Cancer Health Economics Research: The Future Is Now

Abstract

The goals of the “Future of Cancer Health Economics Research” virtual conference were to identify challenges, gaps, and unmet needs for conducting cancer health economics research; and develop suggestions and ideas to address these challenges and to support the development of this field. The conference involved multiple presentations and panels featuring several key themes, including data limitations and fragmentation; improving research methods; role and impacts of structural and policy factors; and the transdisciplinary nature of this field. The conference also highlighted emerging areas such as communicating results with nonresearchers; balancing data accessibility and data security; emphasizing the needs of trainees; and including health equity as a focus in cancer health economics research. From this conference, it is clear that cancer health economics research can have substantial impacts on how cancer care is delivered and how related health-care policies are developed and implemented. To support further growth and development, this field should continue to welcome individuals from multiple disciplines and enhance opportunities for training in economics and in analytic methods and perspectives from across the social and clinical sciences. Researchers should continue to engage with diverse stakeholders throughout the cancer community, building collaborations and focusing on the goal of improving health and well-being.

Synthesis of the “Future of Cancer Health Economics Research” Virtual Conference

In synthesizing the findings and recommendations from the “Future of Cancer Health Economics Research” virtual conference, it is useful to revisit the goals of the conference as stated in the introduction to this Supplement: identify challenges, gaps, and unmet needs for conducting cancer health economics research; and develop suggestions and ideas to address these identified challenges and to support the development of this field (1). The first 2 presentations of this conference, which are described in separate articles in this Supplement, sought to identify current challenges in conducting cancer health economics research. The portfolio analysis of funded National Cancer Institute (NCI) grants from 2015 to 2020 found that only a small proportion of these grants (approximately 1%) included economic analyses, and these were generally not the primary focus of the grants (2). Most of the funded research has been in cancer prevention and screening. In contrast, the review of published studies in cancer health economics research found that most focused on treatment (3).

The graphic Framework for Cancer Health Economics Research, referenced in several presentations, provides an opportunity to think about gaps and unmet needs, including research areas that have not been well studied (Figure 1) (4). Starting with the “inputs” component of the Framework, there has been little research on the effects of health insurance benefit design on cancer care and outcomes. Benefit design may be important not only for reducing out-of-pocket costs and financial hardship for individuals with cancer, but also for reducing use of low-value, unnecessary, and potentially harmful cancer care services (5). These inputs affect economic outcomes across the cancer control continuum for both individuals and populations. Also understudied is the “structural factors” component of the Framework, including the role of the cancer care workforce, health-care organizations, and availability of personnel, services, and technology in the economics of cancer care. “Policy factors” such as payment models, mandates, and innovation and diffusion of technology have received more attention over the past few years, with a particular focus on the impacts of provisions in the Affordable Care Act such as Medicaid Expansion and the Oncology Care Model from the Centers for Medicare and Medicaid Services. “Factors” represent influences that, in general, are at a larger context level than are inputs. Regarding the “outcomes” components of the Framework, the conference focused on emerging areas, including financial hardship, employment impacts, and health equity, where there has been only limited cancer health economics research.

Figure 1.

Framework for cancer health economics research. This figure illustrates key inputs, patient- and population-level outputs, and structural and policy factors important to consider in cancer health economics research (4). Figure reprinted with permission from John Wiley and Sons.

Health Economics Research Across the Cancer Continuum: What We Have Learned, What Lies Ahead

Four presentations highlighted the challenges and unmet needs in cancer health economics research focused, respectively, on prevention, screening and diagnosis, treatment, and survivorship and discussed next steps to address these challenges (6-9). Several common themes emerged from these 4 presentations and the 2 discussions following the presentations, highlighting areas including data, methods, structural and policy factors, and transdisciplinary collaborations. First, a key issue is data limitations and fragmentation. Data are central to health economics research across the cancer control continuum, yet many data sources are limited to individuals who have a specific type of insurance or who receive care in a restricted geographic region or from a specific set of health-care organizations. This can limit the generalizability of research findings and prevent adequate longitudinal assessment, particularly among individuals who change health insurance or move to different regions. Additional limitations of data resources frequently used for cancer health economics research include limited or no information on:

• Exposures and risk factors;

• Social determinants of health, social needs, and social risk factors;

• Provider and provider practice characteristics;

• Molecular markers and other factors that may determine whether a patient’s cancer is appropriate for certain treatments;

• Patient-provider communications, provider treatment recommendations, and patient treatment preferences;

• Patient-reported outcomes; and

• Detailed clinical outcomes such as cancer recurrence.

