Summary

The challenge of cancer control starts with the complex nature of cancers themselves. Throughout history, few other diseases have generated the level of social, scientific, and political discourse or have had the degree of cultural significance as cancers. A collective in the truest sense of the word, “cancer” is a clustering of different diseases that afflict individuals in different ways. In the early 1970s, cancer was still largely thought of as a single target, albeit one that affected different parts of the body. Now, it is well known that cancer is a vast and evolving multitude of individual diseases with different biological mechanisms and different responses to treatments, but with a single overarching characteristic in common—the unchecked proliferation of the body’s own cells. Cancers can occur in many human tissues and organs, and there may be many different subtypes that can be identified based on molecular abnormalities, yielding potentially hundreds of different types of cancers. Even what might seem to be a single cancer type—a cancerous lump in the breast, for example—can come in different versions that respond differently to a given treatment.

The burdens of cancers are also broad and diverse, from the physical, financial, and psychological tolls it imposes on individuals to the costs it inflicts on the nation’s clinical care and public health systems. Decades of concerted efforts to understand and eliminate cancers, often referred to as the “war on cancer,” have produced some significant advances in prevention (e.g., tobacco cessation and vaccines for hepatitis B and human papillomavirus), early detection (e.g., colonoscopy and cervical cancer screening), and treatment (e.g., targeted and combination therapies), but

the burden of cancers is still substantial and growing as the population ages. Although age-adjusted cancer mortality has been steadily declining over the past three decades, about 600,000 people in the United States died from cancer in 2018, and about 1.7 million people received a new diagnosis of cancer. Moreover, significant disparities in cancer incidence and outcome persist across different populations.

The World Health Organization adopted a resolution in 2005 urging the member states to develop and reinforce comprehensive cancer control programs and evaluate their impact, and many countries now have a national cancer control plan. In the United States, however, cancer control efforts have evolved over time without a unifying national plan or centralized guidance. Numerous federal agencies have diverse roles in cancer control, but there is little cross-agency coordination, and each state and territory develops its own cancer control plan, with no overarching strategy or guiding vision of how an ideal cancer control system should operate or perform. Thus, the study sponsors (the American Cancer Society, the Centers for Disease Control and Prevention, and the National Cancer Institute) asked the National Academies of Sciences, Engineering, and Medicine to develop a national strategy for cancer control (see Box S-1). In response to that charge, a committee of independent experts appointed by the National Academies developed a set of recommendations that define the key principles, attributes, methods, and tools needed to achieve the goal of implementing an effective national cancer control plan. In developing these recommendations, the committee reviewed literature on the history and current status of cancer control efforts in the United States and globally. In addition, the committee held two public sessions with sponsors and various stakeholders. The public sessions featured presentations and discussions focused on two overarching questions: “What have we learned in the past decade?” and “What should we be doing differently?” These meetings provided an opportunity for the committee to seek input from a broad range of experts in cancer control in the United States.

THE SCOPE OF CANCER CONTROL

A remarkable number of analyses have already been conducted on the subject of cancer control (perhaps more than for any other disease), with several dozen reports and proceedings issued by the National Academies alone over the past several decades. A National Academies report declared in 2013 that cancer care in the United States was in “crisis.” That statement is no less true today and can be generalized to the full spectrum of cancer control efforts. Hampered by poorly integrated resources, uncoordinated activities and conflicting interests and incentives, the current

cancer control system is underperforming in curbing the burden associated with cancers.

The causes and effects of cancers are complex, and addressing that complexity requires efforts across the continuum of cancer care, starting from basic risk awareness through the processes of cancer prevention, detection, diagnosis, and treatment, as well as palliative care, survivorship care, and hospice care, and all the supporting services linked to these efforts. Moreover, cancer control is affected in various ways by the environment, technologies, economics, policies, research quality, and ethics surrounding or transcending its more traditional aspects. These additional factors include such things as education, food quality and availability, and policies related to housing and urban development. Obviously, there will be no single solution that can succeed across a large percentage of cancers and populations. Therefore, the term “cancer control” as conceived and used throughout this report refers to a much broader range of actions than most people appreciate or practice; it comprises a variety of strategies and tactics aimed at helping people at risk for or diagnosed with cancers

in various ways that can extend beyond the traditional notions of cancer prevention or treatment. This report also advances a strategic vision for transforming cancer control that would require a much broader alliance among federal agencies, state governments, and key participants in the for-profit, nonprofit, and other sectors, including the technology industry, with its growing investments in population health.

