Cancer screening and the periodic health examination


  • See referenced original article on pages 000–000, this issue.


In this issue of Cancer, Chen et al. report outcomes from an integrated model of multiphasic chronic disease screening in Taiwan. The design and outcomes of the Keelung Community-based Integrated Screening program are noteworthy for several reasons and should reopen the debate not only regarding the worth, but also the protocol for periodic preventive health examinations.

See also pages 000–000.

In this issue of Cancer, investigators from the Taiwan Community-based Integrated Screening Group report outcomes from an integrated model of multiphasic chronic disease screening in Keelung, Taiwan.1 Their findings represent an important, modern demonstration of a practical and productive approach to preventive health and case finding. The design and outcomes of the Keelung Community-based Integrated Screening (KCIS) program are noteworthy for several reasons and should reopen the debate not only about the worth, but also the protocol of periodic preventive health examinations.

The underlying philosophy of the KCIS program was the potential for the simultaneous ascertainment of two or more asymptomatic chronic conditions, identification of multiple risk factors for metabolic syndrome, enhanced attendance for screening, and more efficient use of resources. Other “linkage” strategies might be equally effective, or be more appropriate for other target audiences. The program focused on five neoplasms (breast, cervical, colorectal, liver, and oral cancers) and three nonneoplastic chronic conditions (diabetes, hypertension, and elevated cholesterol), and education concerning metabolic syndrome. The decision to use the Papanicolaou (Pap) smear registry as a base to build the program was both innovative and practical because screening for cervical cancer is well established and because women can play a pivotal role in monitoring and attending to the health status of other family members. The investigators increased adherence in nonattenders to the Pap smear program by incorporating cervical cancer screening into a broader testing regimen, thus increasing the screening rate in nonattenders who may have been averse to gynecologic screening, with incentives for additional preventive care opportunities for themselves and other family members. Telephone invitations to 30,384 women resulted in 42,387 participants (24,469 women and 17,918 dependents) being seen at 257 health centers. The program increased adherence to cervical cancer screening by 25% (from 55.5% to 80.5%). For all cancer screening, the overall detection of asymptomatic neoplasia was 16 per 1000. Significant numbers of individuals with previously unknown diabetes, metabolic syndrome, hyperlipidemia, and hypertension also were identified. Based on these findings, the investigators estimated that 959 deaths will be averted over a 10-year period after full implementation of the program. The opportunities for research also were not overlooked. The investigators observed that individuals diagnosed with metabolic syndrome were at an elevated risk for colorectal and liver cancers, and other statistically significant associations between nonmalignant chronic disease and neoplasia were identified. These outcomes in conjunction with information obtained from the questionnaire, including lifestyle factors, family history, menstrual and reproductive history, and the presence of disease suggest the potential for establishing comprehensive profiles for future behavior risk factor surveillance. What are the implications of the KCIS program findings for current or alternative approaches to preventive health in the U.S.?

