Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, ECONOMICS AND FINANCE (oxfordre.com/economics). (c) Oxford University Press USA, 2019. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 17 November 2019

Economics of Cancer Prevention and Control

Summary and Keywords

The goal of cancer prevention and control is to reduce cancer risk, morbidity, and mortality through transdisciplinary collaborations across biomedical, behavioral, and social sciences. Risk reduction, early detection, and timely treatment are the rationales behind policy efforts to promote cancer prevention. Economics makes three important contributions to cancer prevention and control research. Firstly, research built upon the human capital model by Grossman and the insurance model by Ehrlich and Becker offers solid theoretical foundations to study human behaviors related to preventive care. Secondly, economic evaluation provides useful analytical tools to assess the “cancer premium” (through the stated preference research approach) and to identify the optimal screening strategy (through cost-effectiveness analysis). Lastly, the rich set of quantitative methods in applied economics contributes to the estimation of the relative contribution of prevention versus treatment in the reduction of cancer mortality and the evaluation of the impact of guidelines to regulate screening practices or policy initiatives to promote cancer screening.

Keywords: economics of prevention, cancer screening, cancer premium, economic evaluation, overdiagnosis, health economics

Introduction

Cancer prevention and control encompasses collaborative efforts across biomedical, behavioral, and social sciences to reduce the disease burden of cancer. A conceptual framework for cancer prevention and control research was proposed by the Advisory Committee on Cancer Control of the National Cancer Institute of Canada in 1997, adapted by the National Cancer Institute in the United States in the early 2000s, and remains highly relevant for cancer research in 2018 (Best, Hiatt, Cameron, Rimer, & Abrams, 2003). Under this framework, cancer prevention and control research involves five phases, each answering an important question in its scientific pursuit. Critical questions addressed in each research phase begin with “What do we know?” in foundational research, followed by “What works?” in intervention research, “Where are we?” in surveillance research, “What is next?” in knowledge synthesis, and finally “How do we deliver it?” in dissemination and implementation research (Hiatt & Rimer, 2006).

“Transdisciplinarity,” defined by Rosenfield (1992) as a process by which researchers work jointly using a shared conceptual framework that integrates and synthesizes discipline-specific knowledge to address a common problem, has been the guiding principle of training programs for cancer prevention and control for more than two decades (Colditz & Wolin, 2011). It is in the spirit of transdisciplinarity that economics contributes to the grand scheme of cancer prevention and control research, with the ultimate goal to reduce the financial, physical, and psychological devastation that cancer imposes on patients and their families, the healthcare system, and society as a whole.

Economics offers rigorous theoretical foundations to understand human behaviors and a rich set of analytical methods to evaluate policy and intervention effects, examine demand and supply conditions, assess allocation efficiency, and determine factors associated with market failure. To understand the contribution of economics in cancer prevention and control research, it is important to differentiate between primary and secondary prevention of cancer (Spratt, 1981). Primary prevention of cancer consists of activities that prevent the occurrence of either a precancerous or cancerous condition. Activities such as smoking cessation, having a healthy diet, exercising, wearing sun protection clothing and lotion, and minimizing environmental carcinogens are examples of primary prevention. Secondary prevention of cancer involves the detection of asymptomatic cancers, precancerous conditions, or cancer at early stages so as to allow the opportunity of early intervention and timely treatment. A classic example of secondary prevention of cancer is cancer screening, such as screening mammography for breast cancer, and colonoscopy for colorectal cancer.

The objective of this article is to provide a synthetic overview on topics of the economics of cancer prevention and control. The book chapter on prevention, authored by Kenkel (2000), in the Handbook of Health Economics provides an excellent overview of the health economics literature on the economics of prevention. Although that book chapter was published in 2000 and the focus was on prevention in general, many concepts discussed there are directly applicable to the economics of cancer prevention and control discussed here. While this article builds on the knowledge, especially the theoretical models, discussed in Kenkel’s book chapter, there are two important differences. Firstly, this article will focus primarily on secondary prevention of cancer because activities for primary prevention of cancer often are also applicable to other diseases and have been more extensively discussed in the health economics literature. For example, the latest edition of the textbook Health Economics by Folland, Goodman, and Stano (2016) provides a rather comprehensive overview on economic studies related to primary prevention in the chapter “The Health Economics of Bads.” Secondly, this article will focus more attention on economic evaluation studies. Economic evaluation is a fast-growing field in health economics and has made significant contributions to the economics of cancer prevention and control through the evaluation of the cost-effectiveness or benefit–harm trade-off among cancer screening strategies or the assessment of consumers’ willingness to pay for cancer prevention through stated preference methods.

The rest of this article is organized as follows. Section 1 will provide a brief overview on economic theories that explain the behaviors of prevention. Section 2 will explore the role of cancer prevention in reducing cancer mortality. Section 3 will introduce the concept of “cancer premium” by reviewing studies that assessed consumers’ willingness to pay for cancer prevention. Section 4 will highlight practical and analytical considerations for cost-effectiveness analysis designed to identify the optimal screening strategy. Section 5 will discuss policy initiatives to promote cancer prevention and studies that assess the effectiveness of these policies, followed by concluding remarks in Section 6.

Economic Theories of Cancer Prevention and Control

A theoretical foundation for the economics of cancer prevention and control can be traced back to Grossman’s human capital model (Grossman, 1972). Under the framework of the human capital model, demand for prevention serves the same purpose as demand for medical care, which is an investment in health to increase the stock of health capital. The distinction between preventive and curative care became more clear in further extension of the human capital model by Grossman and Rand (1974), in which prevention and cure were modeled as separate inputs in the health production function, with the assumption that individuals with low depreciation rates of health stock (e.g., young people in excellent health) would demand more preventive care whereas those with high depreciation rates (e.g., older people in frail health) would demand more curative care. Hey and Patel (1983) developed a two-state (sickness and health) dynamic model and approached the issue of preventive versus curative care from the perspective of optimal allocation of expenditures between prevention and cure. Both theoretical models assumed prevention and cure were substitutes and did not further differentiate between primary and secondary prevention.

The seminal work by Ehrlich and Becker (1972) offers the most insightful theoretical framework for the understanding of the economics of cancer prevention and control. The model considered three types of behavior related to insurance decision under uncertainty: a market insurance that was purchased to provide some protection of income had a bad state of nature occurred; self-protection referred to activities to reduce the probability of the occurrence of a bad state of nature; and self-insurance involved behaviors to reduce the magnitude of a loss if a bad state of nature occurred. Of these three behaviors, “market insurance” refers to the behavior of purchasing insurance (e.g., health insurance, home owner insurance) from the insurance market, whereas “self-protection” and “self-insurance” are risk protection tools or behaviors that interact with the insurance purchasing decision. Ehrlich and Becker (1972) showed that market insurance and self-insurance were substitutes, whereas market insurance and self-protection can be complements. Kenkel (2000) pointed out that self-protection and self-insurance correspond to primary and secondary prevention, respectively. Putting this model in the context of cancer prevention and control, self-protection (i.e., primary prevention) lowers the risk of cancer whereas self-insurance (i.e., secondary prevention) increases the effect of cancer treatment through early detection. Indeed, the insurance model developed by Ehrlich and Becker (1972) has served as the theoretical foundation for many health economic studies related to cancer prevention and control. For example, the work by Kenkel (1994) on the use of screening for breast and cervical cancer suggested that when it comes to secondary prevention, prevention and cure are complements rather than substitutes. Simon, Sonia, and Cawley (2017) examined whether Medicaid expansion under the Affordable Care Act (ACA), launched in 2010 in the United States, could cause ex ante moral hazard, thereby increasing risky health behaviors, and found no evidence of ex ante moral hazard.

Cancer Prevention and Reduction in Cancer Mortality

The relative contribution of prevention and treatment in reducing cancer mortality has been a topic of heated debate for more than two decades. To understand the relative contribution of these two cancer-fighting forces on the trends of cancer incidence and mortality at the population level, the National Cancer Institute funded a research consortium in 2000, known as the Cancer Intervention and Surveillance Modeling Network (CISNET), to develop simulation models using mathematical or statistical methods to address this research question (Habbema, Schechter, Cronin, Clarke, & Feuer, 2006). There were four phases of CISNET funding as of December 2017. After addressing the relative contribution of primary prevention, secondary prevention, and treatment in breast, colorectal, prostate, and lung cancer in Phase I funding (2000–2005), the CISNET models have been expanded to address other important topics in cancer research and policies and added two more cancer sites, esophagus and cervix, in other funding phases. Findings from these microsimulation models have been used to assist the U.S. Preventive Services Task Force (USPSTF) in the development of screening guidelines for breast, colorectal, and lung cancer (de Koning et al., 2014; Knudsen et al., 2016; Mandelblatt et al., 2016).

