9
Screening for Cancer

Screening for cancer ranges from visual inspection to sophisticated radiologic or biochemical techniques.2,10,26,34,70,81 It is generally associated with the subsequent verification of in situ or invasive cancer and the surgical removal of all identifiable cancerous tissue. A common example is mammography in an asymptomatic woman over the age of 50. If the mammogram yields a suspicious finding, such a screening may be followed by surgical removal of an invasive but localized carcinoma of the breast. Cancer screening is a form of secondary prevention and should be distinguished from primary prevention (e.g., reductions in cigarette smoking to prevent the occurrence of lung cancer) and treatment (e.g., lumpectomy and lymphadenectomy or chemotherapy for breast cancer that is not detected by screening).

In this chapter, the incidence rate refers to the proportion of a population with cancer within a given limited time interval (e.g., 50 cases per 100,000 persons at risk per year), and prevalence rate refers to the number of existing cases of cancer at a particular point in time (e.g., 180 per 100,000 persons on January 1, 1982). Unless otherwise designated, prevalence rates include those with previously diagnosed cancer (whether or not they are free of cancer at the time of the survey).

The prevention of death from cancer is often considered to be the essential outcome for measuring the efficacy of cancer screening;56,68 unless otherwise noted, it is the definition of the efficacy of screening



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability 9 Screening for Cancer Screening for cancer ranges from visual inspection to sophisticated radiologic or biochemical techniques.2,10,26,34,70,81 It is generally associated with the subsequent verification of in situ or invasive cancer and the surgical removal of all identifiable cancerous tissue. A common example is mammography in an asymptomatic woman over the age of 50. If the mammogram yields a suspicious finding, such a screening may be followed by surgical removal of an invasive but localized carcinoma of the breast. Cancer screening is a form of secondary prevention and should be distinguished from primary prevention (e.g., reductions in cigarette smoking to prevent the occurrence of lung cancer) and treatment (e.g., lumpectomy and lymphadenectomy or chemotherapy for breast cancer that is not detected by screening). In this chapter, the incidence rate refers to the proportion of a population with cancer within a given limited time interval (e.g., 50 cases per 100,000 persons at risk per year), and prevalence rate refers to the number of existing cases of cancer at a particular point in time (e.g., 180 per 100,000 persons on January 1, 1982). Unless otherwise designated, prevalence rates include those with previously diagnosed cancer (whether or not they are free of cancer at the time of the survey). The prevention of death from cancer is often considered to be the essential outcome for measuring the efficacy of cancer screening;56,68 unless otherwise noted, it is the definition of the efficacy of screening

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability for cancer that is used here. One reason for this definition is the lack of epidemiologically based research using morbidity indicators. The advantages and disadvantages of this measure are discussed further in the section entitled ''Preventability of Burden." In this chapter, an ecologic study indicates the examination of population-based rates. Case-control studies involve rates of prior screening in cases (e.g., persons who have died from the cancer) as compared with such rates in controls (e.g., persons who have not died from the cancer and usually have not contracted it). Nested case-control studies refer to case-control studies conducted within a longitudinally defined cohort in which careful attempts are made to classify all members as to level of screening. This methodology limits the case-control comparisons to a fraction of the total cohort—for example, all cases who died from the cancer and five times as many age-stratified controls (the remaining controls are excluded). Levels of assurance are based on whether evidence for efficacy involves the following: (1) different research designs, which may include ecologic and case-control studies, observational cohort studies, controlled nonrandomized studies, and randomized controlled trials; (2) the magnitude of the observed protective effect (e.g., an observed reduction in risk of 50 percent versus 10 percent); (3) the extent of care given to collecting data on the screening modality and on the outcome (e.g., death from the cancer), and the extent to which covariables such as socioeconomic status and race were considered; (4) the consistency in the results across different studies; (5) supporting data on sensitivity and specificity; and (6) the extent to which older age groups were studied. BURDEN Incidence, Mortality, Survival, and Prevalence Age is the most consistent and strongest predictor of risk for cancer and for death from cancer. (The effect of age is so dramatic that incidence and mortality rates increase exponentially with age.)21,71 Table 9-1 provides population-based incidence rates and incidence/mortality ratios from the Surveillance, Epidemiology, and End Results (SEER) program in the United States. The adult epithelial cancers are more common in persons over the age of 50 than in those under that age, and the rates of incidence and mortality increase with each decade of life after age 50. The incidence/mortality ratios imply that death rates for cancer generally increase with age more rapidly than incidence rates.

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability TABLE 9-1 Incidence (I) per 100,000 Persons and Incidence/Mortality (I/M) Ratios for Selected Cancer Sites, U.S. Surveillance, Epidemiology, and End Results Program, 1981–1985   Age Groups Site 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74 75–79 80–84 All Sites I 29 47 75 121 199 323 499 780 1,125 1,506 1,898 2,187 2,433 I/M 4.7 4.9 4.5 3.8 3.1 2.6 2.2 2.2 2.1 2.0 2.0 1.9 1.7 Rectum I 0.1 0.4 0.8 1.8 4.4 9.3 19.1 33.4 49.9 68.4 89.7 102.0 120.3 I/M — 4.0 8.0 6.0 6.3 6.2 6.2 6.1 5.8 5.5 5.0 4.4 3.7 Lung I 0.2 0.5 1.6 5.8 19.1 44.2 88.5 156.0 221.8 287.0 329.0 313.2 274.0 I/M 2.0 1.7 1.6 1.4 1.4 1.3 1.3 1.3 1.3 1.2 1.2 1.1 1.1 Breast I 1.1 8.4 27.5 66.0 114.6 169.8 194.6 244.3 292.2 31.5 362.3 380.2 397.3 I/M 5.5 7.0 5.1 5.2 4.9 4.6 3.5 3.3 3.3 3.3 3.2 3.0 2.7 Cervix I 2.0 7.6 12.1 15.1 16.6 16.8 17.1 19.0 19.1 19.9 20.0 19.7 18.9 I/M — 8.4 6.4 4.7 3.5 2.8 2.4 2.3 2.1 2.0 1.9 1.6 1.3 Melanoma I 2.7 5.6 8.7 12.1 13.8 15.6 17.5 20.2 21.5 22.0 23.3 24.6 26.8 I/M 9.0 8.0 6.7 7.1 5.5 5.0 4.5 4.2 4.0 3.5 3.0 2.9 2.6   SOURCE: Data abstracted from material provided by the Division of Cancer Prevention and Control, National Cancer Institute, 1988.

