Journal of Cancer Education
ISSN: 0885-8195 (Print) 1543-0154 (Online) Journal homepage: http://www.tandfonline.com/loi/hjce20
A clinical cancer prevention curriculum in a comprehensive cancer center Victor G. Vogel MD, MHS To cite this article: Victor G. Vogel MD, MHS (1991) A clinical cancer prevention curriculum in a comprehensive cancer center, Journal of Cancer Education, 6:3, 133-139 To link to this article: https://doi.org/10.1080/08858199109528109
Published online: 01 Oct 2009.
Submit your article to this journal
Article views: 2
View related articles
Citing articles: 1 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=hjce20
J. Cancer Education. Vol. 6, No. 3, pp. 133-139, 1991 Printed in the U.S.A. Pergamon Press pic
0885-8195/91 $3.00 + .00 © 1991 American Association for Cancer Education
MINISYMPOSIUM ON TEACHING CLINICAL DECISION MAKING
A CLINICAL CANCER PREVENTION CURRICULUM IN A COMPREHENSIVE CANCER CENTER VICTOR
G.
VOGEL,
MD, MHS*
Abstract — Interns, residents, and medical students who spend one month at a comprehensive cancer center are presented with a strategy to incorporate cancer preventive services into their outpatient practices. This is accomplished through didactic lectures in clinical epidemiology. Terms that are essential to an understanding of disease occurrence are defined including incidence, prevalence, and mortality rate. Odds and probability are explored with the introduction of the odds ratio and the identification of risk factors. Primary and secondary cancer prevention are defined, and parameters related to cancer screening are introduced (sensitivity, specificity, prevalence, and positive predictive value). Risk factors for malignancy are identified where they are known and can be modified. Primary prevention strategies are reviewed by site along with data that support or refute recommendations for site-specific screening. Medical conditions and risk factor profiles that define high-risk groups are reviewed. Trainees are given copies of the US Preventive Services Task Force screening recommendations and a clinical handbook. The impact of this information on clinical practice patterns remains to be evaluated.
INTRODUCTION
to carcinogens. Love and his colleagues have written that if there is to be a shift toward increased practice of clinical preventive medicine, there should be concomitant emphasis upon teaching in basic and clinical disciplines relevant to this field.1 Problems arise, however, in identifying available time during clinical house staff training to schedule the necessary didactic material and practical experience that provides the appropriate increased emphasis. Additionally, a paradigm shift away from the curative model with its biotechnical frame of reference and a reorientation to clinical prevention is required.2 One environment in which a change in orientation can be initiated is the comprehensive cancer center where many house staff receive a portion of their clinical training. Much of this training is concentrated in the care of chronically or acutely ill cancer patients with limited life expectancies for whom the prevention of malignancy is not a consideration. When viewed from another perspective, this setting, with its population of patients who have not responded to conventional therapies, is an ideal venue for the preventive oncologist to interact with house staff for the purpose of teaching clinical cancer prevention. It is here that they are made aware of the need for alternative strategies.
Knowledge held by internal medicine and family practice residents about cancer prevention has not been systematically evaluated. During clinical training there are few opportunities for residents to acquire clinical cancer prevention skills. Once these skills are acquired, there are few opportunities to incorporate them into clinical practice. An understanding of clinical epidemiology is the foundation on which clinical cancer prevention is built, yet few medical schools and even fewer clinical residency programs provide training in the basic concepts of cancer epidemiology. As a result of this lack of training, house staff may be inclined to perceive cancer prevention in a unidimensional orientation. That is, a house officer may believe that cancer prevention involves only activities such as those related to behavior modification, or only those concerned with limiting environmental exposures This paper was presented, in part, at the annual meeting of the American Association for Cancer Education, Cincinnati, Ohio, October 5, 1990. *Assistant Professor of Medicine, and Epidemiology, University of Texas, M. D. Anderson Cancer Center; Assistant Professor of Epidemiology, University of Texas School of Public Health, Houston/Texas. Reprint requests to: Victor G. Vogel, MD, MHS, 1515 Holcombe Boulevard, Box 501, Houston.TX 77030.