Economists approached the above research question by applying decomposition techniques to separate out the effect of prevention from that of treatment on cancer mortality. In his article assessing factors contributing to the reduction in cancer mortality, Cutler (2008) obtained information from published CISNET models and epidemiological studies to decompose the reduction in cancer mortality observed between 1990 and 2004 for breast, colorectal, lung, and prostate cancer into three sources: primary prevention from changes in environmental and behavioral risk factors, secondary prevention from screening, and treatment. He concluded that these three sources combined accounted for 78% of the observed reduction in cancer mortality, with primary prevention, secondary prevention, and treatment accounting for 23%, 35%, and 20% of the reduction, respectively. That is, prevention alone contributed close to 60% of the mortality reduction in the 15 years’ duration (1990–2004) for the four cancers included in Cutler’s analysis.

Another health economic study reached a different conclusion and suggested a much more limited role of prevention in mortality reduction. Using data from the Surveillance, Epidemiology, and End Results (SEER) program, Sun, Jena, Lakdawalla, Reddy, and Philipson (2010) combined the decomposition technique and survival analysis to estimate the relative contribution of prevention and treatment in the reduction of all-cause mortality observed among cancer patients between 1988 and 2000. The SEER program is an epidemiological surveillance system of population-based tumor registries. The program began in 1973, and as of March 2012 it had collected data from 20 geographic areas in the United States, covering approximately 28% of the U.S. population (NCI). The authors restricted their analysis to the nine original SEER sites so as to track survival from the earliest data capture. Statistical models assessing the impact of screening on survival often encounter two biases: lead time and length bias. Lead time is the amount of time by which the diagnosis has been advanced by screening, whereas length bias refers to the situation that slow-progressing tumors are more likely to be detected by screening. Both biases lead to an artificial survival benefit for screening-detected cancers (Duffy et al., 2008). After developing algorithms to adjust for lead time and length biases, Sun et al. (2010) concluded that treatment advances explained 80–90% of the observed survival gains between 1988 and 2000. In analysis stratified by cancer type, they reported that treatment accounted for 83%, 85%, 76%, 100%, and 96% of the survival gains in breast cancer, lung cancer, colorectal cancer, pancreatic cancer, and non-Hodgkin’s lymphoma, respectively.

Given the similarity of time periods between these two studies, why did they reach vastly different conclusions on the role of prevention in reducing cancer mortality? Other than the difference in analytical approaches, there are two plausible explanations. Firstly, mortality in Sun et al. (2010) referred to all-cause mortality, instead of cancer-specific mortality. Secondly, unlike microsimulation models that can “isolate” the effect of screening by comparing survival between screening and screening-free cohorts, the data analyzed in Sun et al. (2010) covered a time period in which screening was already prevalent in some cancers (e.g., breast cancer). Without some form of counterfactual analysis to tease out the effect of screening from that of treatment, the estimates from Sun et al. (2010) might be less reliable. Indeed, a recent analysis of SEER data of breast cancer sheds new light on this inquiry. Using SEER data from 1975 to 2012, Welch, Prorok, O’Malley, and Kramer (2016) explored tumor size distribution and size-specific cancer mortality with a minimum of 10 years’ follow-up data in two periods: a baseline period from 1975 to 1979, representing the duration before widespread use of screening mammography, and a status quo period from 2000 to 2012, capturing contemporary utilization patterns of screening mammography. The study concluded that treatment improvement accounted for approximately two-thirds of the mortality reduction in breast cancer.

Welch et al. (2016) also showed that increasing use of screening mammography was associated with a shift in the size distribution of breast cancer diagnosed over time, with the proportion of in situ carcinoma or small-size (< 2 cm) invasive tumors increasing from 36% in 1975 to 68% in 2012. The authors estimated that while the incidence of large tumors decreased by 30 cases per 100,000 women during the study period, the incidence of small tumors increased by 162 cases per 100,000. And of these 162 cases (per 100,000 women), 132 (over 80%) were overdiagnosed. “Overdiagnosed” cancers are cancers detected from screening that would never have led to clinical symptoms or cancer death during a patient’s lifetime. The rather large magnitude of overdiagnosis from screening mammography estimated from this study is alarming. Overdiagnosis and false-positive test results are two undesirable aspects of screening, as overdiagnosis leads to overtreatment whereas false-positive findings often result in unnecessary biopsies.

Understanding the relative contribution of prevention and treatment in reducing cancer mortality is a topic of immense interest to policymakers as such knowledge provides important insights on how to best allocate limited healthcare resource to cancer prevention and control. To credibly estimate the impact of screening on cancer mortality, studies must consider four methodological issues. Firstly, the study needs to be able to distinguish period effects from cohort effects, including differentiation of period and cohort effects on trends. For example, instead of attributing the increase in breast cancer incidence over time to overdiagnosis driven by screening, it is equally important to recognize that some increase in breast cancer incidence may be related to epidemiological trends such as changes in age at first live birth, increase in obesity, or the use of hormonal replacement therapy. Secondly, the study must accurately quantify exposure to screening. For example, it is not appropriate to estimate prevalence of annual screening based on a self-reported survey question asking about screening behavior “in the past two years.” Thirdly, the study should properly adjust for lead time bias and account for competing risks; the latter is especially important for the elderly population. Lastly, the study must have sufficient duration of follow-up as it can take years for the benefit of screening to take effect. For example, trial evidence has indicated that there would be no reduction in breast cancer deaths until seven years after mammography.

Health economic studies often apply an observational study approach to answer the research question of the relative contribution of screening versus treatment. However, none of the studies discussed in this section has properly addressed the complex methodological issues mentioned above. While modeling studies can potentially handle all these analytical challenges, the quality of models varies widely and hinges critically upon the validity of assumptions, the credibility of data sources, and the reliability of the modeling approach to capture the disease history and progression. When discussing the relative contribution of screening versus treatment in cancer mortality reduction, it is important to bear in mind that the exposure to each cancer-fighting force differs by population. Findings from modeling studies that assume full adherence to screening and treatment will be more applicable to populations residing in countries with organized screening but less so for those in countries with opportunistic screening, such as the United States. Even for populations in the same country, treatment adherence is often suboptimal among certain population subgroups, which could then lower the effect of screening on these subgroups and alter the relative contribution of screening versus treatment. Failure to recognize unequal exposure to screening and treatment among populations of interest can thus lead to misguided policy decisions.

Consumers’ Willingness to Pay for Cancer Prevention

The ultimate goal of cancer prevention and control is to reduce cancer risk, morbidity, and mortality through primary and secondary prevention. For consumers and policymakers, the decision to make an investment in cancer prevention activities will depend on whether the marginal benefit of reducing cancer risk outweighs the marginal cost associated with these activities. The value of cancer prevention and control perceived by consumers can be understood from economic studies assessing “cancer premiums.” A cancer premium refers to a higher willingness to pay (WTP) expressed by consumers in order to avoid cancer risks relative to other types of risks. Van Houtven, Sullivan, and Dockins (2008) applied a stated preference approach to examine individuals’ trade-offs between two types of risks, cancer vs. automobile accident, and explored whether cancer premiums would decrease as the cancer latency period increased. The authors structured a risk–risk trade-off model to explain the concept of cancer premiums:

EQ(1)

E(U)=PDU(D,Y)+PcU(C,Y)+(1PDPc)U(H,Y)

In EQ(1), lifetime utility U is determined by health outcomes (D, C, or H) and wealth (Y), where D, C, and H indicate fatal automobile accident, fatal cancer, and normal health, PD is the probability of dying within a very short period of time from an automobile accident, and PC is the probability of contracting and eventually dying of cancer. Maximizing expected utility and assuming U(D,Y) = 0 yields the following equation:

EQ(2)

MER=VSCVSL=(1U(C,Y)U(H,Y))

, where VSL = dY/dPD is the marginal rate of substitution between income and risk of immediate death from automobile incident, and VSC = dY/dPC is the marginal rate of substitution between income and risk of eventual cancer death. The mortality equivalence ratio (MER) quantified avoided fatal cancer into an equivalent unit of avoided death from automobile accidents, and MER > 1 implies the existence of cancer premiums in that consumers value avoided fatal cancer risks more than avoided immediate mortality risk from accidents. The authors further expanded the above framework to incorporate latency effect by allowing U(C, Y) to be time dependent: U(C(t), Y), with t representing the duration of the cancer latency period.