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability TABLE 9-2 Age-specific Prevalence Rates per 100,000 Persons for Selected Cancer Sites in Connecticut on January 1, 1982   Age Group Site 0–29 30–49 50–59 60–69 70+ Totala Men All sites 134.4 597.6 2,296.3 5,380.3 11,809.7 1,789.0 Pharynx 2.6 32.6 196.5 390.2 608.3 107.7 Colon 0.7 26.2 246.4 723.8 2,053.2 249.3 Rectum 0.3 13.4 161.1 477.3 1,123.2 144.7 Lung 0.8 27.7 238.6 564.5 757.6 134.8 Melanoma 5.6 74.5 200.7 238.3 261.3 80.6 Women All sites 142.7 1,169.8 4,538.0 7,530.7 10,635.0 2,221.6 Pharynx 2.8 24.1 96.3 165.1 198.8 45.5 Colon 1.7 31.8 251.2 662.8 1,887.7 224.1 Rectum 0.1 14.1 124.0 350.3 722.9 97.6 Lung 0.3 24.5 143.3 272.2 233.6 61.2 Melanoma 9.5 102.5 184.3 188.4 183.6 77.0 Breast 4.3 413.3 2,067.3 2,983.4 3,888.7 847.6 Cervix 8.1 120.8 286.6 447.4 528.0 138.4 a Age-adjusted to the 1980 U.S. population. SOURCE: A. R. Feldman, L. Kessler, M. H. Myers, et al., "The Prevalence of Cancer: Estimates Based on the Connecticut Tumor Registry," New England Journal of Medicine, Vol. 315, pp. 1394-1397, 1986. Table 9-2 presents prevalence data for selected cancers.31 (Because the data come from a limited geographic region, the state of Connecticut, any given subgroup of the U.S. population may have a prevalence rate for a given cancer that is quite different from that shown in Table 9-2.) Incidence rates for colorectal cancer are about 5 to 10 percent higher in Connecticut than in remaining portions of the SEER areas; for breast cancer, they are somewhat higher in Connecticut as compared with other SEER areas. In general, however, these prevalence rates are probably not dramatically different from prevalence rates for the United States as a whole.71 As noted earlier, prevalence rates include those with previously diagnosed cancer who are free of the disease and also those who are not free of disease. For cancers with high incidence/mortality ratios, persons free of disease but with a prior diagnosis of cancer would generally represent the majority of prevalent cases. For example, persons aged 70 to 74 with rectal cancer have an incidence/mortality

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability ratio of 5.0 compared with 1.2 in the same age group for lung cancer. For rectal cancer, the percentage of prevalent cases free of disease is much greater than that for lung cancer. Impairment, Disability, Handicap, and Costs A description of the aspects of cancer that pertain to impairment, disability, and handicap is difficult to derive from available published U.S. data, and in the absence of known prevalence rates for these cancer-related problems, a comprehensive description of the cost of cancer is not possible. Several morbidity and quality of life indices associated with specific cancers are listed below.23,31,51,57,58,68,79 Impairment Psychological stress and fear associated with diagnosis of cancer at any site. Bone pain (particularly with lung, breast, and prostate cancers). Impairment of joint motion or joint discomfort (particularly at the shoulder in association with breast cancer). Disfigurement, particularly with respect to surgery for head and neck cancer and breast cancer. Generalized weakness associated with cachexia of cancer and with chemotherapy. Disability Behavioral difficulties, particularly those associated with embarrassment stemming from mastectomy or head and neck surgery. Communication involving difficulties in phonation stemming from pharyngeal and laryngeal cancer and necessary surgery. Handicap Loss of employment (in association with loss of health insurance) or fear of loss of employment as a result of knowledge of an employer or insurer regarding the diagnosis of cancer in an employee. All categories of handicap associated with progression of end-stage cancer.

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability PREVENTABILITY OF BURDEN This section summarizes published data on the efficacy of screening for cancers in persons over the age of 50. The review focuses on those cancers that are generally accepted by most authoritative review groups as screenable without special reference to age: cancers of the female breast, uterine cervix, and colon and rectum; skin cancer (including melanoma); and oral cavity cancer.2,10,26,44,81 Efficacy is defined as postponement or avoidance of death by the screening intervention. Ecologic Considerations The comparison of trends in population-based incidence and mortality rates for aggregate populations provides a useful background for considering the efficacy of screening. Long-term trends in incidence rates since around 1950 have a more positive slope than do corresponding long-term trends in mortality rates for invasive breast cancer, colorectal cancer, cancer of the uterine cervix, and melanoma skin cancer.26,71 These patterns apply to the oldest age groups as well as to younger populations.53,71 The differential patterns in incidence and mortality trends coincide with the introduction and increasing availability of breast physical exam and mammography, pap testing, physical exam of the skin, and increasing scrutiny of the large bowel using rigid sigmoidoscopy, fecal occult blood testing, and, since the 1970s, flexible sigmoidoscopy. In these studies, however, the so-called ecologic fallacy may apply because the unit of analysis is an aggregate population. That is, undefined factors in these aggregate populations could also account for the observed patterns in time trends for incidence and mortality. The remaining portion of this chapter focuses on data in which the unit of analysis is the individual. Female Breast Cancer Table 9-3 summarizes four studies that examined screening efficacy in two or more age groups over the age of 50. Two other publications4,64 provided information on two age groups only, but these studies were nevertheless consistent with the pattern in Table 9-3. (A seventh study in the United Kingdom has published a report indicating a modest protective effect.81) All six of the studies are consistent with a reduction in breast cancer mortality in age groups

