to prevention of disease. The Organisation for Economic Co-operation and Development has developed a framework for the actions that can be taken by nonhealth ministries and departments to affect disease prevention and health promotion, she added.

Bukhman commented that this work should not lose sight of the opportunity to converge health outcomes for the poorest people in the world by applying what has been possible in some countries at reasonable cost (Jamison et al., 2013). There are disparities across multiple disease entities, including NCDs and injuries that extend beyond a few major conditions. However, he raised the concern that bringing NCDs into discussions about convergence for the poor entails different connotations about development and eradicating absolute poverty in general—in a sense, the “flip side” of convergence. The tendency to frame treatments in terms of epidemics, syndemics, and pandemics may occlude the importance of endemics, which are long-term problems that are difficult to apprehend and measure, said Bukhman.

He explained that the foundation of the ICD is William Farr’s work in England in the mid-nineteenth century who categorized zymotic (fermenting in Greek) diseases as those that signal fermenting, evolving public health epidemics such as cholera that warrant attention, in contrast to NCDs. This distinction remains as the basis for the ICD classification system. He suggested that perhaps the current era is focused on “nonfermenting” diseases and their importance. Reflecting on his years spent working in Rwanda, Bukhman described the country’s relative development in its health system and the improvement in its health statistics as community-level achievements that did not necessitate a functioning health system. However, he predicted that the current focus on addressing NCDs will likely require dealing directly with complexity within health systems.