To measure the magnitude of cancer premiums, Van Houtven et al. (2008) designed a survey based on the concept of risk–risk trade-offs that asked respondents to choose between two locations: one had fewer automobile deaths and the other had fewer cancer deaths compared with the respondent’s current location. Respondents were also randomly assigned to cancer type (stomach, liver, or brain), latency period (five, 15, or 25 years), and cancer morbidity period (two or five years). Findings from their empirical analyses showed a significant cancer premium and that the premium declined as the cancer latency period lengthened. Linking VSL in EQ(2) to the value of statistical life literature (Viscusi & Aldy, 2003), the authors estimated that with a 15-year cancer latency period, the value of statistical cancer death avoided or VSC was approximately $14 million, compared with $6 million of VSL for immediate accidental risks reported in the literature. That is, cancer has a value more than twice as large as fatal automobile accidents when the cancer latency period is 15 years. Additional analyses indicated that the cancer latency period would need to be longer than 30 years to offset the cancer premium.

More recently, Viscusi, Huber, and Bell (2014) applied a different stated preference approach to estimate VSL for cancer risks. Instead of risk–risk trade-offs, Viscusi et al. (2014) designed a survey to elicit risk–dollar trade-offs and estimated VSC from the costs that respondents were willing to incur for policies or interventions that reduce their risk of cancer. The survey was framed in the context of fatal bladder cancer risks from exposure to arsenic in drinking water. Based on a latency period of 10 years, the estimated VSL for an immediate risk of cancer was $10.85 million, indicating a cancer premium 21% greater than the median VSL for acute fatalities. It is not clear to what extent the substantially different magnitude of cancer premiums estimated from these two studies was driven by the difference in the elicitation method (risk–risk vs. risk–dollar trade-offs) or the types of cancers included in the survey (stomach, liver, and brain vs. bladder cancer). However, both studies took a more environmental research approach in the sense that cancer was framed as caused by location- or environment-specific exposures. In addition, none of the cancers queried in the surveys have well-established cancer screening modalities. Therefore, it is reasonable to conjecture that when considering actions or policies to reduce cancer risks, survey respondents in neither study were thinking about secondary prevention. A meaningful extension of this line of research will be to apply the stated preference approach to examine whether a cancer premium exists for cancers for which the risk can be reduced through screening. Future research on this topic should be aware that while some screenings (e.g., mammography) reduce the risk of dying from cancer through detecting cancer at an early stage, others can go beyond early detection and prevent the onset of cancer. An example of the latter type of screening is colonoscopy, which removes polyps detected during the procedure and thus prevents these noncancerous polyps from developing into colorectal cancer.

Milligan, Bohara, and Pagan (2010) employed the contingent valuation method to directly assess consumers’ WTP for cancer prevention. The authors inferred WTP from a question in the 2002 Health and Retirement Study survey that asked respondents to state whether they were willing and the dollar amount per month to pay for a new drug that would be fully effective at preventing cancer without any side effects. The study found that 72% of respondents were willing to pay more than $100 per month for the hypothetical cancer prevention drug, and 10% indicated that they were willing to pay at least $1,000 per month. Not surprisingly, the study also found that WTP was negatively correlated with age and positively correlated with income and the probability of developing cancer. An interesting finding was that about half of the respondents felt that their risk of developing cancer in the future was greater than 50%. However, interpreting the findings of this study in the context of cancer prevention and control is challenging because although the survey question asked about “cancer prevention,” the framing of the question as “a new drug without side effects” is somewhat detached from the clinical reality of cancer prevention activities.

Another study applied conjoint analysis, also known as discrete ranking modeling, to assess public preferences for cancer screening programs (Gyrd-Hansen & Sogaard, 2001). This study explicitly included false-positive diagnosis, an undesirable aspect of screening, as one of the four “attributes” associated with screening. The other three attributes were: life cycle costs, lifetime risk reduction, and number of tests over a lifetime. Survey participants were randomly sampled from the Danish population. Specifically, a random sample of men and women aged 50 and over was selected for interviews related to colorectal cancer screening, and another random sample of women ≥ age 50 were recruited for breast cancer screening interviews. The estimated WTP for screening programs was between DKK 1,800 and 3,400 per year, which translated to WTP for a statistical life in the range of DKK 8.4 to 11.9 million. The authors commented that although the amount of WTP estimated from their study seemed high, it was within the range of WTP for saving a statistical life reported in the literature. Although no cancer premium was conferred in this study, it should be noted that the WTP estimated in this study was limited only to “cancer screening programs.”

Consumers’ tendency to overestimate cancer risk and the perception of cancer as a “dread” disease contributed to the cancer premium documented in the literature and could also raise consumers’ demand for screening. Historically, prostate-specific antigen (PSA) testing had been recommended for prostate cancer screening for men ≥ age 50 and mammography for breast cancer screening for women ≥ age 40. The USPSTF recommended against routine screening for prostate cancer with PSA in 1996 (i.e., recommendation D), although the recommendations have changed a few times in the subsequent updates over the years. For breast cancer screening, the 2009 updated guideline from the USPSTF raised the screening initiation age to 50 and concluded that there was insufficient evidence to recommend mammography screening for women in the age range of 40 to 49 (i.e., recommendation I). The high value that consumers place on screening often triggers protests or outbursts from the public for any attempt to scale back screening by either removing it from screening guidelines (e.g., PSA for prostate cancer) or recommending a later starting age or less frequent screening schedule (e.g., raising the starting age of mammography screening from age 40 to 50) (Chapman, 2003; Ehrenreich, 2010; Yamey & Wilkes, 2002). The fear of cancer could prompt some consumers to avoid cancers at all costs, leading to behaviors that have puzzled clinicians. For example, bilateral prophylactic mastectomy (BPM) is one of the preventive strategies recommended by the Society of Surgical Oncology for women who have BRCA mutations or pathological risk factors, or a strong family history of breast cancer (Giuliano et al., 2007). Yet a recent study in the United States found that of the 71 BPMs identified from a retrospective chart review, nine (~13%) women appeared to choose BPM based solely on cancer-risk anxiety or personal preference as these women did not have BRCA mutations, pathological risk factors, or a family history of breast cancer (Rueth et al., 2011).

In Search of the Optimal Screening Strategy

As noted by Kenkel in his book chapter in the Handbook of Health Economics, prevention has been a popular subject for economic evaluation, in particular cost-effectiveness analysis (CEA) (Kenkel, 2000). It is generally accepted that primary prevention measures, such as smoking cessation, exercise, and healthy diet, offer health benefits, especially in the long run. CEA related to primary prevention therefore focuses less on the behavior itself and more on comparing behavior-modification interventions or policies to achieve the goal of primary prevention. Examples of this type of CEA can be found in the collection of articles included in a systematic review of economic evaluations of smoking cessation interventions (Ronckers, Groot, & Ament, 2005). For secondary prevention of cancer, once the clinical benefit of a screening modality is deemed acceptable (and this is not necessarily based on evidence established through clinical trials, but also often includes post-trial studies), the focus is often on using CEA as a tool to determine the optimal screening strategy for the targeted population. Even for the same screening modality (e.g., mammography or colonoscopy), the actual implementation of it in a screening program would involve a minimum of four decisions: at what age to start screening, when to stop screening, how frequently to screen, and for whom. Upon ascertaining the targeted population for screening, the combination of initiation and cessation age, plus screening frequency, will lead to a multitude of screening strategies to be considered. Take, for example, mammography screening for women at average risk of breast cancer: the initiation age of screening explored in the updated guidelines in the United States around 2015 and 2016 includes ages 40, 45, and 50, cessation age could be 75, 80, or no upper limit, and the screening frequency ranges from annual to biennial, resulting in a total of 19 screening strategies, 18 (3 × 3 × 2) strategies involving screening mammography plus the no screening option, to be considered in CEA.