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability TABLE 9-3 Results of Studies of Screening and Mortality from Breast Cancer in Persons Over the Age of 50a Study Results Health Insurance Plan of Greater           New Yorkb           Age groups 40–44 45–49 50–54 55–59 60–64 Relative riskc of death from breast cancer 0.73 0.82 0.78 0.86 0.69 Number of deaths from breast cancer (66) (80) (96) (78) (57) Breast Cancer Detection and Demonstration Projectd Age groups 35–49 50–59 60–74     Observed/expected deaths from breast cancer 0.89 0.76 0.74     Number of deaths from breast cancer (49) (51) (25)     Nijmegen, Netherlandse Age groups 35–49 50–64 65+     Relative risk of death from breast cancer 1.23 0.26 0.81     Number of deaths from breast cancer (19) (27) (16)     Utrecht, Netherlandsf Approximate midpoint age group 54 59 64     Relative risk of death from breast cancer 0.82 0.31 0.05     Number of deaths from breast cancer (12) (14) (20 over age 50)     a Age groups for each study represent ages at entry into screening. b This study recruited 60,000 women who were randomized to two groups: usual care or mammography plus a physical exam. The study was begun in 1963 and included a 16-year follow-up period. See the analyses by Habbema and colleagues (reference no. 37) and Schapiro and coworkers (reference no. 76). c Defined as the probability of death from breast cancer in the screened divided by the probability of breast cancer in the unscreened. d Fifty-five thousand self-referred women participated in this study, which involved two-view mammography as the screening modality. Begun in 1973, the study included a 9-year follow-up period and comparison with rates from the Surveillance, Epidemiology, and End Result (SEER) program areas. (See Morrison and colleagues, reference no. 56, for further analysis.) e Female inhabitants of Nijmegen, Netherlands, were invited to participate in a nested study using single-view mammography as the screening modality. The study was begun in 1975. For further details, see reference nos. 82 and 83. f This study was begun in 1974 and used a nested case-control design to study the effectiveness of xeromammography plus a clinical exam (see reference no. 20).

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability from 50 to 70 years by a factor of 1. 2- to more than 3-fold as a result of the screening modalities. Furthermore, there is no overall indication of a decline in protective effect with increasing age; if anything, there is a suggestion that screening is more effective in postmenopausal women. Table 9-3 is consistent with a recent independent review.28 In fact, there may be some improvement in sensitivity or specificity of breast cancer screening with age, perhaps in part related to menopause, declining estrogen stimulation, and declining benign breast disease. Independent of this feature of the screening test, its predictive value would increase with age because positive predictive value is affected by the underlying prevalence of the cancer in the population subjected to the test, and Tables 9-1 and 9-2 demonstrate the increase in cancer incidence, mortality, and prevalence that occurs with age. Thus, for any given level of sensitivity and specificity, the positive predictive value of breast cancer screening will be much higher in older women. In other words, Table 9-1 suggests that, other factors being equal, the probability that a positive mammogram will, indeed, lead to a diagnosis of breast cancer is more likely in a 75-year-old woman than in a 45-year-old woman. With respect to female breast cancer, the rate of advance of preclinical breast cancer may be somewhat longer in older persons than in younger persons.22 However, this effect is probably not a large one ("… for women over 50 the sojourn time increases slowly with age"22). In particular, the committee did not find evidence in published data that invasive breast cancers in the elderly were, as a group, less aggressive or less likely to kill the woman at any given stage at time of diagnosis.1,6,14,35,39,55,76 (In particular, see the discussion with R. Peto noted in reference 70.) For example, in women who were 60 to 64 years old on entry into the study conducted by the Health Insurance Plan (HIP) of Greater New York (see Table 9-3), the number of incident cases of breast cancer after 16 years of follow-up were 68 and 69 in the screened versus unscreened groups, respectively, as compared with 425 and 443 over all women aged 40 to 64.39 If breast cancer were relatively less lethal in this age group, more incident cases would have been identified in this group than in younger women. In summary, with advancing age, there is a marked decline in the ratio of benign-to-malignant outcomes of biopsies in screened women and an improvement in the positive predictive value of mammography that can be in the range of 45 to 85 percent in women over the age of 65.6,65

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability Cancer of the Uterine Cervix The efficacy of screening for cervical cancer is supported by time trends in incidence and mortality as noted above. In addition, a study by Hakama40 provides differences in cervical cancer mortality trends by geographic region that correspond well with differing geographic trends in screening prevalence. Further supporting evidence comes from three case-control studies that have shown protective effects from pap testing by a factor of 2- to 3-fold or more.8,16 Clarke and Anderson demonstrated that the relative risk of death from cervical cancer was reduced in those who had undergone prior pap testing; the overall reduction in risk was similar to or greater in magnitude than that for invasive cancer.16 The ability to use invasive cervical cancer as the case definition presumably stems from the sensitivity of pap testing to carcinoma in situ. Once the carcinoma had been removed, both invasive cervical cancer and death from cervical cancer would be prevented. The original study by Clarke and Anderson16 provided some detail on the relative risk for invasive cervical cancer by age at diagnosis. Women in the sixth and seventh decades of life benefited from screening as much or more than younger women (see Table 9-4). These results are consistent with other observations. In situ carcinoma of the cervix persists in its occurrence from ages 55 to 75 TABLE 9-4 Estimates of Relative Risk (Derived from Pap Testing) for Invasive Cervical Cancer by Age at Diagnosis from a Case-Control Study   Age Group Risk 20–34 35–44 45–59 60+ Relative riska for invasive cervical cancer 0.48 0.50 0.32 0.29 Number of cases (16) (36) (102) (58)b a Defined as the probability of death from cervical cancer in the screened divided by the probability of cervical cancer in the unscreened. b Overall prevalence of prior cervical cancer screening was 32 percent in cases and 56 percent in controls. SOURCE: E. A. Clarke and T. W. Anderson, "Does Screening by 'Pap' Smears Help Prevent Cervical Cancer? A Case-Control Study," Lancet , Vol. 2, pp. 1-4, 1979.