In the U.S., the importance of preventive care is well appreciated by the population, health care professionals, and health agencies, especially in the presence of a significant, ongoing burden of chronic disease. Federal health agencies issue annual reports, and there are established goals and objectives for the nation's health. In 2003, the National Center for Health Statistics (NCHS) issued its 27th report on the health status of the U.S., which is submitted by the Secretary of Health and Human Services to the President and Congress in compliance with Section 308 of the Public Health Service Act.2 The Secretary's report highlighted that too many Americans still smoke cigarettes and are physically inactive, and that the prevalence of overweight and obesity in adults had risen to 65% in 1999–2000, all of which are factors that confer a significant risk for developing hypertension, heart disease, diabetes, and some cancers. Healthy People 2010, the prevention agenda for the nation, currently lists 476 specific objectives in 28 focus areas, which include cancer, diabetes, heart disease, nutrition and overweight, physical activity and fitness, tobacco use, and access to quality health services.3 In the private arena, the National Committee for Quality Assurance (NCQA) supports the Health Plan Employer Data and Information Set (HEDIS®) so that purchasers and consumers can measure the performance of managed care plans on key public health indicators, including breast, cervical, and colorectal cancer screening; controlling blood pressure; comprehensive diabetes care; and medical assistance with smoking cessation.4 The Agency for Healthcare Research and Quality (AHRQ) currently is the home to the U.S. Preventive Services Task Force (USPSTF), which was first convened in 1984 to evaluate clinical research in order to assess the value of preventive measures.5 The USPSTF has reviewed, and periodically updated, more than 200 preventive services offered in primary care settings, and presently recommends routine screening for cervical, breast, and colorectal cancers and screening for hypertension, lipid disorders, and obesity, and strongly recommends screening for tobacco use and the provision of cessation interventions for those who use tobacco products.6 In addition, AHRQ also supports Put Prevention Into Practice (PPIP), a program of provider and patient tools to increase the appropriate use of clinical preventive services.5 The American Cancer Society (ACS) has guidelines for screening for breast, cervical, and colorectal cancers and, consistent with other organizations, recommends that men should have an opportunity for shared decision making with regard to testing for early prostate cancer detection.7 The ACS also has made recommendations for nutrition and physical activity.8

The investment in problem definition and goal setting is considerable, and there is broad acceptance of these long-range goals and guidelines to improve the health of the nation. It also is reasonable to say that there is a sense of urgency regarding existing and emerging challenges that affect the burden of chronic disease, including chronic disease management in diagnosed patients, identifying the sizable prevalence of individuals with undiagnosed chronic conditions, and identifying the larger number of individuals whose elevated risk for developing a chronic condition could be altered through tailored interventions. Consider the following. In 2000, deaths from malignant neoplasms, heart disease, cerebrovascular disease, and diabetes mellitus cost Americans 18.8 million person-years of life lost.9 These 4 diseases alone accounted for 81% of all premature deaths that year, with the largest share attributable to cancer (8.3 million person-years), and nearly half of the premature mortality from cancer attributable to deaths from cancers of the lung and bronchus, breast, colon and rectum, and cervix. Heart disease accounts for nearly as many person-years of life lost (7.8 million person-years) as cancer. The annual direct and indirect cost of these diseases is estimated to be in the hundreds of billions of dollars.3 According to the Centers for Disease Control and Prevention (CDC), 29% of diabetics in the U.S. currently are undiagnosed, and approximately 6.1% of the population has impaired fasting glucose levels.10 Approximately one in four adults is hypertensive, but the majority do not have their high blood pressure under control.3 Greater than 90 million adults have elevated cholesterol levels, and greater than 50 million have cholesterol levels that require medical advice and treatment.3 Recent estimates from the Third National Health and Nutrition Examination Survey indicate that among insured individuals, 28.6% of adults with hypertension were undiagnosed, as were 51.2% of adults with hypercholesterolemia.11 Although screening rates for breast and cervical cancer are relatively high, a significant percentage of the age-appropriate population does not undergo regular cancer screening, and rates of colorectal cancer screening remain discouragingly low.12, 13 Approximately 33% of breast and cervical cancers, and nearly 67% of colorectal cancers, are diagnosed at an advanced stage.9

The sheer volume of potential preventive care tasks accentuates the challenges facing primary care clinicians today. A recent article by Yarnall et al. noted that a clinician with an average panel of 2500 patients would need to devote 7.4 hours per day in an average work year to fully satisfy the USPSTF recommendations.14 These findings have been viewed as revealing the futility of meeting the full scope of preventive health recommendations. Yet, the challenge is largely futile because the current model for the delivery of preventive care emphasizes finding opportunities for prevention during illness encounters (often called opportunistic preventive care), a strategy that emerged to replace the model of the annual physical examination.

A consensus had evolved, beginning in the mid-19th century, that it was advisable to have an annual checkup.15 However, as Han noted in a recent history of the objectives of the periodic health examination,15 several comprehensive reviews of preventive services that were conducted in the 1970s and 1980s, most notably the report by the Canadian Task Force on the Periodic Health Examination, challenged the value of the annual examination.16–20 Han describes what resulted as an empiric purging of conventional procedures and the logic for periodic encounters. What was left in place was a “radically minimalist” set of procedures that could be performed, not at a preventive health encounter, but “opportunistically” during physician visits for illness. This logic of “fitting prevention into practice” still exists today, and although some preventive care takes place, the present model is hopelessly inefficient and ineffective as a formal strategy for health promotion and disease prevention. The failure of this encounter-based, primary care-based approach to adequately increase screening, risk factor modification counseling, and chronic disease management has been highlighted in numerous studies.21–24

When one considers the barriers to fulfilling preventive care opportunistically, it is rather remarkable that the U.S. achieves even current levels of preventive care. Current patterns of preventive care have been influenced by provider routines that have evolved over time, public demand, mandated benefits, and pressures external to the office setting, such as HEDIS and the widespread use of capitated payments.4 However, as an example, under this model the patient may be referred for breast cancer screening, but cervical cancer screening, colorectal cancer screening, and counseling concerning obesity may be left for the next encounter, if ever.

In the fractionated “nonsystem” that characterizes the delivery of health care in the U.S., a kind of “blame game” emerges. We blame primary care clinicians for their failure to achieve higher rates of preventive care delivery, and we blame individuals for unrealistic expectations. This blame is misplaced and unfair, is unlikely to contribute to improved care, and likely contributes to a defensive posture toward public health goals and even patients. A recent article in The New York Times observed that for those physicians who practice evidence-based medicine, there was not a more inviting target than the annual physical examination although the demanding public had not gotten the message that the examination was largely obsolete.25 In a survey of 500 adults, Schwartz et al. observed that nearly 9 in 10 believed cancer screening was nearly always a good idea. The authors expressed concern that the public's enthusiasm for cancer screening was not dampened by the prospects of false-positive test results.26 As a further demonstration of the public's uncritical acceptance of screening, Schwartz et al.26 described full-body scans in terms more glowing than those usually seen in current advertisements from free-standing centers, and then were incredulous that nearly 3 in 4 respondents would choose the examination over $1000 in cash. (The authors described a total-body computed tomography (CT) scan as a three-dimensional look inside [your] body using a CT scanner, which can provide a very detailed picture of the lungs, liver, heart, and other internal organs, as well as bones and arteries. They also added that a total body scan could detect many diseases such as cancer before they could be found by routine checkups and that the body scan was quick and painless.26 Based on this description, it is more interesting that 27% of the respondents did not want to undergo the examination.) Oboler et al. measured public expectations and attitudes regarding annual physical examinations and testing and found that approximately 67% of respondents believed an annual physical examination was necessary in addition to regular care.27 Interest in undergoing the examination declined by approximately 50% if an out-of-pocket charge of $150 were applied. The authors expressed concern that such a high percentage of the public favors the idea of an annual checkup, observing that it has little screening value,28 and they also noted that the public perceives value in specific tests that currently are not recommended by the USPSTF. Letters in response to the article also chided the pubic for demanding tests when cost was no object,29 and expecting that health insurance should pay for prevention when it truly only was intended to decrease the risk of catastrophic loss by spreading it over many people.30 However, in an editorial that accompanied the article, Laine cited numerous reports showing that individuals who undergo annual checkups have higher rates of screening and other preventive services compared with individuals who only visit a physician during illness. Dr. Laine specifically notes that an annual examination provides a specific time and place for prevention(emphasis added).31