Multiple comparators is an analytical issue commonly encountered in assessing the cost-effectiveness of secondary prevention strategies. As discussed earlier, the combination of screening frequency, initiation, and cessation age leads to a large number of potential screening strategies. While both deterministic analyses accompanied by tornado diagrams and probabilistic analyses are acceptable ways to present findings from model-based CEA per the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) released in 2013 (Husereau et al., 2013), preliminary work from Shih, Dong, Xu, and Yu (2017) showed that because differences in quality-adjusted life years (QALYs) among different screening strategies tended to be small, the optimal strategy chosen from deterministic analysis could vary from one simulation to another. Even though this issue could be mitigated by probabilistic analyses, Shih et al. (2017) demonstrated that the probability of a certain screening strategy being optimal was sensitive to the total number of strategies available in the “choice set” and called for more methodological research in microsimulation model-based CEA involving multiple comparators. This line of research is especially relevant to studies assessing the cost-effectiveness of screening strategies. Another analytical consideration is that the validity of modeling studies hinges critically upon the quality of modeling parameters. For model-based CEA of screening strategies, while cost information may be relatively easier to obtain, the efficacy—especially in the long term—of screening technologies or strategies from randomized clinical trials is often unavailable, let alone effectiveness information from pragmatic trials. The review article by Whynes (2004) provides a nice summary of the challenges in generating rigorous clinical evidence for cancer screening programs. In addition to the extensive resources required for screening trials due to the need to recruit a large number of participants and the long follow-up duration, the biggest challenge is that once a screening modality has been adopted in medical practice, it becomes nearly impossible to recruit trial participants.

The fast diffusion of new technology in cancer screening renders it necessary to continuously update and re-evaluate the cost-effectiveness of screening strategies. As old technologies are replaced by new ones, earlier CEA can quickly become irrelevant to policymakers. For example, the vast majority of mammography screening facilities had transitioned from film to digital mammography by 2010 (Henderson, O’Meara, Braithwaite, Onega, & Breast Cancer Surveillance Consortium, 2015). Digital breast tomosynthesis, also known as 3D mammography, has gained popularity in the United States since its approval by the Food and Drug Administration (FDA) in 2011 (Clark et al., 2017). In the case of colorectal cancer, while colonoscopy, fecal occult blood testing (FOBT), and flexible sigmoidoscopy are the recommended modalities for colorectal cancer screening, estimates from the Centers for Disease Control and Prevention (CDC) indicated that only 65% of the eligible adult population in the United States were up to date with colorectal screening in 2012 (Chowdhury et al., 2016) and the screening rate remained low in 2015 (White et al., 2017). Some have attributed the low rate of screening to the more invasive nature of endoscopy procedures, and new screening modalities developed for colorectal screening, such as CT colonography and fecal DNA testing, often promote themselves as minimally invasive or non-invasive alternatives to colonoscopy. As of July 2018, no conclusive findings had been reached regarding the cost-effectiveness of these new screening modalities for colorectal cancer (Skally, Hanly, & Sharp, 2013; Sweet et al., 2011; Zauber, 2010). For cervical cancer, science has substantially improved our knowledge of the etiology of cervical cancer to the extent that the European guideline for cervical cancer screening, updated as supplements in 2015, no longer recommends primary screening with Papanicolau smear test (Pap test) but recommends primary human papillomavirus (HPV) testing instead (von Karsa et al., 2015). The 2015 U.S. guideline differs from the European guideline in that it recommends a Pap test every three years or co-testing with a Pap test in combination with an HPV test every five years for women between the ages of 30 and 65 (Schlichte & Guidry, 2015). For CEA to be useful to local policymakers, it is therefore important to stay current with the landscape of screening technologies as well as the local clinical and policy environment.

Screening guidelines from professional societies offer a reliable source of knowledge on the current landscape of screening. These guidelines are updated periodically to reflect new evidence from scientific discoveries. Often, guidelines released from different organizations have different recommendations. The lack of consensus across guidelines from different professional associations can cause confusion for consumers as well as clinicians. For example, the American Cancer Society (ACS) updated their breast cancer screening guideline for average-risk women in 2015. The updated ACS guideline recommends a hybrid strategy that initiates annual screening at age 45, transitions to biennial screening at age 55, and continues screening as long as the individual is in good health. The ACS guideline also states that women between the ages of 40 and 44 should be offered the opportunity to consider beginning annual screening before age 45 and that women aged 55 and above can continue annual screening if the decision is consistent with their preference (Oeffinger et al., 2015). This recommendation not only deviates from the ACS’s previous guideline, which recommended annual screening starting at age 40, but also differs from the biennial schedule beginning at age 50 recommended by the USPSTF. Shih, Doug, Xu, and Shen (2019) employed a microsimulation model to assess the cost-effectiveness of U.S.-based mammography screening guidelines from various medical societies. They concluded that the updated ACS guideline was most cost-effective, yielding an incremental cost-effectiveness ratio slightly above $40,000/QALY when compared with the USPSTF guideline. The age to stop screening has been a gray area in most guidelines. Rafia et al. (2016) developed a model to identify the upper age limit where screening mammography would remain cost-effective for England and Wales and concluded that extending screening up to the age of 78 years would be a cost-effective strategy.

Policy Effectiveness to Promote Cancer Prevention and Control

Health policies targeting primary prevention largely focus on regulatory efforts to encourage health-promoting behaviors or disincentivize health-harming behaviors, whereas policies for secondary prevention mostly aim to increase the utilization of various forms of screening. Taxation is the most commonly implemented policy to discourage the consumption of unhealthy goods, such as tobacco and sugar-sweetened beverages. Bans (e.g., smoke-free laws or restrictions on indoor tanning), restrictions on advertisement, and penalties (e.g., jail sentence for drunk driving) have also been enforced in various parts of the world. The rationale for instituting regulations to intervene in markets for these commodities is externality in that the consumption of these “health bads” often incurs a cost to others, such as health risks for secondhand smokers, higher insurance premiums for non-smokers, or accidents caused by drunk driving (Folland, Goodman, & Stano, 2016). The Costs of Poor Health Habits, a very interesting book by Manning, Keeler, Newhouse, Sloss, and Wasserman (1991), estimated that the external lifetime costs associated with smoking, heavy drinking, and sedentary lifestyles were $1,000, $42,000, and $1,650 per person, respectively. The reason that the external lifetime costs of heavy drinking were 42 times the costs of smoking was because a substantial proportion of the external costs imposed by heavy drinking were driven by consequences of drunk driving, such as loss of innocent lives and property damage, and resources consumed by the public systems and programs to deal with alcohol-related issues. Of the $42,000 external lifetime costs estimated for heavy drinking, $24,000 was associated with innocent lives lost in alcohol-related traffic accidents and another $14,000 was from loss of property, strain on the criminal justice system, and social programs. Neither smoking nor sedentary lifestyles would incur these types of external costs. Furthermore, the relatively lower external lifetime cost of smoking compared with other poor health habits was largely because smoking shortened life expectancy and thus the duration of “lifetime” to accumulate external costs.

Although consumers tend to respond negatively to guidelines that tighten the eligibility criteria of cancer screening, the public’s favoritism toward screening does not always translate into a high uptake rate. Screening utilization has been found to be associated with individual as well as institutional factors. Lower rates of cancer screening have been shown to be more common among racial/ethnic minorities, individuals with lower income or education level, and the uninsured (Chen, Kessler, Mori, & Chauhan, 2012; Sambamoorthi & McAlpine, 2003). Studies in the United States found that individuals who resided in areas with higher market share of HMO or enrolled in health insurance plans with gatekeeping requirements were more likely to have screening mammography, clinical breast examination, and Pap smear, but not prostate cancer screening (Baker, Phillips, Haas, Liang, & Sonneborn, 2004; Phillips et al., 2004). The authors of these studies postulated that the lack of association between these institutional factors and prostate cancer screening was probably because screening for prostate cancer was not uniformly recommended at the time of their study.

Several economic studies have examined screening behavior from the angle of information processing. Allocative efficiency, the central hypothesis in this line of research, is rooted in Grossman’s human capital model which postulated that educated individuals were healthier because they were able to produce health more efficiently (Grossman, 1972). Kenkel (1994) analyzed two sets of household survey data to study demand for two preventive services—breast examination and Pap smear—and found a positive association between education and the use of preventive care. The author suggested that two mechanisms could explain this association: allocative efficiency or time preference. Lange (2011) tested this allocative efficiency hypothesis for the relationship between education and health in the context of perceived cancer risk and cancer screening behavior. Using data from the National Health Interview Survey, the author found that educated individuals were more likely to incorporate objective, science-based evidence in the assessment of their personal risk of cancer whereas less-educated individuals tended to be more skeptical about objective evidence. In addition, educated individuals were likely to be screened if they perceived themselves at higher risk of developing cancer based on objective evidence. Lange concluded that this study supported the allocative efficiency hypothesis as educated individuals, by being able to better process health information, appeared to make better health decisions.