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability and older.4 Moreover, in a collaborative study of screening programs in eight countries involving 1.9 million women, there was no detectable decline in the sensitivity of cervical cancer screening up to the age of 65, the last age evaluated.44 Celentano and colleagues13 have published data on the duration since the last screen of the protective effect derived from pap testing and have summarized data from several studies. Protection against invasive cervical cancer probably persists for four to six years after the last screen, although the magnitude of protection may decline by a factor of at least 2 after three years. Such information is relevant to determining whether arbitrary termination of cervical cancer screening in elderly persons is wise; this problem is discussed later in the chapter. Cancer of the Colon and Rectum Data on sigmoidoscopy35,36,43,61,75 and fecal occult blood testing5 demonstrate the potential of these procedures as effective screening tests for colorectal cancer. The predictive value of hemoccult testing improves as the age of the patient undergoing the screening increases.60,84 Flexible sigmoidoscopy yields more data on the presence of cancer62 and is better tolerated than rigid sigmoidoscopy. However, flexible sigmoidoscopy is limited to distal segments of the large bowel and may miss more than two-fifths of the cancers suspected by hemoccult.60,63 Limitations in the overall sensitivity of hemoccult screening are being altered by newer, immunologically based tests;41 these improvements could be particularly relevant to screening in the elderly, who may tolerate fecal occult blood testing much better than flexible sigmoidoscopy. In studies of screening for colorectal cancer, there are few published data that adequately address problems of lead-time and length bias sampling (see the later "Discussion" section). A recently reported case-control study62 (presented at meetings of the American Society for Oncology in 1989) found a 2- to 3-fold protective effect from screening sigmoidoscopy. Analyses by age of the results of this case-control study and other studies in progress42,45,48 may clarify the protective potential of colorectal cancer screening in the elderly. Skin Cancer (Including Malignant Melanoma) The sensitivity and specificity of skin examinations for cutaneous malignant melanoma and for basal cell carcinoma appear to exceed 85 to 90 percent.39,46,47,66,71 The efficacy of a thorough clinical

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability examination of the skin is based primarily on the concept of a radial or horizontal growth phase of sufficient duration and extent so that gross anatomic recognition is feasible. If it is, the sensitivity and specificity noted above should lead to some reduction in mortality from skin cancer. To the committee's knowledge, there are only two sets of data that provide any evidence on this question. Greene and coworkers37 found that, after implementation of surveillance for melanoma and dysplastic nevi, there was total elimination of mortality from melanomas in 14 families with the apparent autosomal dominant familial dysplastic nevus syndrome, a syndrome said to carry a virtually 100 percent risk for melanoma to the eighth decade of life in those with multiple large nevi. In a small set of pilot data that were gathered to prepare for a large field study of melanoma, an inverse association between prior examination of the skin and either late-stage or lethal melanoma was observed.72 To the committee's knowledge, there are no available data on the efficacy of screening for skin cancer in the elderly. It should be noted, however, that, with age, inspection of the skin becomes increasingly complex and may lead to greater difficulty in distinguishing neoplastic lesions.72 Screening for Other Cancers Some attention has been given to screening for cancers of the mouth,5,30,50 lungs,27,73 and prostate.15,30 For lung cancer, case-control data fail to show a protective effect from screening, and prospective data do not support a benefit from cytology as compared with chest x-rays among smokers.73 Regarding prostate cancer, ultrasound screening of the prostate is probably a more sensitive measure than the digital rectal exam.50 Whether use of the ultrasound procedure will lead to actual reductions in morbidity and mortality has not been fully evaluated.50,70 Prostatic antigen assays do not appear to have sufficient specificity to allow their use in screening.9 Controlled studies to account for lead-time and length bias sampling (see the next section) are particularly important in view of the high proportion of men over age 60 with changes of the prostate that are histologically malignant but biologically relatively benign. In summary, there are insufficient data at present to determine whether screening for this cancer produces benefits. DISCUSSION Table 9-5 summarizes evidence on the efficacy of screening in persons over the age of 50. The determination of efficacy in these

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability TABLE 9-5 Summary of Evidence for Efficacy of Screening for Cancer in Persons over the Age of 50 Cancer Site Screening Intervention Extent of Evidence and Limitations Female breast Mammography, with or without physical exam Ecologic, case-control, and observational cohort studies; randomized controlled trial. None are contradictory; ages over 70 not extensively studied. Level of assurance: aA. Uterine cervix Pap smear At least one case-control study showed more than a 3-fold reduction; multicenter ecologic study showed no evidence for decline in sensitivity to age 65. Level of assurance: A to B. Large bowel Hemoccult and flexible sigmoidoscopy Efficacy in elderly not well studied (one case-control study and observational cohort studies tended to favor overall efficacy of screening). Increase in the use of medications in the elderly may affect sensitivity and specificity. Level of assurance: C. Skin cancer, including melanoma Skin exam Efficacy in elderly not well studied (case-control study is in progress). Possible decline in accuracy of exam owing to changes in skin with age. Level of assurance: C. a Levels of assurance (''A" being most confident) are based on whether (1) evidence comes from studies with different designs, which may include ecologic or case-control studies, observational cohort studies, controlled nonrandomized studies, and randomized controlled trials; (2) the extent of care given to collecting data on the screening modality and the occurrence of outcome and whether covariables were considered; (3) the consistency across different studies; and (4) the extent to which older age groups were studied. studies is often based on prevention of death from the cancer. A disadvantage of the mortality measure is that it fails to address directly the question of quality of life and the impairments, disabilities, and handicaps that may be associated with the disease, prevention, or treatment. These deficiencies in the information these studies provide make it particularly difficult to develop cost-benefit analyses that have comprehensive implications. On the other hand, the use of mortality from cancer as an outcome measure has certain advantages in the context of a report on screening in the elderly. For example, one can conceive of a screening modality whose only effect is to advance the time of diagnosis prior to the eventual occurrence