A SYSTEMS APPROACH TO ADDRESS THE COMPLEXITY OF CANCER CONTROL

The ability to systematically collect and analyze large volumes of data has advanced rapidly in recent years, and this has generated novel approaches to continuous, systematic learning and quality improvement in health care. For example, the learning health care system model strives to enable evidence-informed transformations by continually collecting and using data to systematically integrate new knowledge into care delivery processes and to improve outcomes and motivate greater collaboration among all participants. Other systems frameworks have also been used to assess and improve certain aspects of cancer control efforts across the continuum. A socioecological model, for example, has been used to examine the factors contributing to cancer disparities in communities with low-income residents. Although these models can be useful in understanding and improving certain aspects of cancer control, they are unable to holistically view cancer control efforts to obtain an overall perspective on the collective behavior of the numerous participants in the ecosystem.

To overcome the limitations of the current systems-based approaches to cancer control, the committee approached cancer control as a system of systems, with a focus on the concept of a “complex adaptive system.” In short, a complex adaptive system is a system consisting of individual entities that act and interact with one another to advance their own “interests,” modifying their behavior in response to what is happening in the rest of the system. The behavior of a complex adaptive system cannot be understood simply by examining its individual parts in isolation; instead, the overall behavior is a product of the way that the individual components influence one another. The hallmark of complex adaptive systems is that behaviors emerge that could not have been predicted by understanding the behaviors of the individual components. Examples of complex adaptive systems include not only ecosystems and living organisms but also national economies, transportation systems, and population health.

Tools from complexity science and systems engineering have been applied to systems such as manufacturing, banking, air traffic control, weather prediction, homeland security, and the Internet, to name a few. Like the cancer control system, most of these systems developed over

time with no overarching “master plan” and with no one entity in charge, perhaps with many of the same practical challenges as in cancer control, such as how to integrate resources and capabilities and how to coordinate different components that are generally pursuing their own goals and interests. In the United States, multiple federal agencies are involved in cancer control in addition to those principally focused on health promotion, disease control, and medical benefits.

Systems engineering tools make it possible to analyze, understand, and predict the behavior of complex systems through the study of a system’s components and how the interactions of those components produce the system’s behavior. A detailed analysis is generally the first step in understanding a complex system, and it often involves creating models and simulations followed by rigorous testing to see whether the system’s behavior under different situations can be reproduced. Once such a simulation has been constructed and tested, it can be used to test how the system will respond to various stimuli and changes, which in turn makes it possible to learn how to guide—not “command and control”—the system to a certain degree.

Although a search of the literature has uncovered no suggestions for using systems engineering approaches to understand the total system of cancer control, as this report is proposing, there have been a number of ideas and initial efforts related to the use of systems engineering concepts to understand individual components of cancer control. Published papers have discussed applying systems engineering techniques to cancer drug delivery, cancer survivorship, clinical care and patient safety, and efforts to reduce disparities in cancer outcomes. These precedents could inform a broader systems engineering approach to integrate the various resources and efforts currently in use for the nation’s cancer control system, as can be observed through the varied work of at least 13 different federal agencies and numerous other participants in the for-profit and nonprofit sectors.

GUIDING THE TRANSFORMATION OF CANCER CONTROL

This report offers 10 conclusions supported by 25 findings, all based on the overarching message that overcoming the current narrow and uncoordinated approaches that significantly constrain progress and effectiveness across the segments of the cancer control continuum is an imperative. One of those conclusions emphasizes that cancer control needs to be “recognized and approached in practice as a complex adaptive system whose elements are interactive and influential at multiple levels of society, starting with the individual. This change in mind-set is

essential to recognize, reduce, and mitigate risks and make significant progress in diminishing the cancer burden in the United States, a situation challenged by aging and other demographic factors with no apparent blunting of costs across cancer control activities.” The estimates of how much cancer control efforts cost vary, and it becomes a particularly daunting task to arrive at aggregate costs if one considers the complex adaptive nature of the disease and the efforts to control it. However, the total volume of expenditures attributable to, or associated with, cancer in the United States is estimated to be nearly $600 billion annually, and that figure will only increase with escalating cancer incidence due to an aging society and other factors, including behavioral factors. Therefore, a renewed vision to guide the development of new and more effective national approaches to cancer control is essential.

No single volume can issue detailed analyses and be comprehensive on every aspect of cancer control, and this report is no different. Indeed, this report has a different vision and ambition; it provides a higher level view on the progress made and yet to be made in cancer control and on what is still unclear about the various cancer control interventions and policy strategies. This report does not supply a construction blueprint that may be relevant only to one particular time, entity, or context because another conclusion of this report is: “The design of a single top-down, static blueprint for cancer control programs and operations in the United States is currently neither realistic nor productive. Instead, greater effectiveness in cancer control requires centrally available customizable planning tools that are useful across contexts and that can actively support performance monitoring and accountability reviews. Dynamic data feeds, computational and other capabilities, and interactive visual analytics will be required for the supporting systems analyses.”

The necessity of this broader view is captured in another conclusion: “The current processes and systems of cancer control are at best reactive to circumstances. A proactive and progressive planning system for cancer control policies and operations would necessitate a learning mind-set, from individuals to institutions, focused on periodically determining what activities should be initiated, expanded, or terminated, as well as critically analyzing the trade-offs and tracking the consequences of related decisions.”

The operational strategy recommended in this report will invariably require trade-offs, continuous learning, and adaptation as well as a diligent, accountable, and periodic review of initiatives and strategies going forward. Such a discipline might well be a national imperative in order to progressively tackle the wide-ranging effects of cancers. This report argues that the best chance for transforming the U.S. cancer control system is to apply such a systems engineering approach, and it sketches out

what might be involved in such an approach, providing three interlinked recommendations.

RECOMMENDATION A: A U.S. National Cancer Control Plan should principally ensure resource integration and operational coordination across the various components of the cancer control system and should actively do the following:

1. Improve, where feasible, effective, and affordable, the availability of preventive, screening, diagnostic, and therapeutic interventions. Encourage timely palliative care, hospice care, survivorship services, and related social services according to the preferences and values of patients and their families.
2. Leverage the advances in and apply “multi-omic” diagnostics to improve therapies and better understand their scientific, clinical, and economic impacts, including their role in creating additional new prospects for cancer control and overall cost reduction.
3. Integrate the use of social, behavioral, and other information made possible by the convergence of communication, social media, cognitive, financial, and sensor technologies as well as electronic health records, cancer registries, and insurance claims to establish large-scale interoperable data sources.
4. Use cloud computing, machine learning, and artificial intelligence tools for continuous analytics, rapid reporting of trends and patterns, and improved forecasting and performance reviews. Evaluate emerging data-intensive technologies not only for their utility in advancing health and economic parameters but also regarding their ability to protect individual privacy and the security of data systems.
5. Apply the tools of complex systems analyses for assessing the “value” of cancer control interventions, establishing robust policy and incentive assessments to guide the development and commercialization of products and services, developing new financing and payment mechanisms that alleviate overall cost burden, and aiding individual patients and their families in making informed decisions about cancer care.
6. Minimize the waste and harm stemming from disparate clinical practices, interventions lacking evidence of effectiveness, and conflicting clinical practice guidelines.
7. Track and monitor financial links, incentives, and disincentives throughout the processes and systems of cancer control and rigorously require conflict-of-interest disclosures across cancer care, research, and patient advocacy activities.