If the ability to more accurately assess one’s cancer risk would raise cancer awareness and subsequently participation in screening, then a promising public health action is to promote cancer awareness through campaigns. Jacobsen and Jacobsen (2011) evaluated the policy effectiveness of an awareness campaign called National Breast Cancer Awareness Month (NBCAM), a nationwide campaign established in the United States in 1985 to promote breast screening. Given that the NBCAM campaign occurs in October, the authors used the number of breast cancer diagnoses as the effectiveness measure. Using breast cancer cases in the SEER data, the authors found that NBCAM events were effective in increasing November diagnosis in the mid-1990s but the effectiveness appeared to have tapered off later on. The authors speculated that given the long history of the NBCAM, the campaign effect could have become saturated over time. Belkar, Fiebig, Haas, and Viney (2006) explicitly modeled “awareness of screening” in an econometric analysis that explored cervical cancer screening in Australia. This study made an important point, because from a policy perspective it is important to differentiate between individuals who did not participate in screening by choice and those who were not screened due to lack of awareness. Empirically, screening awareness poses a potential selection bias. Using data from the 1995 National Health Survey in Australia, Belkar et al. (2006) reported that 4.3% of the study sample had never heard of a Pap test, and findings from a censored probit model showed that failure to account for screening awareness could generate inconsistent estimates and overestimate screening probabilities for the population. Given the demonstrated importance of screening awareness in understanding screening behaviors, the authors urged national health surveys that collect information on cancer screenings to also include questions regarding respondents’ awareness of various types of screening, as such information is not always collected.

Unlike many European countries that have organized screening programs, the fragmented health insurance system in the United States makes it impossible to establish such programs at the national level. In fact, lack of insurance coverage has been found to be a major barrier to cancer screening uptake. One option is to level the playing field by removing the financial barrier of screening for the uninsured or underinsured. For the underinsured, state mandates are legislative efforts at the state level that require private insurers to cover preventive care, including cancer screening. Different timing of phasing in insurance mandates across states creates a natural experiment environment, and several studies have employed difference-in-differences analyses to examine the impact of state mandates on participation in cancer screening. Findings from these studies have been mixed (Bitler & Carpenter, 2017; Hamman & Kapinos, 2016; Xu, Dowd, & Abraham, 2016). State mandates, however, have limited reach as they only regulate private insurance markets and self-insured health plans are typically exempted under the Employee Retirement Income Security Act (ERISA). For the uninsured, several policy initiatives have been implemented to provide them with insurance coverage. The health reform in Massachusetts brought near-universal insurance coverage to its state residents. Sabik and Bradley (2016) found that by expanding insurance coverage, the Massachusetts health reform was associated with increased use of breast and cervical cancer screening. Several provisions under the ACA are potentially beneficial for the use of preventive care. The two most noticeable provisions are a) a mandate that requires health insurance plans, including the self-insured and Medicaid, to cover preventive services with no cost-sharing for services with a grade A or B recommendation by the USPSTF, and b) to expand state Medicaid programs. It should be noted that although the ACA waives out-of-pocket costs for preventive services recommended by the USPSTF that carry an A or B rating, patients may still be financially responsible for the deductible, copayment, and coinsurance associated with downstream costs subsequent to an abnormal finding from screening, such as costs associated with diagnostic testing or biopsies. For states that chose to participate in Medicaid expansion, it provides a mechanism for the previously uninsured to become insured. Sabik and Adunlin (2017) reviewed empirical studies that examined the impact of the ACA on cancer screening and diagnosis. Of the 14 articles identified for the review, the findings were mixed, with stronger impact reported among individuals with lower education and income.

The role of screening guidelines deserves more attention in the economics of cancer prevention and control. From an information perspective, guidelines reduce the information asymmetry between consumers and physicians, as guidelines are publicly accessible. Throughout the years, there have been changes in the screening initiation age and the frequency of screening, and even the abandonment of certain screening modalities. Coverage decision, however, does not always follow screening guidelines. For example, although the USPSTF recommends biennial mammography for women at average risk of breast cancer, Medicare continues to cover annual mammography. In addition, although the USPSTF no longer advises PSA for prostate cancer screening, Medicare still reimburses for annual PSA. The cancer screening literature suggests that guidelines that tightened the age eligibility criteria for screening mammography had a null or relatively small effect in reducing screening among the age group (i.e., 40–49 years old) for whom screening was no longer recommended (Howard & Adams, 2012; Qin, Tangka, Guy, & Howard, 2017), whereas guidelines that recommended against PSA testing for prostate cancer seemed to have reduced PSA screening rates (Jemal et al., 2015). Research has also found that widening the eligibility criteria of reimbursement policies (e.g., Medicare extended the coverage of colonoscopy from high- to average-risk beneficiaries in July 2001) was effective in increasing screening utilization (Shih, Zhao, & Elting, 2006). It is therefore interesting to explore whether consumers’ decisions to participate in screening are more likely to be influenced by screening guidelines or insurance policies.

Future Direction and Challenges

Several trends in biomedical research around 2018 offer ample research opportunities for the economics of cancer prevention and control in the years to come.

Firstly, for the first time in medical history an effective screening modality for lung cancer has become available. Historically, preventive effort for lung cancer has largely focused on smoking cessation. The effectiveness of low-dose computed tomography (LDCT) in lung cancer screening among heavy smokers was established in the landmark National Lung Screening Trial (NLST) in 2011 (National Lung Screening Trial Research et al., 2011). Many professional societies and associations, including the USPSTF and the ACS, now recommend lung cancer screening with LDCT for high-risk smokers (Moyer & Force, 2014; Wender et al., 2013; Wood et al., 2012). The availability of LDCT for lung cancer screening offers an unprecedented opportunity to reduce lung cancer mortality but also brings many unique challenges from an economic perspective. The eligibility criteria for the NLST trial included being between 55 and 74 years of age and a heavy smoker (defined as having a 30+ pack-year history of smoking), either currently smoking or having quit within the last 15 years. Findings on the cost-effectiveness of LDCT for lung cancer screening are mixed, but the cost-effectiveness appears to hinge upon finding the right population to target for the screening program (Raymakers et al., 2016). Given the public’s fear of lung cancer and that LDCT is relatively inexpensive if one has to pay for it out of pocket, it is possible the availability of LDCT could motivate health-conscious non-smokers to seek lung cancer screening despite the lack of effectiveness of LDCT among non-smokers. Early evidence of such a spillover effect was reported in Huo, Shen, Volk, and Shih (2017).

Secondly, while cancer screening has traditionally been considered “a good thing,” a paradigm shift occurred around 2010 in that concerns regarding overtreatment and overdiagnosis resulting from screening have motivated professional societies and medical associations to take into consideration the harm–benefit trade-off when drafting screening guidelines (Mandelblatt et al., 2016; Nelson et al., 2016; Oeffinger et al., 2015). Overall, economic studies on the utilization of cancer screening have focused largely on how to increase screening uptake. The implicit assumption in these studies was the underuse of cancer screening. As the high cost of cancer care threatens the affordability of cancer care worldwide, cancer researchers from various disciplines are reflecting on what constitutes appropriate, responsible use of limited healthcare resources for cancer care (Shih et al., 2013; Sullivan et al., 2011). Since the turn of the 21st century, discussions have begun to shift toward identifying underuse, overuse, and misuse of medical technologies in cancer treatment and prevention. Evidence has suggested overuse of colonoscopy and mammography (Goodwin, Singh, Reddy, Riall, & Kuo, 2011; Tan, Kuo, Elting, & Goodwin, 2013) and possibly misuse of screening technologies for lung cancer (Huo et al., 2017). Future research should continue to explore optimal screening strategies that balance the benefits and harms of screening. Thinking in terms of health production function, the undesirable consequences of screening should be taken into consideration in economic theories related to cancer prevention and control because with the presence of false-positives and overdiagnosis, secondary prevention will not always increase the marginal productivity of cancer treatment.