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability of death rather than to prevent death itself. In such an instance of lead-time bias, the elderly person is likely to be subjected to the negative effects that arise from being labeled as diagnosed with cancer, as well as the likelihood of additional uncomfortable or painful instrumentation or surgery; these effects occur in the absence of any real postponement of death from the cancer. Also, one could conceive of a screening modality whose effect was to identify tumors that are declared to be cancer but that in fact are without lethal potential or that, as a group, take many years to kill the individual and are much less aggressive in comparison to cancers that are identified on the basis of symptoms or routine clinical exams. In such an instance of length bias sampling, the social and medical consequences are the same as in the prior example: adverse interventions without material prolongation of life. There are few available data on the efficacy of screening for breast cancer in those over the age of 75, but screening efficacy is well supported in the years from ages 50 through 65. Although data on screening for cervical cancer in women over age 50 are sparse, the Clarke and Anderson case-control study16 and the ecologic studies noted earlier are consistent with a strong protective effect from cervical cancer screening through age 65. Taken together, the data on these two very different epithelial cancers support the conclusion that the predictive value of screening tests in the elderly is at least as good if not better than the positive predictive values characteristic of younger age groups. From the standpoint of evidence of efficacy, it is difficult to support the notion of establishing an arbitrary age at which to terminate screening for cancer such as, for example, terminating screening for cancer of the cervix at age 65. Yet the committee notes that recent recommendations regarding cervical cancer are consistent with this practice.32 This may well be counterproductive: if either sensitivity or specificity are, indeed, less in a given screening modality (which seems not to be the case in screening for cervical and breast cancer), this deficit may be partly or entirely offset by much greater predictive value at any given level of sensitivity and specificity owing to the more common occurrence of cancer (see Tables 9-1 and 9-2). The issue of competing causes of death becomes increasingly important in the context of cancer screening and aging in the very elderly; for example, it is recognized18 that average life expectancy is 6.9 years for an 85-year-old woman and that the average, untreated natural history between onset of detectable preclinical cancer and death from it may be nearly as long or longer. The importance of

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability such an effect could be estimated with appropriately controlled studies. Data from a 1987 survey and from other studies indicate that elderly populations have a lower prevalence of prior screening as compared with younger populations.12,65 Table 9-6 summarizes the results for three screening modalities: pap smear, mammography, and fecal occult blood testing. In most instances, the prevalence of the never-screened was higher in those 70 years of age and older, with blacks over the age of 70 having the highest rates of never-screened persons. It is possible, however, that some of these effects might reflect the open-endedness of the oldest age group or failures to report prior screening in the elderly in this survey. The above comments do not consider either factors reflecting quality of life, the discomfort of the screening procedures, or economic cost-benefit analyses. Eddy has developed estimates on the benefits and costs related to screening for breast cancer.28 Increases in life expectancy in those aged 65 to 75 were approximately 60 to 90 percent of the increases in life expectancy that were achieved by screening women in age groups between the ages of 40 and 65; the cost of adding a year of life by screening for breast cancer was about 10 percent greater for women aged 65 to 74 as compared with women aged 55 to 65, but it was actually 31 percent lower than the cost for women in their forties. Systemic biochemical screening tests33 (for example, nuclear resonance spectroscopy of serum) are an interesting possibility, but they TABLE 9-6 Percentage of Persons Reporting No Prior Screening Procedures       Fecal Occult Blood Testing Age Group Pap Smear Mammography Men Women White 40–49 4.3 57.7 67.6 72.7 50–59 4.2 53.7 63.1 58.1 60–69 7.7 61.4 55.1 54.1 70+ 22.6 71.8 61.4 60.4 Black 40–49 2.7 64.0 78.6 77.3 50–59 10.7 69.9 72.7 69.9 60–69 21.9 71.7 62.2 69.2 70+ 43.4 82.4 73.6 78.0   SOURCE: National Health Interview Survey, 1987.

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability are particularly problematic in the elderly because they might lead to extensive diagnostic evaluations without being able to focus on a particular organ site. However, there have been efforts to develop biochemical tests that would localize suspicion to a particular organ site such as the lung, breast, or prostate.9,54,59 RECOMMENDATIONS Implementation of Screening For persons over the age of 50, it is important to consider screening for cancer with the same inherent interest as for younger persons. Initially, at least, all elderly persons should be considered candidates for physical exam of the breast and mammography, pap testing, physical exam of the skin, fecal occult blood testing and flexible sigmoidoscopy to 35 centimeters, and oral exam. The frequency of screening and the balancing of the limited screening options for this group (e.g., hemoccult versus flexible sigmoidoscopy) may be dramatically affected by such factors as the ability of the individual to undergo screening itself, the ability to tolerate surgery if a suspicious lesion is identified, the risk factors other than age that predispose the person to cancer, the incidence of the cancer, prior negative screening tests within the past 2 to 15 years, and the individual's life expectancy, which may change dramatically over a 5- to 10-year period among the very elderly. Research The committee recommends a systematic reanalysis of existing data sets, including those involving population-based rates, case-control data, and prospectively collected data, to evaluate the efficacy of screening in individual age groups over the age of 50. Furthermore, new data on screening should be collected, data that will address the relative importance of the several factors (e.g., tolerance to screening, life expectancy) noted in the previous paragraph. Regarding colorectal cancer, the use of fecal occult blood testing, in comparison with flexible sigmoidoscopy, is particularly appealing for elderly individuals because it is noninvasive, and some of the recent modifications of this technique suggest improvements in specificity in comparison with prior occult blood testing. Studies should be conducted to evaluate these recent tests; their results could be particularly relevant to screening for colorectal cancer in the elderly. On the other hand, it would also be helpful to assess the degree to which elderly persons are capable of conducting fecal