Many of these data items would be considered inputs in the Framework illustrated in Figure 1. Provider practice characteristics and provider treatment recommendations can be classified as structural factors, and patient-reported outcomes and clinical outcomes are clearly “outcomes” in this Framework. In addition, caregiver information is also absent from most data resources. This type of information is critical to evaluate the overall burden of cancer and potential interventions that extend beyond patients to effects on the health and financial well-being of family members.

Another important data limitation is timeliness; there are almost always substantial delays between when cancer care interactions (across the cancer control continuum) occur and when data on these interactions are available to researchers. Much of this delay may be unavoidable due to issues of data transmission, cleaning, verification or validation, deidentification, and other necessary steps. That is, there may be necessary trade-offs between having high-quality data and having timely data. Similarly, there may be trade-offs between having comprehensive data and having timely data: the more complete a cancer data resource is or the more sources it draws from, the greater the time between patient interaction and availability of data to researchers. It is also important to consider when access to timely data is highly important (eg, economic assessments of recently introduced interventions) vs when it is less critical (eg, examining mediating or moderating effects of structural factors on associations between costs and outcomes). Overall, it would be worthwhile exploring approaches that optimize the timeliness of cancer care data without sacrificing data quality or comprehensiveness.

There were multiple opportunities discussed to improve data for cancer health economics research. Opportunities discussed included:

• Creating new linkages between existing data resources. This is an ongoing activity by multiple organizations, for example, by the NCI Healthcare Delivery Research Program involving Surveillance, Epidemiology and End Results (SEER)Program-linked data resources;

• Creating new linkages between existing data resources and novel or newly collected data. New types of information sources such as social media and reports on consumer spending patterns are being incorporated into cancer health economics research; and

• Developing or incorporating standardized economic measures in secondary data sources as well as prospective clinical trials and cohort studies. Examples of economic measures could involve episodes of care, including development of standardized costs, for a defined treatment interval.

These opportunities are discussed in more detail in other articles in this supplement (6–11).

The next common theme to improve cancer health economics research to emerge from the conference related to methods. One example, noted in several presentations, is to increase the use of modeling. Modeling has frequently been used to project costs and outcomes for cancer prevention and screening (eg, studies by CISNET, the Cancer Intervention and Surveillance Modeling Network) but less frequently for survivorship. Moreover, modeling has had only limited use in assessments of cancer-relevant policies, cancer care guidelines, health insurance benefit design, potential effects of alternative payment models, and strategies to enhance equity in cancer care. Other methods-related topics highlighted during the conference included application of evolving data science methods, such as machine learning, which will likely provide important opportunities for advancing cancer health economics research. New methods to better understand and incorporate the perspectives of decision makers are also needed. These may include, for example, use of value of information analysis (ie, formal assessment of the cost of acquiring information or conducting research compared with the benefit of the research results in reducing uncertainty) (12) to help inform decisions about whether to invest in additional data resources that could improve the accuracy or precision of economic analyses.

Another common theme emerging from the conference focused on increased opportunities to study the role and impacts of structural and policy factors. These factors (and their effects on cancer care delivery) include the organization of cancer care and associated market issues; overuse, underuse, and misuse of cancer care services; diffusion of innovation; deimplementation of low-value care; and the role of local context in cancer care delivery. Closely linked to both structural and policy factors, there is also a tremendous need for more cancer health economics research focused on underserved populations. Although there is already a large body of research describing the costs and outcomes associated with cancer health disparities (3, 13–15), research is now needed to examine economic factors associated with (and resulting from) strategies to address disparities in cancer care as well as the implementation and diffusion of strategies shown to be effective.

A final common theme discussed was cancer health economics research as a transdisciplinary field. As such, many of the opportunities for advancement noted during the conference involve collaborations among researchers in different disciplines and the training of individuals with diverse backgrounds. Collaborations may involve health economists, health services researchers, clinicians, epidemiologists, data scientists, policy makers, patient advocates, and patient stakeholders. There are existing opportunities for cancer health economics researchers to interact more frequently and effectively with existing research networks such as clinical trial cooperative groups, some of which are very interested in the economic aspects of cancer care. Researchers from implementation science brought important insights to the conference; there are opportunities to partner with these colleagues as well as learn from their efforts in advancing this field. Parallel to the transdisciplinary collaborations critical for cancer health economics research, cross-disciplinary training will be increasingly important for the next generation of researchers.

Conference Panels

The conference included 3 panels, with focused discussions around the theme “what do we need to do to be successful?” The communications panel highlighted the need to go beyond peer-reviewed publications and consider social media, lay press (eg, “op-eds”), and other approaches for disseminating research to policy makers and nonresearcher audiences. The data resources panel discussed enhancements that included new types of information, novel linkages, and use of social media data. The panel noted the importance of sharing data for novel projects that may be unrelated to the objectives for which the data were originally collected; given the substantial costs of data collection and curation, facilitating strategic reuse is critical. The health equity panel stressed that improving health insurance coverage is necessary but not sufficient to reduce cancer care disparities; diverse policies and other government, health-care system, and employer levers are important for achieving equity. The panel also highlighted disparities in clinical trial participation and the value of using “natural experiments” to explore strategies to address disparities.

There was much discussion about the importance of funding support to enhance the technical quality and policy relevance of cancer health economics research as well as interest in career development awards and training grants to support the future of this field. There was also substantial interest in developing “centers of excellence” in cancer health economics research and in incorporating health economic components in existing NCI-funded networks and potential future networks. Other opportunities discussed included targeted supplements for select health economics research questions, which would permit rapid funding and study initiation; providing expanded opportunities for the Cancer Intervention and Surveillance Modeling Network to investigate economic outcomes and related policy questions; and funding economic analyses in parallel with (ie, alongside) clinical trials.

Conference Breakout Sessions

The 4 conference breakout sessions raised additional important subjects and expanded on topics that had been discussed earlier related to training programs, fostering transdisciplinary collaborations and team science, data accessibility and data security, and methods development. Each group lived up to the moniker of being an “action” breakout session, proposing specific and actionable steps that could be implemented to advance cancer health economics research.

The training program session suggested introducing a webinar series for trainees interested in cancer health economics research across different institutions. This session also recognized the importance of training in communication with lay audiences and of health equity as part of training in this field; this training should be conducted jointly with community-based organizations or other community stakeholders to ensure that trainees have exposure to diverse, nonresearcher audiences and an understanding of the importance of activities such as community-based participatory research. The transdisciplinary collaborations session discussed how collaborations could improve economics research as well as barriers to and promoters for achieving effective collaborations. Having cancer health economics as a central aim in grants may be important for collaborations. Session participants suggested that the NCI could facilitate “matchmaking” between researchers in different disciplines at conferences such as this one.

The data accessibility and data security session highlighted changes in rules at the National Center for Health Statistics data centers that will facilitate support for health economics research. Session participants discussed the need for high levels of protection for confidential patient information and the development of flexibility in government rules related to data sharing. To achieve the dual aims of data accessibility and data security, improvements in data deidentification and data linkages may be important. The methods development session discussed the importance of transparency in details of analyses (eg, the specific billing codes used to identify a set of medical care services); the need for information, such as costs and Quality Adjusted Life Years (QALYs), to be readily accessible as a cancer health economics research “public good” to a range of researchers; and the relevance and trade-offs of different types of “real world data” (ie, representing the range of actual clinical practice) in economic analyses. Recommendations from this session included advocating for funding opportunities for economic methods; recruiting appropriate economic reviewers for study sections; learning to write for nontechnical audiences and producing tutorial documents; sharing programs and coding used for research projects; and facilitating collaborations involving clinicians and clinical fellows, health economics researchers, and policy makers to ensure that studies and findings are methodologically rigorous and have clinical and policy relevance.

Conclusions From the Conference

The introduction to this Supplement posed the (somewhat rhetorical) question: is now the time for the future of cancer economics research? Based on the presentations and discussions at the conference, we believe the answer is “yes”—the future is now. Cancer health economics research has evolved from a relatively niche field to an established discipline. It is poised to have a substantial impact on how cancer care is delivered and how policies are developed and implemented.

There are several conclusions that can be drawn from this conference. First, it is clear that cancer health economics research is a field involving synthesis, that is, “putting together different entities to make a whole which is new and different” (https://www.medicinenet.com/synthesis/definition.htm). Data resources used in cancer health economics research frequently combine information from disparate sources, for example, cancer registries, clinical records, medical claims, and surveys. For research in this field, clinical trials and models are fused, patient-reported outcomes and clinical measures merged, and social risk factors and legislative policies joined. A strength of the field is its fluidity, its ability to adapt, to take information from all sources.

Related to this concept of synthesis, it became clear in planning, conducting, and reviewing the “Future of Cancer Health Economics Research” virtual conference that much of this research is performed by individuals who, although having economics training, do not have graduate degrees in economics. Cancer health economics research is by design and by necessity a transdisciplinary field and not solely a subdiscipline of economics. Cancer health economics researchers have their primary training in fields including epidemiology, health services research, statistics, policy analysis, clinical medicine, and other areas, as well as economics. As such, the future of cancer health economics research will involve not only welcoming individuals from multiple disciplines; it will absolutely need individuals from multiple disciplines to grow and achieve its potential. One component of this will be to enhance opportunities for training in economics and in analytic methods from other disciplines to support this transdisciplinary area of research. Complementing this, it will be important to provide additional training opportunities related to cancer control for individuals with a general economics or health services research background.

Related to this theme of planning for the future is the important allied question: What is the scope of cancer health economics research? As was clear from many of the presentations, panels, and breakout sessions, this is not a field practiced in academic solitude. Researchers in this field are not afraid to “get their hands dirty,” work on policy initiatives, interact with community stakeholders, or speak with members of a legislative body. In addition to its transdisciplinary nature, cancer health economics research also needs on-the-ground input from cancer survivors, patient advocates, caregivers, health-care professionals, and policy makers. An essential component of this field, and one that differentiates it from certain other areas of economics, is a relentless focus on the “real world” and a primary objective of improving health and well-being. It was apparent that conference participants shared a sense of purpose about what this field can and should accomplish.

This shared sense of purpose also leads to thoughts of the future. Although this conference was hosted by the National Cancer Institute’s Division of Cancer Control and Population Sciences, it came about through the efforts of individuals from multiple agencies and organizations. Some of the work discussed at the conference will be continued through HEROiC, the Interagency Consortium to Promote Health Economics Research on Cancer (https://healthcaredelivery.cancer.gov/heroic/). For those who wish to participate in helping set the direction for this field, there are multiple opportunities to be engaged, help build collaborations, and identify next steps. Individuals can obtain more information from the NCI Healthcare Delivery Research Program’s Cancer Health Economics Research website (https://healthcaredelivery.cancer.gov/cancer-health/) and from other organizations and societies involved in research and conferences in this area. Although definitely not a comprehensive list, professional societies (in alphabetic order) whose conferences often involve cancer health economics research presentations include (but are not limited to):

• AcademyHealth (academyhealth.org)

• American Association for Cancer Research (aacr.org)

• American Public Health Association Cancer Forum (apha.org/apha-communities/forums/cancer-forum)

• American Society of Clinical Oncology (asco.org)

• American Society of Health Economists (ashecon.org)

• American Society of Preventive Oncology (aspo.org)

• International Health Economics Association (healtheconomics.org)

• ISPOR (ispor.org)

• Society for Medical Decision Making (smdm.org)

Future activities by the NCI Healthcare Delivery Research Program and the Interagency Consortium to Promote Health Economics Research on Cancer (HEROiC) include plans to bring together government agencies and nonprofit organizations to identify opportunities to further strengthen the agenda for cancer health economics research.

At present, the number of interesting and important cancer health economics research questions almost certainly outstrips the supply of trained investigators to explore these questions. For the future, it is important to consider priorities. Specifically, how can cancer health economics research be most effective in helping patients, families, health care provider, health-care systems, policy makers, and others across the cancer control continuum? Where should research efforts be concentrated, particularly in view of limited research resources and the compelling need to address inequities? Clearly, there are no easy answers to these questions, but they do provide useful guideposts for future development of this field.

To help enhance the future of cancer health economics research, we hope all of those who participated in the conference—and all those who now study its findings and recommendations in this monograph—will take up this call to action. This may involve creating new research collaborations by reaching out to unaccustomed partners, including researchers in unfamiliar departments, institutions, or organizations; policy makers and government officials; health-care systems, health-care providers, and professional societies; or cancer survivors, caregivers, community leaders, and patient advocates and patient advocacy organizations. This may involve offering to help with or lead new activities at future conferences that will strengthen cancer health economics research. It may involve volunteering to be a grant reviewer or to write a grant proposal with new research collaborators, such as community-based groups. It may involve learning new research methods or mentoring a colleague (or future colleague). None of these are easy, and they involve investments of time and additional research resources. However, they are key for the growth of cancer health economics research.

Notes

Role of the funder: No funding was received for this study and the authors indicate no conflicts of interest.

Disclosures: None.

Author contributions: All authors participated in the conceptualization and writing of this manuscript.

Disclaimer: The views expressed here are those of the authors and do not necessarily represent any official position of the National Cancer Institute, National Institutes of Health, or American Cancer Society.

Prior presentation: A version of this synthesis and conclusion was presented at the 2020 Future of Cancer Health Economics Research virtual conference (https://healthcaredelivery.cancer.gov/heroic/conference.html).