8. Expand and support reproducibility strategies for developing reliable evidence in cancer control from biomedical, clinical, public health, and social science research.
9. Discourage direct-to-consumer marketing and advertising of clinical products and services from companies, medical centers, intermediary firms, and other organizations by terminating the tax deductibility of these business expenses. Furthermore, tighten and enforce rules to particularly curb promotional tactics and strategies that are likely to mislead patients about the benefits of products and care services not based on strong evidence.
10. Launch and expand public engagement, literacy, and outreach activities, starting with K–12 curriculums and through technology platforms, to broaden the understanding of cancer prevention as an integral component of a healthy life course.

Historically, cancer control efforts in the United States have prominently involved the federal government—featuring directions from the U.S. Congress or the executive branch—in launching new or expanded national initiatives. Coordinating a wide range of federal agencies active in cancer control efforts could require congressional action if the participating agencies lack a legislative authority, in which case it is urged that the U.S. Congress provide the direction to implement the following recommendations.

RECOMMENDATION B: A U.S. National Cancer Control Plan should be led by the Department of Health and Human Services in cooperation with the Office of Management and Budget, Department of Education, Environmental Protection Agency, Department of Defense, Department of Veterans Affairs, Department of Housing and Urban Development, Department of Agriculture, Social Security Administration, Department of Labor, Department of Commerce, Office of Personnel Management, Equal Employment Opportunity Commission, and Department of the Treasury. The Government Accountability Office should periodically review and report to the relevant congressional committees about the achievement of goals specified in the plan.

A national cancer control plan will need to include all these federal participants, as well as ongoing participation from state and local governments and key participants in the for-profit and nonprofit sectors to undertake a comprehensive review of diverse and shifting needs and an integration of available resources and capabilities. Periodic performance

review and annual reporting, with a rigorous comprehensive review every 3–4 years, similar to the congressionally mandated assessments in other areas, would be essential for both improved accounting and accountability in cancer control. While this extensive level of cooperation may seem daunting, there is precedent for such an approach. For example, the U.S. Global Change Research Program, the NextGen air control system, biodefense initiatives, and intelligence community operations all require resource integration, joint monitoring, and diligent performance review across many different agencies, particularly involving industrial partnership. And, indeed, the iconic Apollo “moon shot” that has since inspired many activities of cancer control was a successful demonstration of more than 20 different government agencies cooperating under a congressional mandate. The ultimate success or failure of a national cancer control plan will depend on gaining a functional understanding of the nation’s cancer control system and being able to predict how it responds to various interests and pressures. Therefore,

RECOMMENDATION C: To support a U.S. National Cancer Control Plan, the Department of Health and Human Services and the federal partner agencies should fund and support an independent organization—or a consortium—with principal competencies in systems engineering, industrial design, software development, and information and visual analytics to prototype and develop a publicly available, interactive, and evolvable planning and monitoring tool.

Moreover,
C-1: Periodic consultations with key participants from state and local governments, and for-profit and nonprofit sectors should focus on ensuring that data feeds to the planning tool are customized and routinely refreshed and that planning parameters are properly applied and extensively tested for transparency and meaningfulness.
C-2: Leaders from multiple sectors—biomedical, consumer products and services, computing, information technology, financial, transportation, agricultural, and construction—should be engaged through an advisory council mechanism.

It would be counterproductive and economically unfeasible if the various stakeholders each went about developing its own platform; hence the need for a “master version.” The tool will also require as much up-to-date data about the nation’s cancer control system as possible, so it will be important, for instance, that each state and territory use its own data—and refresh those data periodically for analyses and comparisons.

Large-scale tools such as this one envisioned for cancer control can be seen in regular use elsewhere in applications for monitoring, for example, the economy, weather, financial markets, labor dynamics, classified intelligence, and the manufacturing supply chain. Cancer affects everyone in one way or another. Thus, everyone has a stake in decisions about cancer control, which makes it crucial that the process of making those decisions be open and accountable. Successful national cancer control efforts will require a significant integration of resources and a major collaborative initiative among multiple participants to develop a joint ability with joint accounting and accountability. Using the science and engineering of complex adaptive systems offers productive possibilities for new progress in guiding the cancer control system to reduce the burden of cancers for individuals, families, and society as whole.