133
134
V.G. VOGEL
Once each month the course is presented in 90-minute didactic sessions during which lunch is provided for the house staff and medical stu1. it is required of all primary care house dents by drug company sponsors, who rotate this responsibility on a regular basis. Attendance staff; is required of internal medicine house staff and 2. it has a defined content contained in a medical students, as well as the family practice printed text; house staff. Handouts are provided that summa3. the content addresses deficiencies in knowlrize the presented information and cite appropriedge of epidemiology and biostatistics that ate references. Participants receive a handbook are common among house staff; and that describes the clinician's role in clinical can4. basic knowledge and clinical competence cer prevention activities. This handbook was required for board eligibility are compoproduced for the Physician Oncology Education nents of the course. Program (POEP) of the Texas Medical Association by the American Cancer Society and BayFew preventive oncology courses exist that lor College of Medicine with funds from the incorporate all of these features.1'3'4 Since 1987, Texas Cancer Council.5 It includes sections on The University of Texas M. D. Anderson Can- patients' attitudes about prevention and on doccer Center has offered a short course in clini- tor-patient communication. It defines types of cal cancer prevention to internal medicine prevention, contrasts preventive and diagnostic residents, interns, and medical students doing a medicine, and examines ways that physicians one-month clinical oncology rotation on its med- can integrate prevention into their practices. The ical teaching service, and to family practice res- handbook summarizes risk factor information idents who participate in an elective rotation in and includes information to give patients about cancer medicine. The content of the preventive self-examinations. A section also addresses the oncology training materials presented to the res- psychosocial issues related to screening for and idents is reviewed here. No attempt has been diagnosis of malignancy. A chapter on special made to assess the impact of this training on the populations provides information addressing house staff who have participated. Others have unique problems in the elderly, children and adevaluated the impact of cancer prevention edu- olescents, ethnic groups, the poor, and those cation on the attitudes and behaviors of medi- with a family history of malignancy. The handcal students3'4, but, to our knowledge, only lim- book concludes with a series of case studies in ited assessment of house staff training in pre- a question-and-answer format that are designed ventive oncology has occurred.1 both to illustrate and to integrate the preventive concepts presented in the didactic chapters. The concluding section also presents information not COURSE FORMAT AND CONTENT presented elsewhere in the handbook such as the American Cancer Society's seven warning sigThe course has three major objectives for nals for cancer. This handbook is given to the participants: internal medicine house staff and students at the completion of the didactic sessions. 1. understand the principles of clinical epiFamily practice residents in the cancer elecdemiology as they pertain specifically to primary and secondary cancer prevention; tive receive a more detailed clinical reference 2. analyze and quantify screening test per- manual entitled ReCap: Recommendations for Cancer Prevention.6 It was developed by Dr. formance; and 3. weigh objectively the evidence supporting Margaret Spitz of the Department of Cancer Preor refuting a recommendation for primary vention and Control, University of Texas M. D. and secondary cancer prevention by ana- Anderson Cancer Center with funding from the tomic site, and clarify existing miscon- POEP. The several hundred pages include an overview of descriptive cancer etiology and a ceptions about these recommendations. An effective preventive oncology course has several features:
Clinical cancer prevention curriculum
Assume: Sensitivity-0.80, specific:ity-0.99 Disease Yes No Totals •
48
999
1.047
12
98,941
98,963
Test
60
99,940
100.000
400
995
1.396
100
98,605
98,605
135
Predictive value of a positive test: 48 -
4.6%
-
29%
1,047
400
Test
500
99,600
100,000
1,395
Figure 1. A hypothetical test is proposed with the given sensitivity and specificity. In the upper panel of the figure, the disease prevalence is 60 per 100,000, while in the lower panel it is 500 per 100,000. The predictive value of a positive test is a fraction expressed as the number of persons whose test is positive and who have the disease divided by the total number whose test is positive.
section on screening and detection. Instructions for taking a family history are included as are data about smoking cessation, occupational exposures with associated cancer risks, and the role of nutrition in cancer epidemiology. There are organ-specific chapters for breast, cervix, colon and rectum, endometrium, head and neck, lung, skin, prostate, and testes. The content of the manual is reviewed with the family practice residents in an individual 1-hour session. The ReCap manual is not given to the medical students. Clinical epidemiology is the basic science that underlies clinical cancer prevention, and few house officers are conversant with its principles.7'8 Therefore, major emphasis in the course is placed on the introduction and review of these principles. The course begins with the concept of risk, and we explain the desirability of identifying risk factors for malignant disease. To do this, we examine the relationship between odds and probability and show how these mathematical parameters are related to risk. We calculate the odds for common events such as horse racing results or meteorological conditions. After this connection between probability and odds for common events is established, we introduce the odds ratio as a measure of risk. Together, the group calculates an odds ratio from hypothetical retrospective data. House staff are told that an odds ratio greater than 1 means that those with the trait or risk factor are at increased risk
for the disease, and that those with the disease have a higher prevalence of the risk factor. It is necessary to teach measures related to screening test performance, ie, sensitivity, specificity, and the predictive value of a positive test. In addition, we enumerate the features of a disease that make it amenable to screening and present the qualities of the ideal screening test. We show that, for a given test, any increase in screening-test sensitivity must come at the expense of specificity and vice versa. We explore the effect of disease prevalence on screening test performance (Figure 1), and also show that by applying a screening test to a population that is at increased risk for the disease (ie, has an increased disease prevalence), the predictive value of a positive test increases dramatically without a change in the test characteristics. To generalize the epidemiologic approach, we provide illustrations from clinical situations unrelated to malignancy. We pose, for example, a situation in which two patients present simultaneously in an emergency room for the evaluation of chest pain. The first is an older middleaged male with numerous risk factors for coronary artery disease. The second is a young woman with no risk factors. Both have elevated serum levels of creatine kinase myocardial isoenzymes.9 The students are asked to predict which patient has the greater chance of having suffered a myocardial infarction and why. We use this illustration to clarify the predictable outcomes associated
136
V.G. VOGEL
with screening for disease in populations with varying risk-factor proifiles. The impact of age as a risk factor for malignancy is cited for reference. We also examine the relationship between disease incidence and prevalence (prevalence equals incidence multiplied by average duration of disease). To complete the picture, we relate the effect of the risk for a disease (expressed as the odds ratio) on the annual incidence of the disease by explaining that an odds ratio can be conceptualized as a multiplier of the age-specific annual incidence for a given disease (Table 1). For example, assume that a woman has risk factors for breast cancer that increase her risk for the disease threefold (ie, odds ratio = 3.0) and that the annual incidence of the disease in women her age is 60 per 100,000. Consequently, the annual incidence of breast cancer for a group of women her age with identical risk factor profiles will be 3 x 60/100,000 = 180 per 100,000. Using these examples, we establish the basis of using risk factors to make both primary and secondary cancer prevention recommendations by equating increased risk to increased annual incidence of disease, improved screening test performance (increased predictive value of a positive test), and greater impact of primary prevention on disease incidence.
CLINICAL APPLICATIONS At all times, the teaching emphasizes the clinical relevance of the epidemiologic principles with clinical examples illustrating our points. We provide a review of known risk factors for malignancy for anatomic sites where primary or secondary cancer prevention is possible11'12 (Table 2). Clinical conditions that increase risk are highlighted (Table 3). In this context, we explore the need for and the debate surrounding the use of difficult, painful, or expensive tests such as colonoscopy. We identify groups of patients for whom these tests may be appropriate. We mention areas of controversy regarding cancer etiology (eg, dietary fat and oral contraceptives), and we encourage participants to evaluate the emerging literature by using an objective and quantitative epidemiologic approach. Primary prevention strategies for patients or subjects at increased risk are reviewed (Table 4). With concentration on available measures (eg, smoking cessation and sun avoidance) we also highlight ongoing research in primary cancer prevention. We review the levels of evidence considered by expert panels to support or refute recommendations for screening.13'14 For each anatomic site we then relate the evidence that supports or
Table 1. Effect of increasing relative risk on the annual incidence of breast cancer
Age (years)
Baseline breast cancer incidence*
20-24 25-29 30-34 35-39 40-44 45-« 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+
1 8 25 63 112*** 159 185 227 262 297 316 327 337 312
RR = 3** 3
23 74 189 335 478 556 682 785 891 949 981 1011 937
RR = 5
RR = 10
5 39 123 315 558 796 927 1136 1308 1486 1582 1635 1684 1562
11 78 246 630 1117 1592 1854 2273 2616 2971 3165 3270 3369 3125
»Cases per 100,000; baseline rates from reference (10). **RR is relative risk compared to baseline. »»»Numbers in bold show the age in each risk category at which the annual incidence is greater than or equal to the baseline incidence in women age 40-44 years.
Clinical cancer prevention curriculum
137
Table 2. Anatomic sites for which primary or secondary prevention is established, possible, doubtful, or not recommended I.
Primary preventive interventions Established Lung Head and neck Skin Possible Breast Colon Uterine cervix Skin
H.
Secondary prevention (screening) Anatomic sites with good or fair evidence to recommend screening Breast Oral cavity Skin Thyroid (Multiple endocrine neoplasia syndromes) Uterine cervix Anatomic sites with poor evidence to recommend routine screening Colon Endometrium Prostate Testicle Urinary bladder (except with known occupational exposure) Anatomic sites with evidence to recommend against routine screening Lung (in smokers) In those occupationally exposed: (no recommendation)
refutes a recommendation for screening. As a part of this task, we estimate the sensitivity and specificity for each available cancer screening test and postulate outcomes derived when using each test in the typical clinical environment in asymptomatic individuals. We emphasize that recommendations are just that, and we make epidemiologically based arguments for initiating screening at earlier ages for individuals who are at substantially increased risk. As a further example, we use the case of a 30-year-old woman whose mother and one sister have breast cancer. This woman has at least a fourfold increase in her lifetime risk of breast cancer.15 Because a relative risk can be thought of as a multiplier of the annual disease incidence, we can show that this woman has an annual risk equal to that of a women 40 to 45 years old who has no other breast cancer risk factors. Therefore, the published screening recommendations do not apply to her because they would preclude screening until age 40. Instead, a risk-specific screening prescription is devised that incorporates her risk factor profile into avail-
able evidence about screening. We remind the house staff that screening efficiency is highest in groups at increased risk (Figure 1). Finally, we challenge the house staff to evaluate their current clinical habits and procedures in light of sound epidemiologic principles.16 Spirited discussion is encouraged as we explore clinical tradition in light of objective evidence. An evaluation of the instructor and the course content is required from the house staff and the medical students at the end of each month. During the 3 years that the course has been offered, both house staff and students have consistently given it favorable evaluations. It is frequently cited as the most useful portion of the 1-month cancer education experience in the evaluation forms. The course remains a permanent component of the house staff and medical student curriculum. DISCUSSION Training house staff to be providers of preventive oncology services is an emerging con-
138
V.G. VOGEL
Table 3. Medical conditions and drugs associated with increased risk of malignancy I.
Medical conditions Acquired immune deficiency syndrome (AIDS) Kaposi's sarcoma Non-Hodgkin's lymphoma (CNS) Dysplastic nevus syndrome Inflammatory bowel disease Oral leukoplakia Peutz-Jeghers syndrome Proliferative benign breast disease Tobacco use Uterine cervical dysplasia Xeroderma pigmentosum
H. Prior malignancy, radiation therapy, or chemotherapy Any site Bladder cancer Breast carcinoma Hodgkin's disease Lung carcinoma Squamous cell of the head and neck/upper aerodigestive tract HI. Drugs ALkylating agents Anabolic steroids Estrogen Immunosuppressive agents Methoxypsoralen Radiodiagnostics (eg, Thorotrast) IV.
V.
Malignancy associated with other malignancy Color-breast-endometrium Breast-ovary Familial syndromes Hereditary/familial breast cancer Li-Fraumeni (SBLA) syndrome Lynch syndromes I and II Multiple endocrine neoplasia syndromes (MEN 1 and 2) Polyposis and nonpolyposis colon syndromes
cept, requiring an orientation that differs from a disease- and symptom-based model. It is not likely that clinicians in training will adopt a preventive orientation if they do not acquire the basic epidemiologic knowledge that provides a framework into which the emerging cancer prevention literature can be incorporated. House staff also require a clinical model to emulate when making clinical decisions about cancer prevention. The cancer center is the ideal locus for preventive oncology training because it is here that the limits of therapy are encountered daily, This confrontation may make house staff more receptive to a comprehensive approach to the cancer problem, although we have not tested this specifically. Primary care physicians will be called on with
increasing frequency by patients to interpret news about cancer etiology and to make recommendations for primary and secondary cancer prevention. Without a firm grounding in the principles of clinical cancer epidemiology, the clinician is handicapped in his understanding, and his knowledge and recommendations cannot be complete. Our approach attempts to complement present knowledge and fill existing gaps, We are somewhat hindered by the paucity of teaching hours available during each clinical training month, and additional faculty members would be helpful. At present, there are few faculty members with clinical epidemiology expertise who can assist with this teaching, The only certain measure of the effectiveness of our methods would be a test of knowledge
Clinical cancer prevention curriculum
Table 4. Proven or hypothetical primary prevention interventions Site
Lung
Intervention Smoking cessation or avoidance
Head and neck Retinoids Smoking cessation or avoidance Smokeless tobacco cessation or avoidance Alcohol moderation Vitamin E Skin
Sun avoidance Para-amino benzoic acid Retinoids
Uterine cervix Barrier contraception Retinoids Colon
Dietary fat (decrease) Dietary fiber (increase) Calcium Difluoromethyl ornithine Prostaglandin inhibitors
Breast
Dietary fat (decrease) Tamoxifen
before and after the course, followed by an audit of subsequent practice patterns. An evaluation in the residency training clinics would be ideal, as would a written evaluation of cancer prevention knowledge. We have not done this yet, but are currently participating in the creation of an instrument that will assess knowledge of cancer detection and prevention among fourth-year medical students in Texas. When this instrument is available, we will use it to measure changes in knowledge that occur as a result of our course. We will use the data to modify both course content and structure in the future. Acknowledgment — Dr. Vogel is a recipient of both an American Cancer Society Clinical Oncology Career Development Award and a research fellowship from Mary Kay, Inc. through the Susan G. Komen Foundation, Dallas.
139
REFERENCES 1. Love RR, Stone HL, Hughes B: Education in cancer prevention for primary care clinicians. Med Pediatr Oncol 14:30-35, 1986. 2. Kuhn T: The Structure of Scientific Revolutions. Chicago: University of Chicago Press; 1970. • 3. Chamberlain RM, Lane M, Weinberg AD, et al.: Application of cancer prevention knowledge: A longitudinal follow-up study of medical students. J Cancer Educ 2:92-106, 1987. 4. Peters AS, Schimpfhauser FT, Cheng J, et al.: Effect of a course in cancer prevention on students' attitudes and clinical behavior. J Med Educ 62:592-600, 1987. 5. Physician Oncology Education Program: Cancer Control: The Physician's Role. Austin: Texas Medical Association; 1989. 6. Spitz MR, Newell GR: Early cancer detection: Update for primary-care physicans. Tex Med 86:44-47, 1990. 7. Fletcher RH, Fletcher SW, Wagner EH: Clinical Epidemiology—The Essentials. Baltimore: Williams and Wilkins; 1982. 8. Last JM: Epidemiology and health information. In: Last JM (ed): Public Health and Preventive Medicine. Norwalk: Appleton-Century-Crofts; 1986: 9-74. 9. Sackett DL, Haynes RB, Tugwell P: Clinical Epidemiology: A Basic Science for Clinical Medicine. Boston: Little, Brown and Company; 1985. 10. Eddy DM. Screening for breast cancer. Ann Intern Med 111:389-399, 1989. 11. Doll R, Peto R: The causes of cancer. J Natl Cancer Inst 66:1191-1308, 1981. 12. Schottenfeld D, Fraumeni JF, Jr. (eds): Cancer Epidemiology and Prevention. Philadelphia: W. B. Saunders Co.; 1982. 13. Canadian task force on the periodic health examination: The periodic health examination. Can Med Assoc J 121:1193-1254, 1979. 14. US Preventive Services Task Force: Guide to Clinical Preventive Services—An Assessment of the Effectiveness of 169 interventions. Baltimore: Williams and Wilkins; 1989. 15. Gail MH, Brinton LA, Byar DP, et al: Projecting individualized probabilities of developing breast cancer for white females who are being examined annually. J Nati Cancer Inst 81:1879-1886, 1989. 16. Oboler SK, LaForce FM: The periodic physical examination in asymptomatic adults. Ann Intern Med 110: 214-226; 1989.
ISSN: 0885-8195 (Print) 1543-0154 (Online) Journal homepage: http://www.tandfonline.com/loi/hjce20
A clinical cancer prevention curriculum in a comprehensive cancer center Victor G. Vogel MD, MHS To cite this article: Victor G. Vogel MD, MHS (1991) A clinical cancer prevention curriculum in a comprehensive cancer center, Journal of Cancer Education, 6:3, 133-139 To link to this article: https://doi.org/10.1080/08858199109528109
Published online: 01 Oct 2009.
Submit your article to this journal
Article views: 2
View related articles
Citing articles: 1 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=hjce20
J. Cancer Education. Vol. 6, No. 3, pp. 133-139, 1991 Printed in the U.S.A. Pergamon Press pic
0885-8195/91 $3.00 + .00 © 1991 American Association for Cancer Education
MINISYMPOSIUM ON TEACHING CLINICAL DECISION MAKING
A CLINICAL CANCER PREVENTION CURRICULUM IN A COMPREHENSIVE CANCER CENTER VICTOR
G.
VOGEL,
MD, MHS*
Abstract — Interns, residents, and medical students who spend one month at a comprehensive cancer center are presented with a strategy to incorporate cancer preventive services into their outpatient practices. This is accomplished through didactic lectures in clinical epidemiology. Terms that are essential to an understanding of disease occurrence are defined including incidence, prevalence, and mortality rate. Odds and probability are explored with the introduction of the odds ratio and the identification of risk factors. Primary and secondary cancer prevention are defined, and parameters related to cancer screening are introduced (sensitivity, specificity, prevalence, and positive predictive value). Risk factors for malignancy are identified where they are known and can be modified. Primary prevention strategies are reviewed by site along with data that support or refute recommendations for site-specific screening. Medical conditions and risk factor profiles that define high-risk groups are reviewed. Trainees are given copies of the US Preventive Services Task Force screening recommendations and a clinical handbook. The impact of this information on clinical practice patterns remains to be evaluated.
INTRODUCTION
to carcinogens. Love and his colleagues have written that if there is to be a shift toward increased practice of clinical preventive medicine, there should be concomitant emphasis upon teaching in basic and clinical disciplines relevant to this field.1 Problems arise, however, in identifying available time during clinical house staff training to schedule the necessary didactic material and practical experience that provides the appropriate increased emphasis. Additionally, a paradigm shift away from the curative model with its biotechnical frame of reference and a reorientation to clinical prevention is required.2 One environment in which a change in orientation can be initiated is the comprehensive cancer center where many house staff receive a portion of their clinical training. Much of this training is concentrated in the care of chronically or acutely ill cancer patients with limited life expectancies for whom the prevention of malignancy is not a consideration. When viewed from another perspective, this setting, with its population of patients who have not responded to conventional therapies, is an ideal venue for the preventive oncologist to interact with house staff for the purpose of teaching clinical cancer prevention. It is here that they are made aware of the need for alternative strategies.
Knowledge held by internal medicine and family practice residents about cancer prevention has not been systematically evaluated. During clinical training there are few opportunities for residents to acquire clinical cancer prevention skills. Once these skills are acquired, there are few opportunities to incorporate them into clinical practice. An understanding of clinical epidemiology is the foundation on which clinical cancer prevention is built, yet few medical schools and even fewer clinical residency programs provide training in the basic concepts of cancer epidemiology. As a result of this lack of training, house staff may be inclined to perceive cancer prevention in a unidimensional orientation. That is, a house officer may believe that cancer prevention involves only activities such as those related to behavior modification, or only those concerned with limiting environmental exposures This paper was presented, in part, at the annual meeting of the American Association for Cancer Education, Cincinnati, Ohio, October 5, 1990. *Assistant Professor of Medicine, and Epidemiology, University of Texas, M. D. Anderson Cancer Center; Assistant Professor of Epidemiology, University of Texas School of Public Health, Houston/Texas. Reprint requests to: Victor G. Vogel, MD, MHS, 1515 Holcombe Boulevard, Box 501, Houston.TX 77030.
133
134
V.G. VOGEL
Once each month the course is presented in 90-minute didactic sessions during which lunch is provided for the house staff and medical stu1. it is required of all primary care house dents by drug company sponsors, who rotate this responsibility on a regular basis. Attendance staff; is required of internal medicine house staff and 2. it has a defined content contained in a medical students, as well as the family practice printed text; house staff. Handouts are provided that summa3. the content addresses deficiencies in knowlrize the presented information and cite appropriedge of epidemiology and biostatistics that ate references. Participants receive a handbook are common among house staff; and that describes the clinician's role in clinical can4. basic knowledge and clinical competence cer prevention activities. This handbook was required for board eligibility are compoproduced for the Physician Oncology Education nents of the course. Program (POEP) of the Texas Medical Association by the American Cancer Society and BayFew preventive oncology courses exist that lor College of Medicine with funds from the incorporate all of these features.1'3'4 Since 1987, Texas Cancer Council.5 It includes sections on The University of Texas M. D. Anderson Can- patients' attitudes about prevention and on doccer Center has offered a short course in clini- tor-patient communication. It defines types of cal cancer prevention to internal medicine prevention, contrasts preventive and diagnostic residents, interns, and medical students doing a medicine, and examines ways that physicians one-month clinical oncology rotation on its med- can integrate prevention into their practices. The ical teaching service, and to family practice res- handbook summarizes risk factor information idents who participate in an elective rotation in and includes information to give patients about cancer medicine. The content of the preventive self-examinations. A section also addresses the oncology training materials presented to the res- psychosocial issues related to screening for and idents is reviewed here. No attempt has been diagnosis of malignancy. A chapter on special made to assess the impact of this training on the populations provides information addressing house staff who have participated. Others have unique problems in the elderly, children and adevaluated the impact of cancer prevention edu- olescents, ethnic groups, the poor, and those cation on the attitudes and behaviors of medi- with a family history of malignancy. The handcal students3'4, but, to our knowledge, only lim- book concludes with a series of case studies in ited assessment of house staff training in pre- a question-and-answer format that are designed ventive oncology has occurred.1 both to illustrate and to integrate the preventive concepts presented in the didactic chapters. The concluding section also presents information not COURSE FORMAT AND CONTENT presented elsewhere in the handbook such as the American Cancer Society's seven warning sigThe course has three major objectives for nals for cancer. This handbook is given to the participants: internal medicine house staff and students at the completion of the didactic sessions. 1. understand the principles of clinical epiFamily practice residents in the cancer elecdemiology as they pertain specifically to primary and secondary cancer prevention; tive receive a more detailed clinical reference 2. analyze and quantify screening test per- manual entitled ReCap: Recommendations for Cancer Prevention.6 It was developed by Dr. formance; and 3. weigh objectively the evidence supporting Margaret Spitz of the Department of Cancer Preor refuting a recommendation for primary vention and Control, University of Texas M. D. and secondary cancer prevention by ana- Anderson Cancer Center with funding from the tomic site, and clarify existing miscon- POEP. The several hundred pages include an overview of descriptive cancer etiology and a ceptions about these recommendations. An effective preventive oncology course has several features:
Clinical cancer prevention curriculum
Assume: Sensitivity-0.80, specific:ity-0.99 Disease Yes No Totals •
48
999
1.047
12
98,941
98,963
Test
60
99,940
100.000
400
995
1.396
100
98,605
98,605
135
Predictive value of a positive test: 48 -
4.6%
-
29%
1,047
400
Test
500
99,600
100,000
1,395
Figure 1. A hypothetical test is proposed with the given sensitivity and specificity. In the upper panel of the figure, the disease prevalence is 60 per 100,000, while in the lower panel it is 500 per 100,000. The predictive value of a positive test is a fraction expressed as the number of persons whose test is positive and who have the disease divided by the total number whose test is positive.
section on screening and detection. Instructions for taking a family history are included as are data about smoking cessation, occupational exposures with associated cancer risks, and the role of nutrition in cancer epidemiology. There are organ-specific chapters for breast, cervix, colon and rectum, endometrium, head and neck, lung, skin, prostate, and testes. The content of the manual is reviewed with the family practice residents in an individual 1-hour session. The ReCap manual is not given to the medical students. Clinical epidemiology is the basic science that underlies clinical cancer prevention, and few house officers are conversant with its principles.7'8 Therefore, major emphasis in the course is placed on the introduction and review of these principles. The course begins with the concept of risk, and we explain the desirability of identifying risk factors for malignant disease. To do this, we examine the relationship between odds and probability and show how these mathematical parameters are related to risk. We calculate the odds for common events such as horse racing results or meteorological conditions. After this connection between probability and odds for common events is established, we introduce the odds ratio as a measure of risk. Together, the group calculates an odds ratio from hypothetical retrospective data. House staff are told that an odds ratio greater than 1 means that those with the trait or risk factor are at increased risk
for the disease, and that those with the disease have a higher prevalence of the risk factor. It is necessary to teach measures related to screening test performance, ie, sensitivity, specificity, and the predictive value of a positive test. In addition, we enumerate the features of a disease that make it amenable to screening and present the qualities of the ideal screening test. We show that, for a given test, any increase in screening-test sensitivity must come at the expense of specificity and vice versa. We explore the effect of disease prevalence on screening test performance (Figure 1), and also show that by applying a screening test to a population that is at increased risk for the disease (ie, has an increased disease prevalence), the predictive value of a positive test increases dramatically without a change in the test characteristics. To generalize the epidemiologic approach, we provide illustrations from clinical situations unrelated to malignancy. We pose, for example, a situation in which two patients present simultaneously in an emergency room for the evaluation of chest pain. The first is an older middleaged male with numerous risk factors for coronary artery disease. The second is a young woman with no risk factors. Both have elevated serum levels of creatine kinase myocardial isoenzymes.9 The students are asked to predict which patient has the greater chance of having suffered a myocardial infarction and why. We use this illustration to clarify the predictable outcomes associated
136
V.G. VOGEL
with screening for disease in populations with varying risk-factor proifiles. The impact of age as a risk factor for malignancy is cited for reference. We also examine the relationship between disease incidence and prevalence (prevalence equals incidence multiplied by average duration of disease). To complete the picture, we relate the effect of the risk for a disease (expressed as the odds ratio) on the annual incidence of the disease by explaining that an odds ratio can be conceptualized as a multiplier of the age-specific annual incidence for a given disease (Table 1). For example, assume that a woman has risk factors for breast cancer that increase her risk for the disease threefold (ie, odds ratio = 3.0) and that the annual incidence of the disease in women her age is 60 per 100,000. Consequently, the annual incidence of breast cancer for a group of women her age with identical risk factor profiles will be 3 x 60/100,000 = 180 per 100,000. Using these examples, we establish the basis of using risk factors to make both primary and secondary cancer prevention recommendations by equating increased risk to increased annual incidence of disease, improved screening test performance (increased predictive value of a positive test), and greater impact of primary prevention on disease incidence.
CLINICAL APPLICATIONS At all times, the teaching emphasizes the clinical relevance of the epidemiologic principles with clinical examples illustrating our points. We provide a review of known risk factors for malignancy for anatomic sites where primary or secondary cancer prevention is possible11'12 (Table 2). Clinical conditions that increase risk are highlighted (Table 3). In this context, we explore the need for and the debate surrounding the use of difficult, painful, or expensive tests such as colonoscopy. We identify groups of patients for whom these tests may be appropriate. We mention areas of controversy regarding cancer etiology (eg, dietary fat and oral contraceptives), and we encourage participants to evaluate the emerging literature by using an objective and quantitative epidemiologic approach. Primary prevention strategies for patients or subjects at increased risk are reviewed (Table 4). With concentration on available measures (eg, smoking cessation and sun avoidance) we also highlight ongoing research in primary cancer prevention. We review the levels of evidence considered by expert panels to support or refute recommendations for screening.13'14 For each anatomic site we then relate the evidence that supports or
Table 1. Effect of increasing relative risk on the annual incidence of breast cancer
Age (years)
Baseline breast cancer incidence*
20-24 25-29 30-34 35-39 40-44 45-« 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+
1 8 25 63 112*** 159 185 227 262 297 316 327 337 312
RR = 3** 3
23 74 189 335 478 556 682 785 891 949 981 1011 937
RR = 5
RR = 10
5 39 123 315 558 796 927 1136 1308 1486 1582 1635 1684 1562
11 78 246 630 1117 1592 1854 2273 2616 2971 3165 3270 3369 3125
»Cases per 100,000; baseline rates from reference (10). **RR is relative risk compared to baseline. »»»Numbers in bold show the age in each risk category at which the annual incidence is greater than or equal to the baseline incidence in women age 40-44 years.
Clinical cancer prevention curriculum
137
Table 2. Anatomic sites for which primary or secondary prevention is established, possible, doubtful, or not recommended I.
Primary preventive interventions Established Lung Head and neck Skin Possible Breast Colon Uterine cervix Skin
H.
Secondary prevention (screening) Anatomic sites with good or fair evidence to recommend screening Breast Oral cavity Skin Thyroid (Multiple endocrine neoplasia syndromes) Uterine cervix Anatomic sites with poor evidence to recommend routine screening Colon Endometrium Prostate Testicle Urinary bladder (except with known occupational exposure) Anatomic sites with evidence to recommend against routine screening Lung (in smokers) In those occupationally exposed: (no recommendation)
refutes a recommendation for screening. As a part of this task, we estimate the sensitivity and specificity for each available cancer screening test and postulate outcomes derived when using each test in the typical clinical environment in asymptomatic individuals. We emphasize that recommendations are just that, and we make epidemiologically based arguments for initiating screening at earlier ages for individuals who are at substantially increased risk. As a further example, we use the case of a 30-year-old woman whose mother and one sister have breast cancer. This woman has at least a fourfold increase in her lifetime risk of breast cancer.15 Because a relative risk can be thought of as a multiplier of the annual disease incidence, we can show that this woman has an annual risk equal to that of a women 40 to 45 years old who has no other breast cancer risk factors. Therefore, the published screening recommendations do not apply to her because they would preclude screening until age 40. Instead, a risk-specific screening prescription is devised that incorporates her risk factor profile into avail-
able evidence about screening. We remind the house staff that screening efficiency is highest in groups at increased risk (Figure 1). Finally, we challenge the house staff to evaluate their current clinical habits and procedures in light of sound epidemiologic principles.16 Spirited discussion is encouraged as we explore clinical tradition in light of objective evidence. An evaluation of the instructor and the course content is required from the house staff and the medical students at the end of each month. During the 3 years that the course has been offered, both house staff and students have consistently given it favorable evaluations. It is frequently cited as the most useful portion of the 1-month cancer education experience in the evaluation forms. The course remains a permanent component of the house staff and medical student curriculum. DISCUSSION Training house staff to be providers of preventive oncology services is an emerging con-
138
V.G. VOGEL
Table 3. Medical conditions and drugs associated with increased risk of malignancy I.
Medical conditions Acquired immune deficiency syndrome (AIDS) Kaposi's sarcoma Non-Hodgkin's lymphoma (CNS) Dysplastic nevus syndrome Inflammatory bowel disease Oral leukoplakia Peutz-Jeghers syndrome Proliferative benign breast disease Tobacco use Uterine cervical dysplasia Xeroderma pigmentosum
H. Prior malignancy, radiation therapy, or chemotherapy Any site Bladder cancer Breast carcinoma Hodgkin's disease Lung carcinoma Squamous cell of the head and neck/upper aerodigestive tract HI. Drugs ALkylating agents Anabolic steroids Estrogen Immunosuppressive agents Methoxypsoralen Radiodiagnostics (eg, Thorotrast) IV.
V.
Malignancy associated with other malignancy Color-breast-endometrium Breast-ovary Familial syndromes Hereditary/familial breast cancer Li-Fraumeni (SBLA) syndrome Lynch syndromes I and II Multiple endocrine neoplasia syndromes (MEN 1 and 2) Polyposis and nonpolyposis colon syndromes
cept, requiring an orientation that differs from a disease- and symptom-based model. It is not likely that clinicians in training will adopt a preventive orientation if they do not acquire the basic epidemiologic knowledge that provides a framework into which the emerging cancer prevention literature can be incorporated. House staff also require a clinical model to emulate when making clinical decisions about cancer prevention. The cancer center is the ideal locus for preventive oncology training because it is here that the limits of therapy are encountered daily, This confrontation may make house staff more receptive to a comprehensive approach to the cancer problem, although we have not tested this specifically. Primary care physicians will be called on with
increasing frequency by patients to interpret news about cancer etiology and to make recommendations for primary and secondary cancer prevention. Without a firm grounding in the principles of clinical cancer epidemiology, the clinician is handicapped in his understanding, and his knowledge and recommendations cannot be complete. Our approach attempts to complement present knowledge and fill existing gaps, We are somewhat hindered by the paucity of teaching hours available during each clinical training month, and additional faculty members would be helpful. At present, there are few faculty members with clinical epidemiology expertise who can assist with this teaching, The only certain measure of the effectiveness of our methods would be a test of knowledge
Clinical cancer prevention curriculum
Table 4. Proven or hypothetical primary prevention interventions Site
Lung
Intervention Smoking cessation or avoidance
Head and neck Retinoids Smoking cessation or avoidance Smokeless tobacco cessation or avoidance Alcohol moderation Vitamin E Skin
Sun avoidance Para-amino benzoic acid Retinoids
Uterine cervix Barrier contraception Retinoids Colon
Dietary fat (decrease) Dietary fiber (increase) Calcium Difluoromethyl ornithine Prostaglandin inhibitors
Breast
Dietary fat (decrease) Tamoxifen
before and after the course, followed by an audit of subsequent practice patterns. An evaluation in the residency training clinics would be ideal, as would a written evaluation of cancer prevention knowledge. We have not done this yet, but are currently participating in the creation of an instrument that will assess knowledge of cancer detection and prevention among fourth-year medical students in Texas. When this instrument is available, we will use it to measure changes in knowledge that occur as a result of our course. We will use the data to modify both course content and structure in the future. Acknowledgment — Dr. Vogel is a recipient of both an American Cancer Society Clinical Oncology Career Development Award and a research fellowship from Mary Kay, Inc. through the Susan G. Komen Foundation, Dallas.
139
REFERENCES 1. Love RR, Stone HL, Hughes B: Education in cancer prevention for primary care clinicians. Med Pediatr Oncol 14:30-35, 1986. 2. Kuhn T: The Structure of Scientific Revolutions. Chicago: University of Chicago Press; 1970. • 3. Chamberlain RM, Lane M, Weinberg AD, et al.: Application of cancer prevention knowledge: A longitudinal follow-up study of medical students. J Cancer Educ 2:92-106, 1987. 4. Peters AS, Schimpfhauser FT, Cheng J, et al.: Effect of a course in cancer prevention on students' attitudes and clinical behavior. J Med Educ 62:592-600, 1987. 5. Physician Oncology Education Program: Cancer Control: The Physician's Role. Austin: Texas Medical Association; 1989. 6. Spitz MR, Newell GR: Early cancer detection: Update for primary-care physicans. Tex Med 86:44-47, 1990. 7. Fletcher RH, Fletcher SW, Wagner EH: Clinical Epidemiology—The Essentials. Baltimore: Williams and Wilkins; 1982. 8. Last JM: Epidemiology and health information. In: Last JM (ed): Public Health and Preventive Medicine. Norwalk: Appleton-Century-Crofts; 1986: 9-74. 9. Sackett DL, Haynes RB, Tugwell P: Clinical Epidemiology: A Basic Science for Clinical Medicine. Boston: Little, Brown and Company; 1985. 10. Eddy DM. Screening for breast cancer. Ann Intern Med 111:389-399, 1989. 11. Doll R, Peto R: The causes of cancer. J Natl Cancer Inst 66:1191-1308, 1981. 12. Schottenfeld D, Fraumeni JF, Jr. (eds): Cancer Epidemiology and Prevention. Philadelphia: W. B. Saunders Co.; 1982. 13. Canadian task force on the periodic health examination: The periodic health examination. Can Med Assoc J 121:1193-1254, 1979. 14. US Preventive Services Task Force: Guide to Clinical Preventive Services—An Assessment of the Effectiveness of 169 interventions. Baltimore: Williams and Wilkins; 1989. 15. Gail MH, Brinton LA, Byar DP, et al: Projecting individualized probabilities of developing breast cancer for white females who are being examined annually. J Nati Cancer Inst 81:1879-1886, 1989. 16. Oboler SK, LaForce FM: The periodic physical examination in asymptomatic adults. Ann Intern Med 110: 214-226; 1989.