Thirdly, as cancer treatment advances with breakthrough discoveries such as immunotherapy and chimeric antigen receptor (CAR) T-cell therapy, the interaction between advances in cancer treatment and the cost-effectiveness of screening must be explored in future research. An increasing number of researchers in the medical community have started sounding the alarm that as cancer treatment improves, the benefits of screening could diminish (Welch et al., 2016). This speculation was refuted in a recent modeling study. Birnbaum, Gadi, Markowitz, and Etzioni (2016) developed a microsimulation model to capture the improved survival benefit of more contemporary systemic therapies in breast cancer. They concluded that advances in systemic treatment in the United States did not substantially change the relative mortality effect of screening mammography. A working paper by Shih, Smith, Dong, Xu, and Shen (2017) used a microsimulation developed for women at average risk for breast cancer to evaluate the association between treatment advances and the cost-effectiveness of screening mammography. They found that for the most cost-effective screening strategy (i.e., a hybrid strategy that began annual screening at age 45 and switched to biennial between age 55 and 75), the incremental cost-effectiveness ratio increased from ~$69,000/QALY to over $71,000/QALY with advances in treatment. At the societal WTP of $50,000/QALY, the probability that the “no screening” strategy was the optimal strategy increased from 8% to 26% with treatment advances. This study tentatively concluded that treatment advances reduced the cost-effectiveness of screening mammography. Discussions on the interaction between cancer treatment and prevention are especially important for developing countries because the limited resources in these countries often force public health officials to choose among many competing priorities. In the extreme case, when a country has no capacity to treat cancer, it begs the question whether investment in primary prevention would be far more efficient than investing in secondary prevention.

Fourthly, while the economic literature on vaccines has traditionally been less relevant to cancer prevention, the role of vaccination as a primary prevention of cancer has become increasingly important. The International Agency for Research on Cancer classified infectious agents as carcinogenic to humans in 2009. It was estimated that over 16% of new cancer cases diagnosed worldwide in 2008 could be attributed to infections; HPV accounted for one-third of these cases (de Martel et al., 2012). More than 90% of cervical cancers were associated with HPV; therefore, HPV vaccination has been recommended for the prevention of cervical cancer by many medical associations (Saslow et al., 2016). CEA of HPV vaccines, especially for low- and middle-income countries, has been an active area of research in economic evaluation (Ekwunife et al., 2017). Estimates from the United States showed that HPV infection was associated not only with the vast majority of cervical cancer cases, but also with a large percentage of vulva, vaginal, penile, anal, and oropharyngeal cancers (Saraiya et al., 2015). The epidemiology of HPV infection in cancer makes HPV vaccinations an interesting topic for a new area of research in economics known as “economic epidemiology” (Pierre-Yves & Philipson, 1996).

Lastly, precision medicine, defined by the US National Institutes of Health as “a revolutionary approach for disease prevention and treatment that takes into account individual differences in lifestyle, environment, and biology,” is believed to be the future of cancer care. The concept of “precision prevention” involves individualized risk assessment using advanced medical technologies so as to develop personalized prevention strategies (Vineis & Wild, 2017). In “Future cancer research priorities in the USA,” a paper commissioned by Lancet Oncology (Jaffee et al., 2017), leaders in cancer research expressed optimism about the potential of precision prevention. The viewpoint was that technologies today are capable of identifying the earliest genetic changes in cells that can promote cancer development, which then allows targeted chemoprevention and vaccination for the population with the specific genetic disposition. The economics aspect of precision prevention is largely unexplored and will be an exciting area for future research on the economics of cancer prevention and control.

Acknowledgments

The author thanks an anonymous referee for his/her insightful comments, and Dr. Gary Deyter, technical writer from the Department of Health Services Research at the University of Texas MD Anderson Cancer Center, for proofreading the manuscript. Some of the work reported here was supported by NCI P30 CA016672 and R01 CA207216.

Further Reading

The following book chapters cover the topic of economics of prevention. Although the materials presented in these two book chapters are not specific to cancer prevention, they provide an excellent introduction to the basic concepts and theories relevant to the economics of cancer prevention and control.

Folland, S., Goodman, A. C., & Stano, M. (2016). The health economics of bads. In S. Folland, A. C. Goodman, & M. Stano (Eds.), The economics of health and health care (pp. 513–530). New York, NY: Routledge.Find this resource:

Kenkel, D. S. (2000). Prevention. In A. J. Culyer & J. P. Newhouse (Eds.), Handbook of health economics (Vol. 1B, pp. 1675–1720). Amsterdam, The Netherlands: Elsevier.Find this resource:

To gain knowledge of various sources of biases commonly encountered in the estimation of the effect of screening on cancer mortality, please read:

Duffy, S. W., Nagtegaal, I. D., Wallis, M., Cafferty, F. H., Houssami, N., Warwick, J., . . . Lawrence, G. (2008). Correcting for lead time and length bias in estimating the effect of screen detection on cancer survival. American Journal of Epidemiology, 168(1), 98–104.Find this resource:

This book is an excellent reference for empirical studies aiming to estimate potential cost savings from prevention interventions:

Manning, W. G., Keeler, E. B., Newhouse, J. P., Sloss, E. M., & Wasserman, J. (1991). The costs of poor health habits. Cambridge, MA: Harvard University Press.Find this resource:

This paper is an exemplar of solid health economic research on the topic of cancer screening:

Lange, F. (2011). The role of education in complex health decisions: Evidence from cancer screening. Journal of Health Economics, 30(1), 43–54.Find this resource:

References

Baker, L. C., Phillips, K. A., Haas, J. S., Liang, S. Y., & Sonneborn, D. (2004). The effect of area HMO market share on cancer screening. Health Services Research, 39(6 Pt. 1), 1751–1772.Find this resource:

Belkar, R., Fiebig, D. G., Haas, M., & Viney, R. (2006). Why worry about awareness in choice problems? Econometric analysis of screening for cervical cancer. Health Economics, 15(1), 33–47.Find this resource:

Best, A., Hiatt, R. A., Cameron, R., Rimer, B. K., & Abrams, D. B. (2003). The evolution of cancer control research: An international perspective from Canada and the United States. Cancer Epidemiology, Biomarkers & Prevention, 12(8), 705–712.Find this resource:

Birnbaum, J., Gadi, V. K., Markowitz, E., & Etzioni, R. (2016). The effect of treatment advances on the mortality results of breast cancer screening trials: A microsimulation model. Annals of Internal Medicine, 164(4), 236–243.Find this resource:

Bitler, M. P., & Carpenter, C. S. (2017). Effects of state cervical cancer insurance mandates on pap test rates. Health Services Research, 52(1), 156–175.Find this resource:

Chapman, S. (2003). Fresh row over prostate screening. BMJ, 326(7389), 605.Find this resource:

Chen, H. Y., Kessler, C. L., Mori, N., & Chauhan, S. P. (2012). Cervical cancer screening in the United States, 1993–2010: Characteristics of women who are never screened. Journal of Women’s Health (Larchmt), 21(11), 1132–1138.Find this resource:

Chowdhury, P. P., Mawokomatanda, T., Xu, F., Gamble, S., Flegel, D., Pierannunzi, C., . . . Town, M. (2016). Surveillance for certain health behaviors, chronic diseases, and conditions, access to health care, and use of preventive health services among states and selected local areas: Behavioral risk factor surveillance system, United States, 2012. MMWR Surveillance Summary, 65(4), 1–142.Find this resource:

Clark, C. R., Tosteson, T. D., Tosteson, A. N. A., Onega, T., Weiss, J. E., Harris, K. A., & Haas, J. S. (2017). Diffusion of digital breast tomosynthesis among women in primary care: Associations with insurance type. Cancer Medicine, 6(5), 1102–1107.Find this resource:

Colditz, G. A., & Wolin, K. Y. (2011). Transdisciplinary training in cancer prevention: Reflections on two decades of training. Journal of Cancer Education, 26(3), 586–590.Find this resource:

Cutler, D. M. (2008). Are we finally winning the war on cancer? Journal of Economic Perspectives, 22(4), 3–26.Find this resource:

de Koning, H. J., Meza, R., Plevritis, S. K., ten Haaf, K., Munshi, V. N., Jeon, J., . . . McMahon, P. M. (2014). Benefits and harms of computed tomography lung cancer screening strategies: A comparative modeling study for the U.S. Preventive Services Task Force. Annals of Internal Medicine, 160(5), 311–320.Find this resource:

de Martel, C., Ferlay, J., Franceschi, S., Vignat, J., Bray, F., Forman, D., & Plummer, M. (2012). Global burden of cancers attributable to infections in 2008: A review and synthetic analysis. Lancet Oncology, 13(6), 607–615.Find this resource:

Duffy, S. W., Nagtegaal, I. D., Wallis, M., Cafferty, F. H., Houssami, N., Warwick, J., . . . Lawrence, G. (2008). Correcting for lead time and length bias in estimating the effect of screen detection on cancer survival. American Journal of Epidemiology, 168(1), 98–104.Find this resource:

Ehrenreich, B. (2010). The uproar over new breast cancer screening guidelines.