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability occult blood testing and to examine the tolerance of flexible sigmoidoscopy (to 35 and 60 centimeters) in subgroups of the elderly population. Also of interest and concern are the rates of complication of endoscopy, such as bowel perforation, in the age group over 50. Prostate cancer is common in the elderly and is a potentially "screenable" cancer. Screening using ultrasound appears to have the potential to enhance screening sensitivity markedly; however, it will be impossible to evaluate the true efficacy of such screening without collecting data that take into account the problems of lead-time and length bias sampling. Epidemiologically grounded methodologic research would be valuable in examining how to incorporate nonmortality measures of screening efficacy, including measures that would reflect impairment, disability, and handicap. This methodologic research should be statistical (e.g., the development of measures to estimate the prevalence of morbidity from existing incidence, mortality, and survival data); it should also consider ways to incorporate measures of cancer prevalence in routinely collected survey data (e.g., minimally, a self-report as to prior diagnosis and a self-report as to impairment associated with surgery on the skin, lymph nodes, or large bowel). Such research would then form the basis for studies of screening efficacy that would not have to be limited solely to measures of the avoidance of mortality from cancer but could also include quality of life. In addition, this research could address the costs of screening in the elderly, an area in which the reassuring findings of cost-benefit studies of cervical and breast cancer offer hopes of further favorable results.11,44,52,77,85 Education The committee recommends educational programs primarily for the relatively noninvasive screening modalities: skin exams, breast physical exams, and mammography. Because of uncertainties regarding colorectal screening and the need to judge individually whether elderly persons should undergo fecal occult blood testing and endoscopy, education regarding screening for large bowel cancer is not recommended. It is difficult to recommend a vigorous educational campaign to convey specific information on two common cancers, colorectal cancer and prostate cancer, because of imprecision regarding estimates of efficacy and individual variations in tolerance to screening protocols. No educational effort is recommended for screening programs for lung cancer until there is better evidence regarding their efficacy.

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability REFERENCES 1. Adami, H. O., Malker, B., Holmberg, L., et al. The relation between survival and age at diagnosis in breast cancer. New England Journal of Medicine 1986; 315:559-563. 2. American Cancer Society. Cancer Facts and Figures. Atlanta, Ga.: American Cancer Society, 1988. 3. American Cancer Society. Summary of Current Guidelines of the Cancer-related Checkup: Recommendations. New York: American Cancer Society, 1988. 4. Anderson, I., Aspegren, K., Janzon, L., et al. Mammographic screening and mortality from breast cancer: The Malmo mammographic screening trial. British Medical Journal 1988; 297:943-948. 5. Baden, E. Prevention of cancer of the oral cavity and pharynx. CA 1987; 37:49-62. 6. Baden, J. P. Screening of colorectal cancer. Digestive Diseases and Sciences 1986; 31(Suppl. 9):43S-56S. 7. Baker, L. H. Breast cancer detection demonstration project: Five-year summary report. CA 1982; 32:4-35. 8. Berrino, F., Gatta, G., D'Alto, M., et al. Use of case-control studies in evaluation of screening programmes. In: P. C. Prorok and A. B. Miller (eds.), Screening for Cancer. 1. General Principles on Evaluation of Screening for Cancer and Screening for Lung, Bladder and Oral Cancer. Geneva: International Agency for Research on Cancer (IARC), 1984, pp. 29-43. 9. Brawer, M. K., and Lange, P. H. Prostate-specific antigen and premalignant change: Implications for early detection. CA 1989; 39:361-375. 10. Canadian Task Force on the Periodic Health Examination. The Periodic Health Examination. 2. 1985 Update. Canadian Medical Association Journal 1986; 134:724-729. 11. Carter, A. P., Thompson, R. S., Bourdeau, R. V., et al. A clinically effective breast cancer screening program can be cost-effective, too. Preventive Medicine 1987; 16(1):19-34. 12. Celentano, D. D., Shapiro, S., and Weisman, C. S. Cancer preventive screening behavior among elderly women. Preventive Medicine 1982; 11:454-463. 13. Celentano, D. D., Klassen, A. C., Weisman, C. S., et al. Duration of relative protective screening for cervical cancer. Preventive Medicine 1989; 18:411-422. 14. Charlson, M. E., and Feinstein, A. R. Rapid growth rate: A method of identifying node-negative breast cancer patients with a high risk of recurrence. Journal of Chronic Disease 1983; 36:847-853. 15. Chodak, G. W., Keller, P., and Schoenberg, H. Routine screening for prostate cancer using the digital rectal examination. Program Clinical Biology Research 1988; 269:87-98. 16. Clarke, E. A., and Anderson, T. W. Does screening by "pap" smears help prevent cervical cancer? A case-control study. Lancet 1979; 2:1-4. 17. Clarke, E. A. Re: "Case definition in case-control studies of the efficacy of screening" (letter). American Journal of Epidemiology 1982; 115:518-519. 18. Clinical Practice Committee, American Geriatrics Society Board of Directors. Screening for cervical carcinoma in elderly women. Journal of the American Geriatrics Society 1989; 37:885-887. 19. Clinical Practice Committee, American Geriatrics Society Board of Directors . Screening for breast cancer in elderly women. Journal of the American Geriatrics Society 1989; 37:883-884.

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability 20. Collette, H. J. A., Day, N. E., Rombach, J. J., and deWaard, F. Evaluation of screening for breast cancer in a non-randomized study (the DOM project) by means of a case-control study. Lancet 1984; 1:1224-1226. 21. Cook, P. J., Doll, R., and Fellingham, S. A. A mathematical model for the age distribution of cancer in man. International Journal on Cancer 1969; 4:93-112. 22. Day, N. E., Bains, C. J., Chamberlain, J., et al. UICC project on screening for cancer: Report of the Workshop on Screening for Breast Cancer. International Journal on Cancer 1986; 38:303-308. 23. Dietz, J. H., Jr. Rehabilitation of the patient with cancer. In: P. Clabresi, P. S. Schein, and S. A. Rosenberg (eds.), Medical Oncology. New York: MacMillan, 1985, pp. 1501-1522. 24. Division of Cancer Prevention and Control, National Cancer Institute. 1987 Annual Cancers Statistics Review. NIH Publ. No. 88-2789. Bethesda, Md.: National Institutes of Health, 1988. 25. Drago, J. R. The role of new modalities in the early detection and diagnosis of prostate cancer. CA 1989; 39:326-336. 26. Early Detection Branch, Division of Cancer Prevention and Control, National Cancer Institute. Working Guidelines for Early Cancer Detection: Rationale and Supporting Evidence to Decrease Mortality. Bethesda, Md.: National Cancer Institute, 1987. 27. Ebeling, K., and Nischan, P. Screening for lung cancer—results from a case-control study. International Journal' of Cancer 1987; 40:141-144. 28. Eddy, D. M. Screening for breast cancer. Annals of Internal Medicine 1989; 111:389-399. 29. Editorial. Diagnostic and therapeutic technology assessment (DATTA): Transrectal ultrasonography in prostatic cancer. Journal of the American Medical Association 1988; 259:2757-2759. 30. Elwood, J. M., and Gallagher, R. P. Dental surveillance produces earlier diagnosis of oral cavity cancers. Canadian Dental Association Journal 1986; 10:845-847. 31. Feldman, A. R., Kessler, L., Myers, M. H., et al. The prevalence of cancer. Estimates based on the Connecticut Tumor Registry. New England Journal of Medicine 1986; 315:1394-1397. 32. Fink, D. J. Change in American Cancer Society checkup guidelines for detection of cervical cancer. CA 1988; 38:127-128. 33. Fossel, E. T., Carr, J. M., and McDonagh, J. Detection of malignant tumors: Water-suppressed proton nuclear magnetic resonance spectroscopy of plasma. New England Journal of Medicine 1986; 315:1369-1376. 34. Frame, P. S. A critical review of adult health maintenance. Part 3. Prevention of cancer. Journal of Family Practitioners 1986; 22:511-520. 35. Friedman, G. D., Collen, M. F., and Fireman, B. H. Multiphasic health checkup evaluation: A 16-year follow-up. Journal of Chronic Disease 1986; 39:453-463. 36. Gilbertson, V. A. Proctosigmoidoscopy and polypectomy in reducing the incidence of rectal cancer. Cancer 1974; 34:936-939. 37. Green, M. H., Clark, W. H., Jr., Tucker, M. A., et al. High risk of malignant melanoma in melanoma-prone families with dysplastic nevi. Annals of Internal Medicine 1985; 102:458-465. 38. Green, A., Leslie, D., and Weedon, D. Diagnosis of skin cancer in the general population: Clinical accuracy in the Nambour survey. Medical Journal of Australia 1988; 148:447-450. 39. Habbema, J. D. F., van Dort Marssen, G. J., Van Putten, D. J., et al. Age-specific

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability reduction in breast cancer mortality by screening: An analysis of the results of the Health Insurance Plan of Greater New York Study. Journal of the National Cancer Institute 1986; 77:317-320. 40. Hakama, M. Trends in the incidence of cervical cancer in nordic countries. In: K. Magnus (ed.), Trends in Cancer Incidence: Causes and Practical Implications. Washington, D.C.: Hemisphere, 1982, pp. 279-292. 41. Hakkinen, I., Paasivup, R., and Partanen, P. Screening of colorectal tumors using an improved fecal occult blood test. Quantitative aspects. Gut 1988; 29:1194-1197. 42. Hardcastle, J. D., Armitage, N. C., Chamberlain, J., et al. Fecal occult blood screening for colorectal cancer in the general population: Results of a controlled trial. Cancer 1986; 58:397-403. 43. Hertz, R. E., Dedish, M. R., and Day, E. Value of periodic examinations in detecting cancer of the rectum and colon. Postgraduate Medicine 1960; 27:290. 44. International Agency for Research on Cancer (IARC) Working Group. Screening for squamous cervical cancer: Duration of low risk after negative results of cervical cytology and its implication for screening policies. British Medical Journal 1986; 293:659-664. 45. Kewenter, J., Bjork, S., Haglind, E., et al. Screening and rescreening for colorectal cancer: A controlled trial of fecal occult blood testing in 27,700 subjects. Cancer 1988; 62:645-651. 46. Koh, H. K., Lew, R. A., and Prout, M. N. Screening for melanoma skin cancer. Journal of the American Academy of Dermatology 1989; 20:159-172. 47. Kopf, A. W. Prevention and detection of skin cancer-melanoma. Cancer 1988; 62:1791-1795. 48. Kronborg, C., Fenger, O., Sondergaard, O., et al. Initial mass screening for colorectal cancer with fecal occult blood test. A prospective randomized study at Funen in Denmark. Scandinavica Journal of Gastroenterology 1987; 22:677-686. 49. Lee, F., Littrup, P. G., Torp-Pedersen, S., et al. Prostate cancer: Comparison of transrectal ultrasound and digital rectal examination for screening. Radiology 1988; 168:389-394. 50. Lee, F., Torp-Pedersen, S. T., and Siders, D. B. The role of transrectal ultrasound in the early detection of prostate cancer. CA 1989; 39:337-360. 51. Loescher, L. J., Welch-McCaffrey, D. W., Leigh, S. A., et al. Surviving adult cancers. Part 1. Physiologic effects. Annals of Internal Medicine 1989; 111:411-432. 52. Mandelblatt, J. S., and Fahs, M. C. The cost-effectiveness of cervical cancer screening for low-income elderly women. Journal of the American Medical Association 1988; 259:2409-2413. 53. McKay, F. W., Hanson, M. R., and Miller, R. W. Cancer Mortality in the United States: 1950-1977. National Cancer Institute Monograph No. 59. Washington, D.C.: U.S. Government Printing Office, 1982. 54. Moroz, C., Kahn, M., Ron, E., et al. The use of oncofetal ferritin-bearing lymphocytes as a marker for the screening, diagnosis and follow-up of patients with early breast malignancy: Screening of 34,000 women. Cancer 1989; 64:691-697. 55. Morrison, A. S. Screening in Chronic Disease. New York: Oxford University Press, 1985. 56. Morrison, A. S., Brisson, J., and Khalid, N. Breast cancer incidence and mortality in the breast cancer detection and demonstration project. Journal of the National Cancer Institute 1988; 80:1540-1547.

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability 57. Mullan, F. Seasons of survival: Reflections of a physician with cancer. New England Journal of Medicine 1985; 313:270-273. 58. Mullan, F. Re-entry: The educational needs of the cancer survivor. Health Education Quarterly 1984; 10(Suppl.):88-94. 59. Mulshine, J. L., Tockman, M. S., and Smart, C. R. Considerations in the development of lung cancer screening tools. Journal of the National Cancer Institute 1989; 81:900-905. 60. Nelson, H., Pemberton J. H. Asymptomatic patients with positive stool guaiac (letter). Journal of the American Medical Association 1989; 261:1800. 61. Neugut, A. I., and Pita, S. Role of sigmoidoscopy in screening for colorectal cancer: A critical review. Gastroenterology 1988; 95:492-499. 62. Newcomb, P., Norfleet, R. G., Surewics, T. S., et al. Efficacy of fecal occult blood screening (abstract). Preventive Medicine, in press. 63. Nivatvongs, S., Gilbertsen, V. A., Goldberg, S. M., et al. Distribution of large bowel cancer detected by occult blood test in asymptomatic patients. Diseases of the Colon and Rectum 1982; 25:420-421. 64. Palli, D., Del Turco, M. R., Buiatti, E., et al. A case-control study of efficacy in a non-randomized breast cancer screening program in Florence, Italy. International Journal on Cancer 1986; 38:501-504. 65. Peters, R. K., Bear, M. B., and Thomas, D. Barriers to screening for cancer of the cervix. Preventive Medicine 1989; 18:133-146. 66. Presser, S. E., and Taylor, J. R. Clinical diagnostic accuracy of basal cell carcinoma. Journal of the American Academy of Dermatology 1987; 16:988-990. 67. Prorok, P. C., and Miller, A. B. (eds.) Screening for Cancer. 1. General Principles on Evaluation of Screening for Cancer and Screening for Lung, Bladder and Oral Cancer. Geneva: International Agency for Research on Cancer, 1984. 68. Rice, D. P., Hodgson, T. A., and Kopstein, A. N. The economic costs of illness: A replication and update. Health Care Finance Reviews 1985; 7:61-80. 69. Ries, L. G., Pollack, E. S., and Young, J. L., Jr. Cancer patient survival. Journal of the National Cancer Institute 1983; 70:693-707. 70. Robie, P. W. Cancer screening in the elderly. Journal of the American Geriatric Society 1989; 37:888-893. 71. Roush, G. C., Holford, T. R., Schymura, M. J., and White, C. Cancer Risk and Incidence Trends. Washington, D.C.: Hemisphere, 1987. 72. Roush, G. C., Nordlund, J. J., Forget, B., et al. Independence of dysplastic nevi from total nevi in determining risk for nonfamilial melanoma. Preventive Medicine 1988; 17:273-279. 73. Sanderson, D. R. Lung cancer screening: The Mayo study. Chest 1986; 89(Suppl.):324S-326S. 74. Seidman, H., Gelb, S. K., Silverberg, E., et al. Survival experience in the breast cancer detection demonstration project. CA 1987; 37:258-290. 75. Selby, J. V., Friedman, G. D., and Collen, M. F. Sigmoidoscopy and mortality from colorectal cancer: The Kaiser Permanente multiphasic evaluation study. Journal of Clinical Epidemiology 1988; 41:427-434. 76. Shapiro, S., Venet, W., Stract, P., et al. A randomized trial of breast cancer screening. In: Selection, Follow-up and Analysis in the Health Insurance Plan Study. National Cancer Institute Monograph No. 67. NIH Publ. No. 85-2713. Bethesda, Md.: National Cancer Institute, 1985. 77. Sickles, E. A., Weber, W. N., Glavin, H. B., et al. Mammographic screening: How to operate successfully at low cost. Radiology 1986; 160:95-97.

OCR for page 136
The Second Fifty Years: Promoting Health and Preventing Disability 78. Sondek, E. J. Cancer Statistics Review, 1973-1986. NIH Publ. No. 89-2789. Washington, D.C.: U.S. Department of Health and Human Services, 1989. 79. Strauss, R. P. Psychosocial responses to oral and maxillofacial surgery for head and neck cancer. Journal of Oral Maxillofacial Surgery 1989; 47(4):343-348. 80. U.K. Trial of Early Detection of Breast Cancer Group. First results on mortality reduction in the U.K. Trial of Early Detection of Breast Cancer Group. Lancet 1988; 2:411-416. 81. U.S. Preventive Services Task Force. Guide to Clinical Preventive Services. Washington, D.C.: U.S. Department of Health and Human Services, 1989. 82. Verbeek, A. L., Hendriks, J. H., Holland, R., et al. Reduction of breast cancer mortality through mass screening with modern mammography: First results of the Nijmegen project, 1975-1981. Lancet 1984; 1:1222-1224. 83. Verbeek, A. L. M., Hendriks, J. H., Holland, R., et al. Mammographic screening and breast cancer mortality: Age-specific effects in the Nijmegen Project, 1975-82 (letter). Lancet 1985; 1:865-866. 84. Winawer, S. J., Andrews, M., Flehinger, B., et al. Progress report on controlled trial of fecal occult blood testing for the detection of colorectal neoplasia. Cancer 1980; 45:2959-2964. 85. Yu, S., Miller, A. B., and Sherman, G. J. Optimising the age, number of tests and test interval for cervical screening in Canada. Journal of Epidemiology and Community Health 1982; 36:1-20.