Laine's point is central to the expectation that we will continue to have substantial unmet needs in preventive care unless we create dedicated opportunities for a preventive health encounter. In the U.S. we presently encourage primary care physicians to find the time to insert at least some preventive care into acute and chronic care visits, but a model that depends on “if-time-permits” is doomed to fail. Questioning the logic for the annual checkup has resulted in a situation where there is no consensus recommendation for a periodic preventive health encounter at any interval. However, arguing that the annual checkup is not supported by evidence misses the point entirely, because it sidesteps that question of what interval and panel of preventive health tests and counseling based on age/gender/risk for healthy individuals would be effective. Preventive health care in the U.S. holds great promise but numerous personal, cultural, and systemic barriers must be overcome. Some barriers are more challenging than others, and it behooves us to address those that can be dealt with structurally so that we can concentrate on the real challenges. Setting aside the national disgrace that greater than 40 million Americans have no health insurance, even for insured Americans, inconsistent policies related to the coverage of preventive health care (particularly health maintenance visits) creates disincentives for individuals to seek care and for physicians to focus on prevention during visits.32, 33

The commitment to improve the health of the public by providing appropriate preventive care to more people is widespread; agencies at all levels of government, not-for-profit health organizations, employers, health care providers, and members of the public share a desire to make an impact on the burden of chronic disease. Yet, many well-meaning efforts fail to take advantage of one of the basic realities of preventive care. The single most important determinant of whether a patient does or does not receive preventive care is whether they receive a recommendation from a trusted primary health care provider whom they view as their regular source of care.34, 35 Furthermore, in several studies of prevention in primary care, the strongest predictor of whether a patient received a recommendation for preventive care was determined to be whether a health maintenance visit was performed.23, 36 The KCIS study in Taiwan demonstrates that if patients are encouraged to have prevention visits that are designed to address multiple health issues in a primary care setting, a significant amount of preventive care occurs, which eventually will translate into reduced premature death and disability.1 The KCIS program also creates a platform for the facilitation of health promotion after the introduction of a series of screening programs. Although this particular model might not be easily adopted in the U.S., there is no reason why high rates of population screening and counseling could not take place within the current framework of primary care.

Despite the consistent observation that individuals who have preventive care visits receive more preventive care, and this is particularly true for cancer screening,21–23 to our knowledge there is virtually no consensus, no simple set of age-specific and gender-specific recommendations, for the periodicity and conduct of periodic preventive health encounters. A common, underlying thread in many of the current discussions concerning preventive health encounters is that the logic supporting an annual examination without a specific clinical protocol truly cannot be defended. However, neither can the alternative, which promotes the idea that preventive care should occur on the occasion when there is a coincidence of encounters and interests between patients and physicians.

The time has come to reinvigorate thought about how to tap the potential of health maintenance visits. Evidence-based guideline groups should address the elements that are appropriate to include in periodic health maintenance visits, and define a schedule for the periodicity of these visits based on age, gender, and other relevant considerations. It is particularly critical to identify which aspects of the old annual visit model need to be abandoned. For example, some of the time currently devoted to physical examinations (including listening to the heart and lungs in younger asymptomatic patients) most likely would better be devoted to evidence-based activity. Clinicians need to be reimbursed for preventive care, including visits devoted exclusively to prevention. There also are likely to be economies that can be achieved through a division of labor between physicians and other health disciplines, including physician assistants, nurse practitioners, and others. Furthermore, office systems and centralized reminder systems offer the potential to stay in contact with patients without face-to-face encounters.

A common set of expectations for preventive health care between payers, providers, and the public, and the means to achieve those goals, likely will accelerate progress toward preventive health objectives. Some may be skeptical that the public will readily accept an evidence-based approach to periodic preventive health examinations. However, it is important to appreciate that the public's attitudes are shaped by history, practice patterns that frequently are consumer-driven (thus reinforced) rather than evidence-based, and by competing, entrepreneurial interests. It is our belief that the public's positive orientation toward prevention and early detection, combined with dependable health education from their physician and other trusted sources, can lead to an informed consumer who will readily accept an effective, evidence-based model for periodic preventive health encounters. The results of the KCIS study challenge us to move beyond opportunistic preventive care.