Ehrlich, I., & Becker, G. (1972). Market insurance, self-insurance, and self-protection. Journal of Political Economy, 80, 623–649.Find this resource:

Ekwunife, O. I., O’Mahony, J. F., Gerber Grote, A., Mosch, C., Paeck, T., & Lhachimi, S. K. (2017). Challenges in cost-effectiveness analysis modelling of HPV vaccines in low- and middle-income countries: A systematic review and practice recommendations. PharmacoEconomics, 35(1), 65–82.Find this resource:

Folland, S., Goodman, A. C., & Stano, M. (2016). The health economics of bads. In S. Folland, A. C. Goodman, & M. Stano (Eds.), The economics of health and health care (pp. 513–530). New York, NY: Routledge.Find this resource:

Giuliano, A. E., Boolbol, S., Degnim, A., Kuerer, H., Leitch, A. M., & Morrow, M. (2007). Society of Surgical Oncology: Position statement on prophylactic mastectomy. Approved by the Society of Surgical Oncology Executive Council, March 2007. Annals of Surgical Oncology, 14(9), 2425–2427.Find this resource:

Goodwin, J. S., Singh, A., Reddy, N., Riall, T. S., & Kuo, Y. F. (2011). Overuse of screening colonoscopy in the Medicare population. Archives of Internal Medicine, 171(15), 1335–1343.Find this resource:

Grossman, M. (1972). On the concept of health capital and the demand for health. Journal of Political Economy, 80(2), 223–255.Find this resource:

Grossman, M., & Rand, E. (1974). Consumer incentives for health services in chronic illness. In S. J. Mushkin (Ed.), Consumer incentive for health care (pp. 114–151). New York, NY: Milbank Memorial Fund.Find this resource:

Gyrd-Hansen, D., & Sogaard, J. (2001). Analysing public preferences for cancer screening programmes. Health Economics, 10(7), 617–634.Find this resource:

Habbema, J. D., Schechter, C. B., Cronin, K. A., Clarke, L. D., & Feuer, E. J. (2006). Modeling cancer natural history, epidemiology, and control: Reflections on the CISNET breast group experience. Journal of the National Cancer Institute Monographs, 36, 122–126.Find this resource:

Hamman, M. K., & Kapinos, K. A. (2016). Colorectal cancer screening and state health insurance mandates. Health Economics, 25(2), 178–191.Find this resource:

Henderson, L. M., O’Meara, E. S., Braithwaite, D., Onega, T., & Breast Cancer Surveillance Consortium (2015). Performance of digital screening mammography among older women in the United States. Cancer, 121(9), 1379–1386.Find this resource:

Hey, J. D., & Patel, M. S. (1983). Prevention and cure? Or: is an ounce of prevention worth a pound of cure? Journal of Health Economics, 2(2), 119–138.Find this resource:

Hiatt, R. A., & Rimer, B. K. (2006). Principles and applications of cancer prevention and control interventions. In D. Schottenfeld & J. F. Fraumeni (Eds.), Cancer epidemiology and prevention (3rd ed., pp. 1283–1291). New York, NY: Oxford University Press.Find this resource:

Howard, D. H., & Adams, E. K. (2012). Mammography rates after the 2009 US Preventive Services Task Force breast cancer screening recommendation. Preventive Medicine, 55(5), 485–487.Find this resource:

Huo, J., Shen, C., Volk, R. J., & Shih, Y. T. (2017). Use of CT and chest radiography for lung cancer screening before and after publication of screening guidelines: Intended and unintended uptake. JAMA Internal Medicine, 177(3), 439–441.Find this resource:

Husereau, D., Drummond, M., Petrou, S., Carswell, C., Moher, D., Greenberg, D., . . . Force, I. H. E. E. P. G.-C. G. R. P. T. (2013). Consolidated Health Economic Evaluation Reporting Standards (CHEERS)—explanation and elaboration: A report of the ISPOR Health Economic Evaluation Publication Guidelines Good Reporting Practices Task Force. Value Health, 16(2), 231–250.Find this resource:

Jacobsen, G. D., & Jacobsen, K. H. (2011). Health awareness campaigns and diagnosis rates: Evidence from National Breast Cancer Awareness Month. Journal of Health Economics, 30(1), 55–61.Find this resource:

Jaffee, E. M., Dang, C. V., Agus, D. B., Alexander, B. M., Anderson, K. C., Ashworth, A., . . . Yung, A. (2017). Future cancer research priorities in the USA: A Lancet Oncology Commission. Lancet Oncology, 18(11), e653–e706.Find this resource:

Jemal, A., Fedewa, S. A., Ma, J., Siegel, R., Lin, C. C., Brawley, O., & Ward, E. M. (2015). Prostate cancer incidence and PSA testing patterns in relation to USPSTF screening recommendations. JAMA, 314(19), 2054–2061.Find this resource:

Kenkel, D. S. (1994). The demand for preventive medical care. Applied Economics, 26, 313–325.Find this resource:

Kenkel, D. S. (2000). Prevention. In A. J. Culyer & J. P. Newhouse (Eds.), Handbook of health economics (Vol. 1B, pp. 1675–1720). Amsterdam, The Netherlands: Elsevier.Find this resource:

Knudsen, A. B., Zauber, A. G., Rutter, C. M., Naber, S. K., Doria-Rose, V. P., Pabiniak, C., . . . Kuntz, K. M. (2016). Estimation of benefits, burden, and harms of colorectal cancer screening strategies: Modeling study for the US Preventive Services Task Force. JAMA, 315(23), 2595–2609.Find this resource:

Lange, F. (2011). The role of education in complex health decisions: Evidence from cancer screening. Journal of Health Economics, 30(1), 43–54.Find this resource:

Mandelblatt, J. S., Stout, N. K., Schechter, C. B., van den Broek, J. J., Miglioretti, D. L., Krapcho, M., . . . Cronin, K. A. (2016). Collaborative modeling of the benefits and harms associated with different U.S. breast cancer screening strategies. Annals of Internal Medicine, 164(4), 215–225.Find this resource:

Manning, W. G., Keeler, E. B., Newhouse, J. P., Sloss, E. M., & Wasserman, J. (1991). The costs of poor health habits. Cambridge, MA: Harvard University Press.Find this resource:

Milligan, M. A., Bohara, A. K., & Pagan, J. A. (2010). Assessing willingness to pay for cancer prevention. International Journal of Health Care Finance and Economics, 10(4), 301–314.Find this resource:

Moyer, V. A., & Force, U. S. P. S. T. (2014). Screening for lung cancer: U.S. Preventive Services Task Force recommendation statement. Annals of Internal Medicine, 160(5), 330–338.Find this resource:

National Lung Screening Trial Research, T., Aberle, D. R., Adams, A. M., Berg, C. D., Black, W. C., Clapp, J. D., . . . Sicks, J. D. (2011). Reduced lung-cancer mortality with low-dose computed tomographic screening. New England Journal of Medicine, 365(5), 395–409.Find this resource:

NCI. SEER Program Overview. Bethesda, MD: NIH.Find this resource:

Nelson, H. D., Pappas, M., Cantor, A., Griffin, J., Daeges, M., & Humphrey, L. (2016). Harms of breast cancer screening: Systematic review to update the 2009 U.S. Preventive Services Task Force recommendation. Annals of Internal Medicine, 164(4), 256–267.Find this resource:

Oeffinger, K. C., Fontham, E. T., Etzioni, R., Herzig, A., Michaelson, J. S., Shih, Y. C., . . . American Cancer, S. (2015). Breast cancer screening for women at average risk: 2015 guideline update from the American Cancer Society. JAMA, 314(15), 1599–1614.Find this resource:

Phillips, K. A., Haas, J. S., Liang, S. Y., Baker, L. C., Tye, S., Kerlikowske, K., . . . Spetz, J. (2004). Are gatekeeper requirements associated with cancer screening utilization? Health Services Research, 39(1), 153–178.Find this resource:

Pierre-Yves, G., & Philipson, T. (1996). Rational epidemics and thier public control. International Economic Review, 37(3), 603–624.Find this resource:

Qin, X., Tangka, F. K., Guy, G. P., Jr., & Howard, D. H. (2017). Mammography rates after the 2009 revision to the United States Preventive Services Task Force breast cancer screening recommendation. Cancer Causes & Control, 28(1), 41–48.Find this resource:

Rafia, R., Brennan, A., Madan, J., Collins, K., Reed, M. W., Lawrence, G., . . . Wyld, L. (2016). Modeling the cost-effectiveness of alternative upper age limits for breast cancer screening in England and Wales. Value Health, 19(4), 404–412.Find this resource:

Raymakers, A. J. N., Mayo, J., Lam, S., FitzGerald, J. M., Whitehurst, D. G. T., & Lynd, L. D. (2016). Cost-effectiveness analyses of lung cancer screening strategies using low-dose computed tomography: A systematic review. Applied Health Economics and Health Policy, 14(4), 409–418.Find this resource:

Ronckers, E. T., Groot, W., & Ament, A. J. (2005). Systematic review of economic evaluations of smoking cessation: Standardizing the cost-effectiveness. Medical Decision Making, 25(4), 437–448.Find this resource:

Rosenfield, P. L. (1992). The potential of transdisciplinary research for sustaining and extending linkages between the health and social sciences. Social Science & Medicine, 35(11), 1343–1357.Find this resource:

Rueth, N. M., McMahon, M., Arrington, A. K., Swenson, K., Leach, J., & Tuttle, T. M. (2011). Preoperative risk assessment among women undergoing bilateral prophylactic mastectomy for cancer risk reduction. Annals of Surgical Oncology, 18(9), 2515–2520.Find this resource:

Sabik, L. M., & Adunlin, G. (2017). The ACA and cancer screening and diagnosis. Cancer Journal, 23(3), 151–162.Find this resource:

Sabik, L. M., & Bradley, C. J. (2016). The impact of near-universal insurance coverage on breast and cervical cancer screening: Evidence from Massachusetts. Health Economics, 25(4), 391–407.Find this resource:

Sambamoorthi, U., & McAlpine, D. D. (2003). Racial, ethnic, socioeconomic, and access disparities in the use of preventive services among women. Preventive Medicine, 37(5), 475–484.Find this resource:

Saraiya, M., Unger, E. R., Thompson, T. D., Lynch, C. F., Hernandez, B. Y., Lyu, C. W., . . .Workgroup, H. P. V. T. o. C. (2015). US assessment of HPV types in cancers: Implications for current and 9-valent HPV vaccines. Journal of the National Cancer Institute, 107(6), djv086.Find this resource:

Saslow, D., Andrews, K. S., Manassaram-Baptiste, D., Loomer, L., Lam, K. E., Fisher-Borne, M., . . . American Cancer Society Guideline Development, G. (2016). Human papillomavirus vaccination guideline update: American Cancer Society guideline endorsement. CA: A Cancer Journal for Clinicians, 66(5), 375–385.Find this resource:

Schlichte, M. J., & Guidry, J. (2015). Current cervical carcinoma screening guidelines. Journal of Clinical Medicine, 4(5), 918–932.Find this resource:

Shih, Y. C., Ganz, P. A., Aberle, D., Abernethy, A., Bekelman, J., Brawley, O., . . . Schnipper, L. (2013). Delivering high-quality and affordable care throughout the cancer care continuum. Journal of Clinical Oncology, 31(32), 4151–4157.Find this resource:

Shih, Y. C., Zhao, L., & Elting, L. S. (2006). Does Medicare coverage of colonoscopy reduce racial/ethnic disparities in cancer screening among the elderly? Health Affairs (Millwood), 25(4), 1153–1162.Find this resource:

Shih, Y. C. T., Dong, W., Xu, Y., & Yu, S. (2017). Use of relative risk ratios to present uncertainty in microsimulation models with multiple comparators: An application to breast cancer screening strategies. Value in Health, 20(5), A324.Find this resource:

Shih, Y. C. T., Doug, W., Xu, Y., & Shen, Y. (2019). Assessing the cost-effectiveness of updated breast cancer screening guidelines for average-risk women. Value in Health.Find this resource:

Shih, Y. C. T., Smith, B., Dong, W., Xu, Y., & Shen, Y. (2017). Do advances in breast cancer treatment render screening mammography not cost-effective for average-risk women? Value in Health, 20(5), A133.Find this resource:

Simon, K., Soni, A., & Cawley, J. (2017). The impact of health insurance on preventive care and health behaviors: Evidence from the first two years of the ACA Medicaid expansions. Journal of Policy Analysis and Management, 36(2), 390–417.Find this resource:

Skally, M., Hanly, P., & Sharp, L. (2013). Cost effectiveness of fecal DNA screening for colorectal cancer: A systematic review and quality appraisal of the literature. Applied Health Economics and Health Policy, 11(3), 181–192.Find this resource:

Spratt, J. S. (1981). The primary and secondary prevention of cancer. Journal of Surgical Oncology, 18(3), 219–230.Find this resource:

Sullivan, R., Peppercorn, J., Sikora, K., Zalcberg, J., Meropol, N. J., Amir, E., . . . Aapro, M. (2011). Delivering affordable cancer care in high-income countries. Lancet Oncology, 12(10), 933–980.Find this resource:

Sun, E., Jena, A. B., Lakdawalla, D., Reddy, C., & Philipson, T. J. (2010). The contributions of improved therapy and early detection to cancer survival gains, 1988–2000. Forum for Health Economics & Policy, 13(2), 1–20.Find this resource:

Sweet, A., Lee, D., Gairy, K., Phiri, D., Reason, T., & Lock, K. (2011). The impact of CT colonography for colorectal cancer screening on the UK NHS: Costs, healthcare resources and health outcomes. Applied Health Economics and Health Policy, 9(1), 51–64.Find this resource:

Tan, A., Kuo, Y. F., Elting, L. S., & Goodwin, J. S. (2013). Refining physician quality indicators for screening mammography in older women: Distinguishing appropriate use from overuse. Journal of the American Geriatrics Society, 61(3), 380–387.Find this resource:

Van Houtven, G., Sullivan, M. B., & Dockins, C. (2008). Cancer premiums and latency effects: A risk tradeoff approach for valuing reductions in fatal cancer risks. Journal of Risk and Uncertainty, 36, 179–199.Find this resource:

Vineis, P., & Wild, C. P. (2017). The science of precision prevention of cancer. Lancet Oncology, 18(8), 997–998.Find this resource:

Viscusi, W. K., & Aldy, J. E. (2003). The value of a statistical life: A critical review of market estimates throughout the world. Journal of Risk and Uncertainty, 27(1), 5–76.Find this resource:

Viscusi, W. K., Huber, J., & Bell, J. (2014). Assessing whether there is a cancer premium for the value of a statistical life. Health Economics, 23(4), 384–396.Find this resource:

von Karsa, L., Arbyn, M., De Vuyst, H., Dillner, J., Dillner, L., Franceschi, S., . . . Anttila, A. (2015). European guidelines for quality assurance in cervical cancer screening: Summary of the supplements on HPV screening and vaccination. Papillomavirus Research, 1, 22–31.Find this resource:

Welch, H. G., Prorok, P. C., O’Malley, A. J., & Kramer, B. S. (2016). Breast-cancer tumor size, overdiagnosis, and mammography screening effectiveness. New England Journal of Medicine, 375(15), 1438–1447.Find this resource:

Wender, R., Fontham, E. T., Barrera, E., Jr., Colditz, G. A., Church, T. R., Ettinger, D. S., . . . Smith, R. A. (2013). American Cancer Society lung cancer screening guidelines. CA: A Cancer Journal for Clinicians, 63(2), 107–117.Find this resource:

White, A., Thompson, T. D., White, M. C., Sabatino, S. A., de Moor, J., Doria-Rose, P. V., . . . Richardson, L. C. (2017). Cancer screening test use—United States, 2015. Morbidity and Mortality Weekly Report, 66, 201–206.Find this resource:

Whynes, D. K. (2004). Economics of cancer screening programs. Expert Review of Pharmacoeconomics & Outcomes Research, 4(1), 61–69.Find this resource:

Wood, D. E., Eapen, G. A., Ettinger, D. S., Hou, L., Jackman, D., Kazerooni, E., . . . Yang, S. C. (2012). Lung cancer screening. Journal of the National Comprehensive Cancer Network, 10(2), 240–265.Find this resource:

Xu, W. Y., Dowd, B., & Abraham, J. (2016). Lessons from state mandates of preventive cancer screenings. European Journal of Health Economics, 17(2), 203–215.Find this resource:

Yamey, G., & Wilkes, M. (2002). The PSA storm. BMJ, 324, 431.Find this resource:

Zauber, A. G. (2010). Cost-effectiveness of colonoscopy. Gastrointestinal Endoscopy Clinics of North America, 20(4), 751–770.Find this resource: