3. a .uxhah “3.5;... 7. 4 ....t 7 a: .rtdmvw .I..,...... x (1 Eu,» :.., i , A (:7 n .r . 1.} 1 a i 5. V 2 Guam Tr . A Ankh? . l‘ ‘ 5.5% «Law 5-... l .p .flmm findnb‘mm. v . 0%.!”th Mg 333??! , r93 Viv. III' LIBRARY ) Michigan State '_ 3 University 6?.6/3'M‘ This is to certify that the thesis entitled DESCRIPTION OF THE DATA CLEANING PROCESS AND DESCRIPTIVE ANALYSIS OF THE ENTIRE DATASET AND THE ABNORMALITIES FOUND IN A MULTI-SITE BREAST CANCER STUDY CONDUCTED BY MICHIGAN STATE‘UNIVERSITY. presented by NAGESH NARAYAN BORSE has been accepted towards fulfillment of the requirements for the Master of degree in EPIDEMIOLOGY Science x r , ' Sf 9.1.4} ’LL‘ A I \ A L) CV MUg/(f‘d' Majbr Professor’s Signature Tuesday, July 26, 2005 Date MSU is an Amn'native Action/Equal Opportunity Institution PLACE IN REIURN BOX to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 2105 chW DESCRIPTION OF THE DATA CLEANING PROCESS AND DESCRIPTIVE ANALYSIS OF THE ENTIRE DATASET AND THE ABNORMALITIES FOUND IN A MULTI-SITE BREAST CANCER STUDY CONDUCTED BY MICHIGAN STATE UNIVERSITY By Nagesh Narayan Borse A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Epidemiology 2005 ABSTRACT DESCRIPTION OF THE DATA CLEANING PROCESS AND DESCRIPTIVE ANALYSIS OF THE ENTIRE DATASET AND THE ABNORMALITIES FOUND IN A MULTI-SITE BREAST CANCER STUDY CONDUCTED BY MICHIGAN STATE UNIVERSITY By Nagesh Narayan Borse Data for the analytic portion of my thesis came from a study supported by the Department of Defense (DOD), Dorothy Pathak, PI, entitled “Improved Follow-up of Breast Abnormalities through Comprehensive Breast Care in Women 40 to 70 Years of Age”. This was a community-based randomized controlled trial whose aim was to enhance primary care physicians’ skills in secondary prevention, diagnosis and follow-up of abnormal findings in the control of breast cancer for woman 40 to 70 years of age. Data from all breast-related encounters were abstracted for visits between August 1, 1998 and July 31, 2000 in Microsoft Access software using four forms. Data cleaning was done using SAS Version 8 and was done primarily to identify any duplicate information and recoding required. For analysis purpose subsets were created based on intervention and control sites, age of the patient, normal and abnormal findings etc. In the final step, frequency analysis was carried out on main variables in the study. Screening rates were calculated using three methods for overall study, intervention sites and control sites. The total number of patients in the final database is 10,101 for Year 1 and 12,816 with almost 30,000 breast care entries. Two year patient based screening rate was 68% for CBE done, 42% for mammogram ordered and done and 74% for mammogram either ordered or done. The combined (CBE and Mammogram done) screening rate was 54% based on patient based screening rate, 59% based on physician based screening rate and 46% based on practice based/public health screening rate. Copyright by NAGESH NARAYAN BORSE 2005 DEDICATION To my aunt ‘A kka-aatya ’ and all those other women in developing world who could not seek any type of care and died due to undetected breast cancer. iv ACKNOWLDEGEMENTS My thesis project at the Michigan State University has contributed a lot to my knowledge and skills. Firstly, thanks to my parents Mr. Narayan Borse and Mrs. Laxmi Borse who supported my decision to come to the USA and attend my master’s degree program at the ' Michigan State University. My parents deserve a special thank for being so accommodating with my plans. I also want to thank my American parents Mr. and Mrs. York for their contribution and endless support during my master’s degree studies. Thanks to Dr. Gardiner for his continuous support and for signing a research contract with MDCH to provide me necessary financial support for my master’s degree studies. I owe a great debt to many people who provided me the necessary support to complete my thesis. Special thanks go to my advisor Dr. Dorothy Pathak for her guidance during my studies at Michigan State University. I also want to thank Dr. Henry Barry for his guidance during my project. My special thanks go to Dr. Janet Osuch, who played many important roles during this project as a teacher, as a guide and as a friend. I also want to thank her for maintaining the work-book (Bible) for this project and for providing uninterrupted supply of Earl Grey tea and healthy food. I also want to thank Dorota for creating the data dictionary for this project and many others who were involved directly or indirectly in the data collection process. Last but not least, I want to thank everyone who directly and indirectly helped me during my studies at Michigan State University. Faith, Hope, Love you need all of the above. If you want to live, then you have got to be positive. There is a rumor I have a tumor. I used to be a dancer, then I got cancer. I used to have hair all down my back, but now it's shorter than Kojak. But that is all right, Cuz I am gonna win the fight. Kristine Kirsten Poem written by a breast cancer survivor vi TABLE OF CONTENTS: LIST OF TABLES ................................................................................ viii LIST OF FIGURES ................................................................................. ix LIST OF ABBREVIATIONS ...................................................................... x INTRODUCTION ................................................................................... 1 CHAPTER 1 Literature Review ................................................................. 6 CHAPTER 2 Department of Defense Study in Nine Michigan Clinics .................... 14 CHAPTER 3 Data Construction and Cleaning ................................................ 22 CHAPTER 4 Recoding of Variables from Text ............................................... 27 CHAPTER 5 Scoring Techniques for Abnormalities ......................................... 38 CHAPTER 6 Screening Rate Calculations ..................................................... 43 CHAPTER 7 Results: Descriptive Analysis and Screening Rates ........................... 52 CHAPTER 8 Discussion ............................................................................. 59 APPENDICES APPENDIX A — CHART ABSTRACTION FORMS USED FOR DOD STUDY ........ 65 APPENDIX B — DATA DICTIONARY FOR DOD DATASETS ......................... 76 APPENDIX C — TABLES ......................................................................... 99 APPENDIX D -— FIGURES ..................................................................... 108 REFERENCES ..................................................................................... 120 vii LIST OF TABLES: Table l: A Woman’s Chances of Breast Cancer Increases with Age ....................... 100 Table 2: Five Year Survival Rate by Age ........................................................ 100 Table 3: Survival vs. Treatment Cost ........................................................... 100 Table 4: Guidelines by various organizations for Breast Cancer Screening .............. 100 Table 5: Staging and Survival Rates .............................................................. 101 Table 6: Overall Survival Rate .................................................................... 101 Table 7: American women who have had a Mammogram within past 2 Years. . . . . . 101 Table 8: List of intervention and control sites ................................................. 101 Table 9: List of site and assigned site numbers ................................................. 102 Table 10: List of number of subjects per site ................................................... 102 Table 11: Cleaning of Duplicate Data Entries ................................................. 102 Table 12: Symptom Codes ........................................................................ 103 Table 13: Formula for CBE screening rate calculation ....................................... 104 Table 14: Formula for Mammogram Ordered ................................................. 104 Table 15: Formula for Mammogram Done ..................................................... 105 Table 16: Formula for Mammogram Ordered and Done ..................................... 105 Table 17: Formula for Mammogram either Ordered or Done ............................... 106 Table 18: Formula for Combined Screening Rate ............................................. 106 Table 19: Screening rates for the DOD study .................................................. 107 viii LIST OF FIGURES: Figure 1: Cancer Incidence Rates (Age-adjusted to the 2000 US standard population) for Women, US, 1975-2000 ........................................................... 109 Figure 2: Cancer Death Rates (Age-adjusted to the 2000 US standard population) for Women, US, 1930-2000 ........................................................... 109 Figure 3: Mammography 2000 ................................................................... 110 Figure 4: Breast Cancer Deaths 1999 ........................................................... 110 Figure 5: State Specific Annual Breast Exam Guidelines .................................... 111 Figure 6: Triad of Breast Cancer Screening ................................................... 111 Figure 7: Continuum of Breast Care .......................................................... 112 Figure 8: Early Diagnosis by Mammogram ................................................. l 13 Figure 9a: DOD Study Forms used for Breast Care data abstraction ........................... .113 Figure 9b: DOD Study and Breast Care ........................................................ 114 Figure 10: Permanent Dataset Sub Setting Strategy ........................................... 115 Figure 11: Method of Screening Rate Calculation ............................................ 116 Figure 12: Comparison of the rates calculated by methods for Screening Rate ........... 116 Figure 13: Criteria for Screening Rate Calculation ........................................... 117 Figure 14: Type of Mammogram Screening Rate ............................................. l 17 Figure 15: Age distribution of Patients in the study .......................................... 118 Figure 16: Age distribution of the patients with self history of Breast Cancer ............ 118 Figure 17: Time interval between Mammogram ordered and mammogram done ........ 119 ix LIST OF ABBREVIATIONS For Variable names Please refer Appendix B — Data Dictionary for DOD Dataset ACS: American Cancer Society BCT: Breast Conserving Therapy BRFS: Behavioral Risk Factor Survey BSE: Breast Self Examination CBE: Clinical Breast Exam CDC: Center for Disease Control and Prevention Cum. Freq.: Cumulative Frequency DOD study: Department of Defense Study DK: Don’t Know FNA: Fine Needle Aspiration FNAB: Mass — Fine Needle Aspiration Biopsy E-code: Eligibility Code FPs: Family Physicians OB/GYN: Obstetrician and Gynecologist NCI: National Cancer Institute NVSS: National Vital Statistic Systems SEER: Surveillance, Epidemiology, and End Results StudyID: Study Identification Number US: United States of America Undoc: Undocumented Introduction Breast cancer is the second leading cause of cancer deaths in women after lung cancer and is the most common cancer among women, excluding non-melanoma skin cancers. According to the World Health Organization, more than one million cases of breast cancer occur worldwide annually, with some 580,000 cases occurring in developed countries (>300/100,000 population per year) and the remainder in developing countries (usually <1500/100,000 population per year), despite their much higher overall population and younger age. (1) In 2000, the last year for which global data exists, some 400,000 women died fi'om breast cancer, representing 1.6 per cent of all female deaths. (1) For United States, the American Cancer Society (ACS) estimates that in 2005, 269,730 new cases of breast cancer will be diagnosed: 211,240 invasive breast cancers and 58,490 cases of in situ breast cancer, of which, 85% will be ductal carcinoma in situ (DCIS). (2) According to the ACS, the chance that breast cancer will be responsible for a woman's death is about 1 in 33 (3%). The incidence rate of breast cancer (number of new breast cancers per 100,000 women) increased by approximately 4% during the 1980s but leveled off to 100.6 cases per 100,000 women in the 19903. (2) (Figure 1) Breast Cancer Screening in the United States Population statistics indicate that age-adjusted breast-cancer mortality rates began to decline during the early 19903 in many developed countries. For several decades before 1990, breast-cancer mortality rates in these countries had been either stable or increasing. (3) In the US, the death rates from breast cancer also declined significantly between 1992 and 1996, with the largest decreases among younger women. Medical experts attribute the decline in breast cancer deaths to earlier detection by screening and more effective treatments. (Figure 2) The Behavioral Risk Factor Survey (BRF S) 2000 presented a map with the age adjusted percentage of women aged GT 40 who reported receiving a mammogram within the past two years by States. (Figure 3) Age adjusted percentage of women aged 40 and more who reported receiving a mammogram within the past 2 years was 77%, over the target set by Healthy People 2010 of 70%. (4) The BRFS map also demonstrates a mammography utilization rate in Michigan of 82%, in excess of the Healthy People 2010 target. However, one has to be careful when interpreting self reported rates used in the BRF S study. Possible limitations of the BRFS survey is that one it excluded women living in households without a telephone another is that self-reported information about cancer screening practices may differ from information obtained from the records of healthcare providers. Persons tend to over report their use of screening and to underreport the time since their last screen. (5) In 1999, the National Vital Statistic Systems (NV SS) reported an age-adjusted death rate due to breast cancer in the US of 27.0 per 100,000 females. Individual age-adjusted death rates by State ranged from 20.5 to 30.0 per 100,000 females. Michigan is among highest breast cancer death rate states (28 per 100,000 females). The Healthy People 2010 target for death rate due to breast cancer is 22. With the exception of Utah and Alaska, no other state is near to the target set by Healthy People 2010. (Figure 4) Incidence and Survival Rate by Age Each woman's breast cancer risk may be higher or lower, depending upon several factors, including family history, genetics, age of onset of menstruation, and other factors, many of which have not yet been identified. According to the National Cancer Institute (NCI), the chance of getting breast cancer goes up as a woman gets older. The risk of breast cancer is greatest for women over age 60. (6) (Table 1) While breast cancer is less common at a young age, some studies have shown that breast carcinoma in young women is more aggressive biologically, which may explain why survival rates are lower among younger women. (7, 8) This can be supported by the ACS’s five year survival rate by age which is lower in younger women and higher in older women. (Table 2) Treatment Cost It is critical to screen for and diagnose breast cancer as early as possible. If the cancer is detected and treated at an early stage survival rates are highest and recurrence and treatment costs are lowest. Screening mammograms generally cost between $100 and $150. Most states now have laws requiring health insurance companies to reimburse all or part of the cost of screening mammograms. The overall 5-year survival for breast cancer is 85%. However, 5-year survival for women diagnosed at Stage 0 is 100% and for those with Stage I, 98%. Therefore, if all Americans participated in regular cancer screenings the overall survival rate could increase to more than 95% (9) For example a mammogram and diagnostic workup will not cost more than $200 and $2000 respectively. However, cost differences between early stage treatment and late stage treatment can range anywhere from $10,000 to $150,000. (Table 3) According to Barlow study with SEER data, at 6 months after diagnosis, the adjusted mean costs were $12,987, $14,309, $14,963, and $15,779 for mastectomy alone, mastectomy with adjuvant therapy, Breast Conserving Therapy (BCT) plus radiation therapy, and BCT plus radiation therapy with adjuvant therapy, respectively. The 1-year adjusted mean costs were $16,704, $18,856, $17,344, and $19,081, respectively, for the four groups. By 5 years, BCT was less expensive than mastectomy (P :< .001), with 5- year adjusted mean costs of $41,930, $45,670, $35,787, and $39,926, respectively. (10) Recommended Guidelines for Breast Cancer Screening: All major US medical/cancer research organizations recommend screening mammography for women aged 40 years and older. However, there is no one single recommended guideline for breast cancer screening that consists of the clinical breast examination (CBE) and mammogram. Each organization in the US has its own guidelines. (Table 4) CDC displays a US map on its website with state specific annual CBE guidelines which is an additional concern to the variations in mammography guidelines. (Figure 5) This is especially true for women living in Michigan who live in warm places like Florida for the winter. Whereas Michigan recommends annual CBE, in Florida it is recommended every two years for ages 40 — 50 years and annually exam for women aged 50 and above. Additionally these guidelines need adjustment based on women’s family history and risk factors. Chapter 1 Literature Review Clinical Preventive Medicine in Primary Care: Clinical evidence supports the value of preventive medicine, defined as the maintenance and promotion of health and the reduction of risk factors that result in injury and disease. Primary prevention activities deter the occurrence of a disease or adverse event, e. g., smoking cessation. Secondary prevention (screening) is early detection of a disease or condition in an asymptomatic stage so treatment delays or blocks occurrence of symptoms, e.g., mammographic detection of breast cancer. Tertiary prevention attempts to decrease adverse consequences of existing clinical disease, e.g., cardiac rehabilitation to prevent the recurrence of a myocardial infarction. (11) Preventive services have decreased morbidity and mortality from both acute and chronic conditions. However, these services are underutilized for numerous reasons. Barriers to their use include physician, patient, and health system factors. (11) Cancer Screening Screening makes it possible to detect cancer before the disease gives rise to symptoms. A more effective treatment could thus be offered, and patients would then have a better prognosis. (12) Early detection of malignant tumors, preferably before symptoms present, is important because the earlier the stage at diagnosis, the less chance that the cancer will spread to distant organs, the major reason for mortality from malignancy. Triad of Breast Cancer Screening Three breast cancer screening methods are commonly employed in combination: mammography, breast self examination (BSE), and Clinical Breast Examination by trained personnel (CBE). (Figure 6) Breast cancer screening by a combination of BSE, CBE and mammography is recommended by the American Cancer Society as effective in detecting abnormalities in all age groups for years 40 and above. (13) Figure 7 was created to illustrate the continuum of breast care. The key to surviving breast cancer is early detection and treatment. According to the ACS, when breast cancer is confined to the breast, the five-year survival rate is close to 100%. The early detection of breast cancer helps reduce the need for aggressive treatment and minimizes pain and suffering, allowing women to continue leading happy, productive lives. Results from large clinical trials also indicate that adjuvant systemic therapy, adjuvant radiotherapy, and screening can reduce breast cancer mortality. (14, 15) For a screening test to be effective, that test must be capable of diagnosing disease prior to it becoming symptomatic. That is, it must be capable of disease detection during the latent phase. Mammography is capable of detecting breast cancer in asymptomatic women and therefore meets the criteria for a screening test. (Figure 8) As shown by the middle portion of the graph, the portion of the latent phase during which breast cancer is detectable by mammography is termed the pre-clinical phase. Mammography is the single most effective method in obtaining the mortality reductions from screening. The overall results of the randomized controlled trials indicate that mammographic screening in women 50 and over can reduce breast cancer mortality by about 25%. (16) The CBE can be done safely by both physicians and other health professionals properly trained in the CBE technique. Screening clinical breast examination adds information at times not apparent on mammography and has been shown to detect some cancers missed by mammography. However, its sensitivity reported in randomized trials is low compared to mammography, about 54%. (17) Breast self-examination (BSE) is useful in detecting breast abnormalities in early stages. Controversies in Mammography: All randomized breast cancer screening trials have shown a reduction in breast cancer mortality in the 'invited for mammography' screening arm compared with the 'control arm' for women aged 50 years and older at randomization (overall 25%). (18) Annual screening mammography can decrease breast cancer mortality by 45% in women over fifty and 23% in women between forty and fifty years of age. (19) However, concerns about screening mammography are raised and those include questions of efficacy, high recall rates, false positives, and age at which to institute annual screening. Approximately 95% of women with abnormalities on screening mammograms do not have breast cancer with variability based on such factors as age of the woman and assessment category assigned by the radiologist. (l 7) Younger women (40-49 years) have lower mammographic sensitivity (i.e., greater proportion of cancers detected after a negative mammogram) than older women (> or =50 years). (20-22) Greater breast density explained 67.6% of the decreased mammographic sensitivity in younger women at 12 months, whereas at 24 months breast density explained 37.6% and rapid tumor growth explained 30.6% of the decreased sensitivity in younger women. (23) Breast density largely explained decreased mammographic sensitivity at 12 months, whereas rapid tumor grth contributed to decreased mammographic sensitivity at 24 months. A 12-month versus a 24-month mammography screening interval may therefore reduce the adverse impact of faster growing tumors on mammographic sensitivity in younger women. (23) Staging and Survival Rates. Staging is the process physicians use to assess the size and spread of location of a patient’s cancer at diagnosis. This information helps determine the most optimal form of treatment. Breast cancer stages range from Stage 0 (in-situ) to Stage IV (advanced, metastatic breast cancer). Breast cancer survival continues to decline after five years post-diagnosis. (Table 5 & 6) Mortality and Breast Cancer Screening A study by Aubard in 2002 observed that more widespread use of mammography screening for breast cancer led to smaller tumors being discovered during the second screening period, with less lymph node involvement and less initial metastasis. It has been shown that, at least for patients aged 50 to 70, properly organized mass screening for breast cancer led to a reduction in mortality rate. (24) Annual screening mammography can decrease breast cancer mortality by 45% in women above age 50 and 23% in women between 40 and 50 years of age. ( 19) Screening mammography reduces breast cancer mortality by about 20% to 35% in women aged 50 to 69 years and slightly less in women aged 40 to 49 years at 14 years of follow-up. (17) The mammography service screening programme in Copenhagen, Denmark showed reduction in breast cancer mortality in the screening period by 25% (relative risk 0.75, 95% confidence interval 0.63 to 0.89). For women actually participating in screening, breast cancer mortality was reduced by 37%. (25) In Switzerland, breast cancer mortality rates for Swiss national females aged 50-79 years fell between 1990 and 2000 by 25% in all regions. It has been suggested that the decrease in breast cancer mortality in Switzerland is not solely due to mammography screening but 10 also partly due to treatment developments and changes in cause-of-death coding (26) The Swedish study on the long—term effects of a screening program in women aged 40-64 years found a significant 20% reduction of breast cancer excess mortality. (27) In another Swedish Two-County Trial of mammographic screening for breast cancer, invitation to screening was associated with a reduction in deaths from all causes among breast cancer cases, consistent with high participation rates in screening. (28) Seven out of eight published randomized controlled trials found a significant decrease in breast cancer mortality among women who underwent screening mammography. The data indicated that screening mammography does indeed assist in early diagnosis, and most published studies show a significant reduction in breast cancer-related mortality in the screened population. (29) A combined analysis of data from five major screening studies indicates that annual screening of all women aged 40 and over by means of state-of—the-art mammography, with two views per breast and physical examination, could reduce breast cancer mortality by at least 40% and possibly as much as 50%. (30) In England and Wales, both screening and improvements in treatment have resulted in substantial reductions in mortality from breast cancer. Many deaths in the 19908 will reflect women diagnosed in the 19805 and early 19905, before invitation to screening was instituted. Further major effects from screening and treatment are expected, which together with cohort effects will result in further substantial reductions in mortality from breast cancer, particularly for women aged 55-69, over the next 10 years. (31) 11 Treatment of breast abnormalities The most common breast abnormality other than benign breast pain is a new lump or mass, although, even with this symptom, most breast lumps are benign. Other physical signs include a generalized swelling of part of a breast (even if no distinct lump is felt), skin irritation or dimpling, nipple pain or retraction (turning inward), redness or scaliness of the nipple or breast skin, or a discharge other than breast milk. Treatment is most successful when it is detected early, depending on the situation and the patient’s choices; treatment may involve breast conservation surgery (surgical removal of only the tumor and surrounding tissue) or mastectomy (surgical removal of the breast). Cost — Benefit Analysis When balancing the benefits of screening women for breast cancer against the harms and costs of screening, the relative reduction in the risk that will result from screening women in different age groups are important considerations. Seven randomized controlled trials provide evidence of the relative risk reduction that results from screening women in different age groups; other studies estimate the harms and costs of screening. These studies indicate that the benefit of screening, expressed as the absolute number of lives extended per 1000 women screened, increases with age and that the harm of screening, expressed as the number of follow-up procedures per cancer detected, decreases with age. Thus, the tradeoff between the benefits and the harms and costs of screening is better for older than for younger women. (32) 12 Barriers to Screening: Many published studies on breast cancer screening have identified specific barriers to screening. (11) Some of those barriers are listed as follows: Patient Factors: Young women, particularly if married or women of color have shown poor compliance with screening. Low socio-economic status, lack of education or awareness, lack of insurance coverage or embarrassment are some other factors which have cause poor patient compliance for breast cancer screening. (Table 7) Physician Factors: Physician gender, specialty group, and age category were significant predictors of breast cancer screening rate. Male physicians in young and middle age have shown poor rates of breast cancer screening.(33) For CBE screening, male physicians reported a greater barrier due to inadequate reimbursement for CBE than female physicians. (34) This may be due to issues of embarrassment. Lack of knowledge and belief in the importance of screening are other factors. Health System Factors: As shown in table 4, there is poor consensus on screening guidelines by various organizations and institutes working in this field. Hospitals with lack of screening strategies, poor utilization of reminders and poor training of physicians are factors affecting poor implementation of breast care and screening programs. Disincentives of Mammography: Poor reimbursement for mammography and high prevalence of breast cancer-related litigation are disincentives for radiologists to provide mammography services. ( 19) The public must be educated so that reasonable expectations on the benefits and limitations of mammography will develop. Concerns about screening mammography include questions of efficacy, high recall rates, false positives, and age at which to institute annual screening. 13 Chapter 2 Department of Defense Study in Nine Sites in Michigan 14 Purpose of the Department of Defense (DOD) Study The purpose of this study was to test a three-component intervention designed to enhance primary care physicians’ skills in secondary prevention, diagnosis and follow—up of abnormal findings in the control of breast cancer. The study was canied out to test an innovative educational intervention designed to optimize secondary prevention, diagnosis and follow-up of abnormal findings. It was directed at a population of physicians (residents and faculty) in which a pilot study has shown sub-optimal management of breast problems. It implemented education about breast cancer screening and management of abnormal findings. The study had three specific aims as follows: Specific Aim 1 To determine the effect of a three-component intervention consisting of educational material on comprehensive breast care; a CBE skills course, and a Chart Reminder/Guideline System on rates of CBE and mammography, documentation of findings, and timeliness and appropriateness of follow-up of abnormal findings. Specific Aim 2 To determine the immediate effect of the educational session on knowledge, attitudes and beliefs about breast cancer screening, early detection and follow-up of abnormalities detected. In addition, the effect of the Clinical Skills Course on the confidence and competence with which family physicians (FPS) and residents perform CBE will be measured. Specific Aim 3 To describe the long term effect of the educational session on knowledge, attitudes and beliefs about breast cancer screening, early detection and follow-up of abnormalities detected, as well as the long-terrn effect of the Clinical Skills Course on the confidence and competence with which FPs and residents perform CBE. 15 Type of Study A randomized controlled trial (with randomization based on the location of the residency) was designed to measure the impact of breast care training provided to physicians. During Year 1 of the study, sites were selected and randomly assigned to the intervention and control arms. The sites designated as Intervention and Control is listed in Table 8. Study Period Data was collected through chart audit during the baseline year which was from 08/01/1998 to 07/31/1999. Year 2, the post-intervention year, audited charts from 08/01/1999 to 07/31/2000. For each woman 40-70 years of age, each breast care-related encounter was abstracted. In addition, total number of office visits, irrespective of the reason, during the given time period was abstracted since each office visits can be viewed as an opportunity for the FPs to review the current status of breast cancer screening for the patient. It was determined that the relevant time period to abstract breast care activity for calculation of annual screening rate should include the 15 months prior to the last visit to the office in a given year. The auto-calculated fifteen-month intervals from the last office visit in Year 1 and Year 2 were then audited for the occurrence of breast care activity. Database Development Initially databases were created in Microsoft Access and exported in Microsoft Excel. There were 9 sites in total for this study. Data was exported in four separate sheets into a Microsoft Excel file, representing the four forms created in the Access database. The forms used for chart abstraction are included in Appendix 1. Figure 9a and 9b explains l6 how the study attempts to collect all aspects of breast care information, using all four forms for a particular patient in the study. 1. Form I - Front End Form 2. Form 11 - Visit Entry Form 3. Form III - Test Result Entry Form 4. Form IV - Follow-up Form Database development started with assigning unique identification number by site to subjects whose charts were abstracted for the study. StudyID Structure Studle Site Number Ecode Subject Number I 123456 1 2 3456 | It was anticipated that the dataset created at the end of study will be large; hence it was necessary to have a unique variable to follow the subject throughout the study period. Each patient was assigned a unique study identification number (StudyID) consisting of six digits. The first digit of the identification number matches with the number that was assigned to each study site. The nines sites in this study were assigned a unique number and those numbers were used as the first digit of the StudyID. Table 9 lists the assigned unique site number. The second digit represents the eligibility code (E-code ranging from 1 to 3) numeral, which is discussed below. The remaining four digits are consecutive numbers starting with 0001. 17 Eligibility Code (E-code) The second digit of the StudyID represented the eligibility code that identified the type of patient abstracted in year 1. The E-code had three possible values; 1, 2 or 3. For the first year, the E-code was defined based on following five criteria: 1. Is the patient a female? 2. Has the patient been seen in the last three years? 3. Was the patient’s date of birth between 8/1/1928 and 7/1/1959? 4. Has breast care been provided by a Family Practice Physician (FPS)? 5. Has the patient had any visit to FP between 8/1/98 and 7/31/99? E-code 1 To be assigned this code, the patient had to have satisfied all 5 of the above criteria. This made her eligible for having her chart abstracted. Additionally, at the intervention sites, these patients were eligible for insertion of the Chart Reminder Guideline System (CRGS) into their charts. E-code 2 Patients who did not satisfy criteria 5, i.e. there was no visit by the patient to the given Health Care Facility during the time period 8/1/98-7/31/99 (baseline year), were thus not eligible to be abstracted. At intervention sites, these patients were still eligible for insertion of CRGS into their charts. E-code 3 An E-code of 3 was assigned when the patient did not satisfy at least one of the first 4 criteria listed above. This made her chart ineligible for abstraction. 18 In Year 2, a new variable, current year eligibility code (E-code), was created which allowed for specification of the eligibility code during Year 2. Thus it was possible to identify all patients whose eligibility code changed between Yearl and Year2 (designated E-code old to E-code). Otherwise, E-code was similar to E-code old for all patients. Only patients that turned 40 during Year 2 and new patients to the practice had a new StudyID assigned in year 2 and added to our study. If the patient had an E-code of 2 or 3, after the patient identification number was assigned, the Microsoft Access program prompted the abstractor to discontinue chart audit, and go to the next patient. Form I - Front End Form This form contains general information about the patient and includes approximately 60 different descriptive variables such as the patient’s first and last name, medical record number, date of birth, abstractor’s ID, abstraction eligibility code (E-code) and date of most recent visit (DMRVis) etc. Form 1 Front End was changed in Year 2, the post intervention year, to gather new information and update previously entered information. The new questions which were added are as follows: a) Date of the very first visit to the FP and b) Any documentation that patient left practice before 7/31/00. The reason for this was to ensure that if there was an abnormality that needed to be followed there was documentation that FP physician did not have the opportunity to follow-up this abnormality since the patient left practice. 19 Criteria for the E-code were changed as follows 1. Was the patient’s date of birth between 8/1/ 1928 and 7/31/1960? (The limit on date of birth was changed to 7/31/1960 to ensure that the new group of patients who were turning 40 prior to August 1, 2000 were included in the study). 2. Has the patient had any visit to FP between 8/1/99 and 7/31/2000? Form 11 - Visit Entry Form Form 11 - Visit Entry was used to enter each breast care encounter the patient received during the 15 month interval. This form had important variables such as ‘Texttel’ and ‘Purpose’ which provided information about the type of contact and purpose of contact made. Type of Contact (Texttel) The abstractor recorded the date of each breast care activity and the type of contact made: 1 = Office visit, 2 = Office initiated phone consultation, 14 = Office response phone consultation, 3 = Patient initiated phone consultation, 4 = Screening/routine/regular mammogram, 5 = Regular Diagnostic Mammogram, 6 = Diagnostic/cone- compression/magnification mammogram, 7 = Ultrasound result, 8 = Fine needle aspiration (FNA) for cyst result; 9 = Fine needle aspiration biopsy (FNAB) result, 10 = Pathology report for radiological/image guided biopsy, 11 = Pathology report for open biopsy; 12 = Surgeon's letter, or 13 = Other. Purpose of this visit/call (Purpose) 6‘ The variable purpose of this visit/call” contained the following options: 1 = Screening/well women exam/annual exam, 2 = Presenting symptom(s), 3 = Follow-up of 20 a previous abnormality, 4 = Prompted by results of screening mammogram, 5 = Prompted by results of other test(s), 6 = Routine care/other health problems, and 8 = Other. The rest of Form 11 contains specifics of for any symptoms with which the patient presented, and the findings of CBE, entered by left and right breast. Form III - Test Result Entry Form Form 111 - test result entry form was created to note the breast care related test results. It includes the results of mammography, Cyst — Fine Needle Aspiration (FNA), Solid Mass — Fine Needle Aspiration Biopsy (FNAB), Ultrasound, and Image-guided biopsy/Open bi0psy results. For each test performed, options were provided to enter the results obtained from that test. For all of the tests, documentation of test dates was tracked very carefully using date of order, date of test performed, date of results obtained and reviewed and date when the results were given to the patient. Form IV - Follow-up Entry Form Form IV - follow-up entry is intended to record the follow-up that occurred or was recommended by the physician associated with each breast care encounter. It is divided into follow-up for normal test results, specific abnormalities, follow-up common to any abnormality, and surgeon’s letter. 21 Chapter 3 Data Construction and Data Cleaning 22 Data Construction In the dataset for each patient in the study the following numbers of forms are expected to be completed. Each patient should have an exclusive and individual “Form 1” which provides unique information about that particular patient. If the patient is eligible for the study and breast care was provided during the 15 month interval of interest, then the patient should have “Form 11” filled out for each breast encounter that was made. The number of times “Form 11” is filled out for a given patient, equals the number of times breast care occurred (“encounters”) during the fifteen-month interval of interest. Additionally patients will have “Form III” filled out for every time “Form 11” records the type of visit as a “test result”. Lastly, for every “Form 11” there will be a “Form IV” recording the follow-up recommended by the health care provider for that breast care “encounter”. Overall a patient will have 1. One “Form 1”. 2. If the patient is eligible and breast care is provided, at least one or more copies of “Form 11” will be filled out, each recording a specific type of breast care “encounter” 3. If “Form 11” describes the breast care encounter as ordering a “test result” or actual findings on a test result, then a “Form III” describing the test result will be filled out. 4. For every “Form 11” or “Form 11 and III” combined, there should be “Form IV” describing the follow-up recommended. If information was provided in the medical chart that breast care was performed at an outside facility or by another physician such as OB/GYN, the patient was not eligible for chart abstraction, and was assigned E-code 3, but also received a special code of “6” for 23 criteria 4 “Has breast care been provided by a FP?”. Before the data cleaning process began, the number of subjects per site is shown in Table 10. Data Cleaning In general, the academic community focuses more on quantitative results of studies, especially if statistical analysis is involved. In addition, we believe in the accuracy of the computerized statistical analysis performed by the various programs that are in common use. Data processing errors can be very subtle and difficult to trace for those doing the research. The larger the study, the more difficult it is for procedures to be kept under control—thus the close supervision of the data gathering process and data cleaning is a must. Form 1 (Front End) in Excel for all sites were imported in SAS 8.0 Version to identify any duplicate entries or Study Identification Number (StudyID) duplicates. Types of Duplicates Mainly there were four types of duplicates found in the datasets. (1) Those StudyIDs with the exact same information entered more than once. We called them Exact Duplicates. This may be because on same day, the chart was abstracted more than once by the abstractor. (2) Those StudyIDs with same information entered on different date of abstraction. We called them Duplicate Entry with different dates. (3) Same StudyIDs were assigned to different people. This might have occurred due to data entry error or the same StudyIDs having been used by different abstractors to abstract the data within the same site. 24 (4) There were some subjects who were abstracted more than once who had the same Date of Birth, but a different Last Name. All information was the same for a given visit. This occurred because of a name change after marriage or divorce etc. Search for Duplicates in the Front End Forms (Form 1) Search for duplicate was carried out using SAS. Four subsets were created and printed out for the above mentioned description of duplicates. There were unique variables in the dataset which were used to identify the duplicates. These unique variables were Medical Record Number (MRNum), Date of Birth, First Name and Last Name. Treatment of Duplicates For the four different types of duplicates, a different method of treatment was applied. (1) Exact Duplicate This type of duplicate was the easiest one to identify and treat. Exact duplicates were identified based on Last Name, First Name, Date of Birth and Medical Record Number. The latest form was kept as a final record and duplicate information was deleted. (2) Subjects entered more than once with different dates of abstraction Based on unique variables such as medical record number these duplicates were identified. Other than the date of abstraction (Variable Label — Date), the Last Name, First Name, Date of Birth and Medical Record Number was the same. Front End information that was missing from the latest abstraction forrrr but present on the earliest entry was appended onto the latest form. Only the latest abstracted information form was retained. 25 (3) Same StudyID assigned to different people There were a few StudyIDs which were used for more than one subject. This was the most difficult type of duplicate to handle as the same StudyID was used in the other forms for two different people as well. Before joining data from three separate files at one of the sites, we used the original files to understand any reason why abstractors used a given StudyID for different subjects. We then created a NewStudyID for these duplicates, one for each unique patient in the dataset, which would be present on all four forms. (4) Same subjects with different StudyIDs Based on Last Name, First Name, Date of Birth and Medical Record Number, duplicate entries for the same person with different StudyIDs were identified. The reason for this duplicate entry were: 1) a typographical error in last name or first name which created new StudyID, 2) the subject changed her first name and or last name, 3) for the site with two clinics a given subject could have separate files thus was abstracted twice. Only one entry per person was kept in the Front End form. However, necessary changes were made in the other forms to join the visits to same person’s record, if appropriate. Except Site 1 where there were two different clinics for a site where more than one abstractor abstracted charts at these clinics, other datasets had fewer numbers of duplicates. A list of the number of patients in the new data sets created after cleaning duplicates, along with the number of duplicate entries found per site, are presented in Table 11. 26 Chapter 4: Recoding of variables in Text 27 Recoding of variables in Text Initially to separate normal findings fi'om abnormal findings, selected variables were used. However after doing this separation of normal and abnormal we found that when variables named as “Other” for a given section were coded (yes) there was additional information contained in the next variable that prompted abstractors to “specify”. This data structure was put in place to ensure that if abstractor was not able to interpret the results/ findings of exam or test result in order to enter it as a normal or abnormal finding they had an option to provide a description of the finding. There were four variable/tests (two subcategories to categorize left and right breast) which had this text information entered. This information needed to be coded back into numeric data. These variables are as follows (Please refer attached Appendix A and B for variable names and details) 1. For Symptoms: If Syrnfinol and/or Symfinor were coded ‘1’ (Yes = 1), then SYMOTHER and SYMOTHERR were recoded. 2. For CBE Test results: If LCBEfino and/or RCBEfino were coded ‘1’ (Yes = 1), then LOTHERA and ROTHERA were recoded. 3. For Mammogram Test results a. If Mamfinol and/or Mamfinor were coded ‘1’ (Yes = 1), then MAMFINLS and MAMFINRS were recoded. b. If MamDesol and/or MamDesor were coded ‘ l ’ (Yes = 1), then MamDesLS and MamDesRS were recoded. Using SAS version8, text which was entered for these variables was exported into Microsoft Excel sheets. New codes for numerical coding of this text were created. New codes created for these variables are as follows: 28 1. For Symptoms two new codes were created. a. One code was created as CODEL and CODER to identify normal and abnormal results for symptoms based on text entered in SYMOTHER and SYMOTHERR variables. This code had two values ‘0’ and ‘1’. If the symptom was abnormal, then a code of ‘1’ was used for CODEL or CODER. b. CODESYML and CODESYMR was another code created to identify the exact type of symptom. This was useful to identify abnormalities by type of symptom. The major codes for this variable are as follows: LUMP = 1, NIPPLE DISCHARGE = 2, SKIN CHANGE = 3, PAIN = 4 and OCCULT = 5. If the symptom had no abnormal presentation in the text, then no code was assigned. However this code was also extended to add other descriptions found in the text. Refer Table 12 for details of the codes and its description. 2. For CBE LCBE-CODE and RCBE-CODE were the new codes created for text entered in variables LOTHERA and ROTHERA. LCBE-CODE and RCBE-CODE were coded as ‘0’ for normal CBE and ‘1’ for abnormal CBE. 3. For Mammogram 3. Based on Mammogram impression, MAMFINLS and MAMFINRS were recoded into ML_Code and MR_Code. 29 i. ML_CODE and MR_CODE / I b. Mammogram 0 is further work needed 1 is Category I — Normal or No Finding 2 is Category H — Normal/Benign Appearing 3 is Category III - Probably Benign/ Possibly Malignant 4 is Category IV - Suspicious for Malignancy 5 is Category V - Malignant until proven otherwise 1111 was used for findings that were missing 999 was used when a mastectomy was done (and mammography therefore impossible). Finding Description had text variables MAMDESLS and MAMDESRS which were coded into CODE_DESLS and CODE_DESRS. 0 is further work needed/from GComment 1 is Category I - Normal or No Finding 2 is Category II - Normal/Benign Appearing 3 is Category HI — Probably Benign/ Possibly Malignant 4 is Category IV — Suspicious for Malignancy 5 is Category V - Malignant until proven otherwise Abnormalities found from Follow-up Form There were some patients who were not confirmed as having abnormality based on clinical visits but the physician recommended a follow-up that he/she considered appropriate for resolution of this abnormality. Therefore we also looked at the recommended follow-up to identify additional patients with potential abnormalities. This 30 included patients that had in their follow-up recommendation for an immediate work up such as Extra Mammography views, (Cone compression and Magnification views), there was an interval follow-up for a Mammogram or CBE, or the patient was asked to undergo Ultrasonography or had a surgical referral letter. The General Comment which was in the Form IV Follow-up form (GComment) was another useful text variable which was used to understand why there was any additional work up or any other abnormal finding not coded in the appropriate sections. A new code was entered in CODE_DESLS and CODE_DESRS (In the Test Results form variables) as appropriate, when abnormal findings were listed in GComment. Merging New Codes with Datasets Data sets with names FINALDOD.DOD_YR1_ALLOTH and FINALDOD.DOD_YR1_ALLOTH were created for the additional coded abnormalities. Table below specifies which data sets were joined together STRATEGY USED TO ADD CODED VARIABLES TO EXISTING DATASETS E FOR YEAR 1: i FINALDOD.DOD_Y1 + SHEET="YR1" (MAMFINOLMAMFINOR CODING.xls + CBEFINOLCBEFINOR CODING.xls + SYMFINOLSYMFINOR CODING.xls + DESOTHER CODING.xls) = FINALDOD.DOD_YR1_ALLOTH FOR YEAR 2: FINALDOD.DOD_Y2 + SHEET="YR2" (MAMFINOLMAMFINOR CODING.xls + CBEFINOLCBEFINOR CODING.xls + SYMFINOLSYMFINOR CODING.xls + DESOTHER CODING.xls) = FINALDOD.DOD_YR2_ALLOTH 31 Based on StudyID and Date of Visit, recoded variables and new codes were added to the original dataset. For identification of all the abnormal findings, this was necessary in order to separate normal and abnormal findings. Using the SAS PROC SQL ‘Right Join’ function, these codes were added to existing datasets using the above described process. Creating Subsets and SAS Permanent Files In the end, final dataset had approximately 12,900 patients with almost 30,000 breast care entries. 1. Form I - Front End Form — 12,816 patients 2. Form II - Visit Entry Form — 29,623 visit entries 3. Form 111 - Test Result Entry Form — 21,147 visit entries 4. Form IV - Follow-up Form — 29,297 visit entries At the end of this process there were 45 permanent files and 405 subsets created fi'om the original dataset. These subsets were created mainly for analysis purpose. Refer to figure 10 for the explanation of the sub-setting strategy. Creating Year One and Year Two Subsets Subsets were created based on study period. Data collected during the time period from 08/01/1998 up to 07/31/1999 was considered Year One data. Year Two data, representing the post-intervention year, was considered from 08/01/1999 up to 07/31/2000. The study also had extra periods of chart abstraction. All visits prior to 8/1/98 and 3 months post 7/31/2000 (to capture up to 3 month follow-up of abnormalities detected at the end of Year Two) were categorized in file labeled “DODothers”. 32 Creating Intervention and Control Subsets Based on StudyID numbers, the entire dataset was categorized into Intervention and Control. StudyIDs with less than 600000 are from the intervention arm and StudyIDs with more than 600000 are from the control arm. Creating Normal and Abnormal subsets The dataset with all of the recoding attached was used to create patients with normal and abnormal findings. The criteria used to separate abnormalities from the entire dataset follows. Please refer to the attached Data Dictionary in the appendix IV for the variable details. In short, any patient with any symptom or abnormal CBE finding or abnormal Mammogram finding was categorized as a patient with an abnormal finding. By the end of this process, there were three datasets which were categorized based on study year (Year 1, Year 2 and DoDOther). These were further categorized as intervention and control and in the end were categorized as normal and abnormal findings. There were 21 subsets created fiom the original dataset. Criterion used to extract abnormal was as follows: If any visit had any of following abnormal findings by symptom or CBE or Mammogram, it was abstracted as an abnormal finding. As mentioned above, new codes were created which were also used to create abnormal. Abnormal Symptom: (Please refer appendix A and B for the variable name and details) SYMLUMP = Yes OR SYMDIS= Yes OR SYMCHA= Yes OR SYMPAIN= Yes OR SYMOCC= Yes OR SYMLUMPR= Yes OR SYMDISR= Yes OR SYMCHAR= Yes OR SYMPAINR= Yes OR SYMOCCR= Yes 33 Recoded Symptom: CODEL = Yes OR CODER = Yes Abnormal CBE: LCBEFLU= Yes OR RCBEFLU= Yes OR LCBEFDIS = Yes OR LCBEFOBS = Yes OR LCBEFP = Yes OR RCBEFDIS = Yes OR RCBEF OBS = Yes OR RCBEFP = Yes Recoded CBE: LCBE-CODE = Yes OR RCBE-CODE= Yes Abnormal Mammogram: MAMFINSL = Yes OR MAMFINML =Yes OR MAMFINPL= Yes OR MAMFINPR= Yes OR MAMFINSR= Yes OR MAMFINMR= Yes Recoded Mammogram: ML_CODE = Category 3 OR ML_CODE = Category 4 OR ML_CODE = Category 5 OR ML_CODE = Category 0 OR MR_CODE = Category 3 OR MR_CODE = Category 4 OR MR_CODE = Category 5 OR MR_CODE = Category 0 Recoded Mammogram Finding: CODE_DESLS= Category 3 OR CODE_DESLS = Category 4 OR CODE_DESLS = Category 5 OR CODE_DESLS= Category 0 OR CODE_DESRS= Category 3 OR CODE_DESRS = Category 4 OR CODE_DESRS = Category 5 OR CODE_DESRS= Category 0 Other types of subsets created were as follows: The dataset was subdivided into three categories based on the patient’s age. This was calculated using each woman’s date of birth from the medical record. Ages were categorized into less than 50 years of age, 50 to 59 years of age and 60 to 70 years of age. 34 For individual site comparison, subsets were created based on site from which data was abstracted. A new variable was created named as ‘Place’ which was used to categorize data by site. The abnormal subset was further classified based on the abnormality detected by type. Abnormal subsets were categorized into three nonexclusive groups: Abnormality presenting as a Symptom, an abnormal finding from the CBE, and an abnormal finding from the Mammogram (the same person could be present in all three subsets). 35 List of FINAL DATASETS: Based on Forms: FINALDODDODFRONTEND FINALDODDODVISITI F INALDODDODVISITZ FINALDODDODFOLLOWUP Based on study period year 1 and Year 2: FWALDODDODFRONTEND FINALDODDODFRONTEND_YR1 F1NALDOD.YR1_ALL F1NALDOD.YR2_ALL FINALDOD.DOD_YR1_ALLOTH F1NALDOD.DOD_YR2_ALLOTH Based on intervention and control Sites: FINALDOD.1NV_FE_Y1 FINALDODCNTRL_FE_Y1 FINALDODJNV_FE_Y2 FINALDOD.CNTRL__FE_Y2 FINALDODINVY1_ALL FINALDODCNTRLYLALL FINALDODINVY2_ALL FINALDODCNTRLYLALL FINALDOD.1NV_VISIT_Y1 FINALDODCNTRL_VISIT_Y1 FINALDOD.INV_VISIT_Y2 FINALDODCNTRL_VISIT_Y2 Based on age of patient and intervention and control site: FINALDOD.INV_YR1_LT50 FINALDOD.1NV_YR1_GT50 FINALDOD.1NV__YR2_LT50 FINALDODINV_YR2_GT50 FINALDODCNTRL_YR1_LT50 FINALDOD.CNTRL_YR1_GT50 FINALDODCNTRL_YR2_LT50 F1NALDOD.CNTRL_YR2_GT50 36 Based on age of patients: FINALDOD.FE_YR1_LT50 FINALDOD.FE_YR1_GTSO FINALDOD.FE_YR2_LT50 FINALDOD.FE__YR2_GT50 Based on study period and forms: FINALDOD.DOD_VISIT1_YR1 FINALDOD.DOD_VISIT2_YR1 FINALDOD.DOD_FUP_YR1 FINALDOD.DOD_VISIT1_YR2 FINALDOD.DOD_VISIT2_YR2 FINALDOD.DOD_FUP_YR2 FINALDOD.DOD_VISIT1_OTHER FINALDOD.DOD_VISIT2_OTHER FINALDODDOD_FUP_OTHER 37 Chapter 5: Scoring Technique for Abnormalities 38 Descriptive Analysis of Abnormal A scoring technique was used to identify type of abnormality represented during the visit or test finding. Scores were assigned to each type of abnormality based on the severity of abnormality. This was done to get a score that summarized all of the abnormalities that were observed during a given visit. Frequency of each score was obtained using SAS Proc Freq function. Scoring used for Symptom For patient who presented with a given symptom, a ‘1’ was recorded in the dataset for the variable that represented presence of that symptom on the given side of the breast. The most common symptoms that presented were Lump, Pain or Tenderness, Nipple Discharge, Skin or Nipple Change and Occult Mammographic abnormality. Coding was separate for left and right breast. For example for symptom of lump, the variable names are: SYMLUMP and SYMLUMPR for left and right breast respectively. Scoring was done as follows: 0 Lump was assigned 10000: SYMLUMP and SYMLUMPR 0 Pain was assigned 1000: SYMPAIN and SYMPAINR o Nipple Discharge was assigned 100: SYMDIS and SYMDISR 0 Skin Change was assigned 10: SYMCHA and SYMCHAR o Occult Mammogram was assigned 1: SYMOCC and SYMOCCR 39 Then the total score was calculated to obtain the total abnormal finding for that visit for that patient. SYML=SYMLUMP+SYMLUMPR SYMP=SYMPAIN+SYMPAINR SYMD=SYMDIS+SYMDISR SYMC=SYMCHA+SYMCHAR SYMO=SYMOCC+SYMOCCR Tot_SYMP=SYML+SYMP+SYMD+SYMC+SYMO; For example, a lump that presented as a symptom during a visit was given a score of 10000. .If a patient presented with a lump as a symptom in both breasts, she would be given a score of SymLump * 10000 and SymLumpR * 10000; therefore the total for that patient for that particular visit will be 20000. By this method we could calculate the total symptom presenting during that visit for each patient. Scoring was useful as it could easily identify the combinations of symptoms with which the patient presented. For example: If the total row score is 21200: That means for that particular visit there were lump on both sides, pain in the breast on one side and nipple discharge from both breasts. The same technique was used to score CBE and Mammogram. 40 For CBE, scoring was done as follows LCBEFLU1=LCBEFLU*1 0000; RCBEF LU l =RCB EFLU“ 1 0000; LCBEFP1=LCBEFP*1 000; RCBEFP1=RCBEFP*1 000; LCBEFDIS 1=LCBEFDIS*1 00; RCBEFDIS l =RCBEFDIS*1 00; LCBEF OBS l =LCBEF OBS"'1 0; RCBEFOBS 1 =RCBEFOBS" 10; CBEFL=LCBEFLU1+RCBEFLU1 ; CBEFD=LCBEFDIS 1 +RCBEF DIS 1 ; CBEFO=LCBEFOBS 1 +RCB EFOB S l ; CBEFP=LCBEFP l +RCBEFP l ; Tot_CBE=CBEFL+CBEFD+CBEFP+CBEFO; For Mammogram, scoring was done as follows (Please refer Appendix A and B for variable names and other details) MAMFINML1=MAMFINML*100000; MAMFINMR1=MAMFINMR*100000; MAMFINSL1=MAMFINSL*10000; MAMFINSR1=MAMFINSR* 10000; MAMFINPL1=MAMFINPL*1000; MAMFINPR1=MAMFINPR*1000; MAMFINBL1= MAMFINBL’IOO; MAMFINBR1= MAMFINBR‘100; MAMFINNL1= MAMFINNL*10; MAMFINNR1= MAMFINNR*10; MAMFINM=MAMFINML1+MAMFINMR1 ; MAMFINS=MAMFINSL1+MAMFINSR1; MAMFINP=MAMFINPL1+MAMFINPR1 ; MAMFINB=MAMFINBL1+MAMFINBR1; MAMFINN=MAMFINNL1 +MAMFINNR1 ; MAM=MAMFINN+MAMFINB+MAMFINP+MAMFINS+MAMFINM; The total score for abnormalities for a given visit were calculated by adding the row totals. 41 Scoring method was also applied to calculate total abnormality during that visit SAS program is as follows: SYML=SYMLUMP+SYMLUMPR SYMD=SYMDIS+SYMDISR SYMC=SYMCHA+SYMCHAR SYMP=SYMPAIN+SYMPAINR SYMO=SYMOCC+SYMOCCR CBEFL=LCB EFLU+RCB EF LU; CBEFD=LCBEFDIS+RCBEFDIS; CBEFO=LCBEFOBS+RCBEFOBS; CBEFP=LCBEFP-+RCBEFP; CBEFIN=LCBEFINO+RCBEFINO; MAMFINO=MAMFINOL+MAMFINOR; MAMFINS=MAMFINSL+MAMFINSR; MAMFINM=MAMFINML+MAMFINMR; MAMFINP=MAMFINPL+MAMFINPR ; MAMF IN L=MAMF INL+MAMF INL; MAMFINB=MAMFINBL+MAMFINBR; MAM=MAMFINO+MAMFINS+MAMFINM+MAMFINP; CBE=CBEFL+CBEFD+CBEFO+CBEFP+CBEFIN; Tot_SYMP=SYML+SYMD+SYMC+SYMP+SYMO; ABNORMAL=MAM+CBE+SYMP; 42 Chapter 6: Screening Rate Calculation 43 Screening Rate Calculation Screening rate calculations were done using three different methods and were calculated for two year study period 8/1/1998—7/31/200 for all sites combined. These methods distinguish themselves based on the denominator used for the calculation of the screening rate. (Figure 11) These three methods are as follows 0 Patient Based Screening Rate: In this method the denominator used for the screening rate calculation uses only active patients in the practice. For the purpose of two year screening rate calculation, active patients are those with E-code 1, which means that the patient is a female, has been seen in the last three years, date of birth was between 8/1/ 1928 and 7/ l/ 1960, breast care has been provided by a FPP and the patient had a visit to FP between 8/1/98 and 7/31/2000. 0 Physician Based Screening Rate: In this method, in addition to all active patients, those who were provided breast care by other specialties such as gynecologists, obstetricians or surgeons were also added to the denominator. From the eligible for screening dataset, E-code 3 (Ineligible for abstraction) and Care 6 (care provided by others) were counted and added to numerator and denominator due to this it will always be little higher than patient screening rate. a Public Health or Practice Based Screening Rate: In this method of the screening rate calculation, patients with E-code 2 who did not satisfy criteria 5, i.e. there was no visit by the patient to the given Health Care Facility during the time period 8/1/98-7/31/2000 were included in the denominator. 44 Denominator for this screening rate includes denominator of physician based screening rate and in addition to that it includes patients with E-code 2. However numerator remains same as physician based screening rate. Interpretation of different methods of screening rates Different methods for calculating screening rates lead to different rates and different interpretations. The patient-based screening rate is liberal method (yields high screening rate values) of calculation as it takes only those patients in the denominator who had a visit to the clinic in the time period of interest and in whom there was a potential for FPs to screen that women for breast care. The public health or practice based screening rate is more conservative (giving lowest values for screening rate). However the practice based screening rate which is also called Physician based screening rate provides highest screening rate values and considered as a most liberal method of calculation. In this method we also include patients who had breast care from someone other than Family physician because of this it is always higher than patient based screening rate. In the physician based approach, the main consideration is that the responsibility of physicians in practice is not only to screen those who visit the clinic, but also those active patients who do not, by reminding them of the importance of regular screening. The active patients who had no visit in the last year to the clinic would then be enticed to visit. The public health screening rate yields lowest screening rate values hence believed as a very conservative way of calculating the screening rate and it is likely that physicians will not accept these numbers readily. (Figure 12) 45 Usefulness of different methods of screening rate calculations: 0 Patient Based Screening Rate: This method is useful in determining the physician’s potential to screen patients who visit the clinic. If all women who are active patients (seen within last three years) make a visit to the clinic once a year, then the physician has a potential to have 100% screening rate. This can be reinforced to physicians by providing proper training on breast care and by stressing the importance of screening to the physicians in the practice. 0 Physician Based Screening Rate: Some patients get breast care by doctors other than F Ps such as Obstetricians, Gynecologists, or Surgeons. Under those circumstances, the FP physician only needs to note in the patient’s chart when and what type of breast care the patient is receiving. Therefore, when calculating this rate, the additional patients screened outside of the FP office are included both in the denominator and numerator of this rate. 0 Public Health or Practice Based Screening Rate: This rate is very useful in determining a local, regional, or national screening rate. This rate reflects the actual potential to screen any eligible patient in the practice. This rate can be increased by applying different screening strategies, reminder letters and awareness campaigns. 46 Types of Screening Rates: There are three different types of screening rates that can be calculated. 0 Clinical Breast Examination (CBE) Rate In this type of screening rate calculation, patients who had annual clinical breast examination are evaluated. 0 Mammogram Screening Rate There are four subtypes of screening rates calculated in a mammogram screening rate. 0 Mammogram Ordered o Mammogram Done 0 Mammogram Ordered and Done 0 Mammogram either Ordered or Done 0 Combined CBE Done and Mammogram Done Rate Process of screening rate calculation: A screening test is a test applied to an asymptomatic individual with no clinical manifestations of the disease. (35) The above definition states two main criteria which were used to calculate CBE and mammography screening rate. The first criterion was to use only asymptomatic patients for screening rate calculations, and remove individuals who presented with any breast symptom 30 days prior to the documented CBE or Mammogram. The second criterion was to remove those who had abnormal finding in CBE for screening mammogram or abnormal finding in mammogram for screening CBE. The total number of patients who are eligible for screening will be asymptomatic patients with no abnormal finding. This 47 became the denominator for the screening rate. For the schematic representation of this calculation please refer to Figure 13 in the appendix. To calculate the ntunerator for the screening rate, all the patients in the denominator were analyzed to identify how many of them had a documented screening CBE or screening mammogram. This number was used as a numerator for the screening rate calculations. CBE Screening Rate: The final data set with all recoding was used for the calculation of CBE screening rate. The first step was to identify all of the visits in which patients presented with symptoms and the date at which symptoms first presented to the clinic. In order to be eligible for CBE screening, any patient with no presentation of a symptom 30 days prior the first CBE documented were included. All other CBEs done after presentation of a symptom were ineligible in the screening CBE calculations. The next step was to identify abnormal mammograms in asymptomatic patients prior to screening CBE. Using the file of asymptomatic patients, identification of patients with any finding with category 3 or more on a mammogram was carried out. In short, the following four important numbers were required for the CBE screening rate calculation: 1) All active people with breast care 2) All those who had a symptom within 30 days of CBE 3) All those who had an abnormal mammogram anytime before CBE 4) CBE Documentation for all eligible people for screening Where 1, 2 and 3 are used for the denominator [(1 — 2) — 3] and number 4 is a numerator for the CBE screening rate calculation. (Table 13) 48 Mammogram Screening Rate: For the Mammography screening rate, four different types of screening rates were calculated. Each has its own importance and explains different dynamics in the mammography screening process. (Figure 14) o Mammogram Ordered In this type of calculation of mammography screening rate, the total number of mammograms ordered by the physician was taken into consideration. No effort was made to identify of those ordered, how many were actually performed. This is liberal method of screening rate calculation. 0 Mammogram Done In this type of mammography screening rate, the actual numbers of mammograms performed in the pertinent year are calculated. This rate shows patient compliance to a physician’s call for a mammogram. It is relatively conservative approach of screening rate calculation, as a physician may have no control over a women’s wish to get a mammogram or not. 0 Mammogram Ordered and Done This type of mammography screening rate is a very conservative way of a screening rate calculation, as those patients who had a mammogram ordered, and done, are only taken into the numerator. o Mammogram either Ordered or Done This is very liberal way to calculate a mammogram screening rate. The calculation includes a numerator which counts all mammograms either ordered or done. 49 The four types of mammogram screening rates were calculated and the formulas for the calculations are as follows: 1. The entire dataset was used to calculate the mammogram screening rate. In the first step we identified office visits and screening or regular mammogram visits which were coded as texttel=' l' or texttel='4'. 2. In the second step, identification of a screening or well women exam, routine care, or other reasons for the purpose of the visit were identified. This was done by adding following SAS part to existing code. [AND (PURPOSE='1' OR PURPOSE = '6' OR PURPOSE = '8' OR PURPOSE=' ')] 3. In the third step, any visits with symptoms 30 days prior to the mammogram were identified and the visit was excluded from the screening rate calculation. There were a few visits with abnormal mammographic findings reported that represented diagnostic rather than screening mammograms. These cases were deleted from the analysis of screening rates. Also deleted were the few mammograms that had no documented dates. (Table 14 — l7) Combined CBE Done and Mammogram Done Rate In this type of screening rate, complete breast care which includes annual CBE and mammography is calculated. Those patients who had both a CBE documented and a Mammogram done are the only patients included in numerator. The entire dataset was used to calculate the combined screening rate. (Table 18) 1. In the first step, we identified office visits and screening or regular mammogram visits which were coded as texttel='l' or texttel='4'. 50 2. In the second step, identification of a screening or well women exam, routine care, other reasons for the visit were identified. This was done by adding the following SAS part to existing code. [AND (PURPOSE='1' OR PURPOSE = '6' OR PURPOSE = '8' OR PURPOSE=' ')] Steps one and two provided the denominator for the calculation. 3. In the third step, any patients who had symptoms that presented 30 days prior to the CBE or mammogram were removed. 4. In the fourth step, the numerator was calculated. Using the same data as was used for the denominator, the number of patients with documented CBEs and Mammograms were counted. 5. In the last step we removed all those visits with no mammogram or CBE documented and ran a PROC FREQ. 51 Chapter 7: Results: Descriptive Analysis and Screening Rates 52 Descriptive Analysis: With the development of large data sets it is important to run initial descriptive analysis to detect any particular pattern shown in the collected data in terms of patients by site, age, eligibility codes and care providers. Eligibility and New Patients in Year 2: The total number of patients abstracted into the database is 10,101 for Year 1 and 12,816 for Year 2. An increase of 2,715 patients in Year 2 included 1,436 patients joining as active patients; 296 patients who did not make any visit in year 2; and, 983 patients who did not satisfy eligibility criteria for chart abstraction. Table of E-code by E—code Old: E-Code Old E-Code 1 2 3 New Patients Total Active Not Active Ineligible in Yr 2 1 = Active 5344 284 261 1436 7325 2 = Not Active 941 958 46 296 2241 3 = Ineligible 115 113 2039 983 3250 Total 6400 1355 2346 2715 12816 E-code was the eligibility code for Year 2 and E-code old was the eligibility code assigned for baseline year. Only 42% patients were active throughout the study period. Eleven percent (1,436 patients) became active in year two. However, 941 patients did not make any visit in year two and became inactive (E-code 2) in the second year. Sixteen percent of the patients were not eligible for chart abstraction for both years and 8% of patients did not make any visit for breast care in either study years. 53 Eligibility code (E-code) for Year 2: E-Code # % Cum. Freq. 1 = Active 7325 57.16 7325 2 = Not Active 2241 17.49 9566 3 = Ineligible 3250 25.36 12816 Overall, 57% of patients were active in Year 2, 25% patients were ineligible for abstraction either because of age or not meeting eligibility criteria, and 17% patients did not have any visit in second year for breast care. Patient’s Age: Age Range # % 31 TO 40 463 3.61 41 TO 50 6455 50.37 51 TO 60 3590 28.01 61 TO 70 1970 15.37 In order to be eligible for chart abstraction, the patient’s age had to be in between 40 to 70 years. Overall, 96% patients were eligible for chart abstraction based on age. (Figure 15) However, 338 2.6%) patients had either their date of birth missing or entered incorrectly. Later on it was found out that due to different versions of Microsoft Access birth years were automatically modified from 1928 to 2028. For the purpose of this analysis, these were treated as missing. 54 Table of Age Range by Eligibility code: Overall, 50% patients in the study were in the age range 41 to 50 years. Of the total active patients 47% patients were in the age range of 41 to 50 years followed by 29% who were Age Range E-code # l 2 3 Total % Active Not Active Ineligible 246 69 148 463 31 To 4° 1.92 0.54 1.15 3.61 3498 1226 1731 6455 41 To 5° 27.29 9.57 13.51 50.37 2156 595 839 3590 51 To 6° 16.82 4.64 6.55 28.01 1241 301 428 1970 61 To 7° 9.68 2.35 3.34 15.37 Miss“ 184 50 104 338 g 1.44 0.39 0.81 2.64 Total 7325 2241 3250 12816 57.16 17.49 25.36 100.00 between 51 and 60 years of age. Who provided Breast Care? Family physicians delivered breast care for 78% of the patients the Family Physician. Gynecologist or Obstetrician or Surgeon assumed responsibility for breast care in 7% of patients. Care # % Missing 10 0.08 1 = FPC 9974 77.82 6 = Other 894 6.98 9 = DK 1938 15.12 55 Personal or Family History of Breast Cancer: Table of History by ActY: ActY History N0 = 0 Yes = 1 Missing Total 0 = N 148 3560 23 3731 ”“e 1.15 27.78 0.18 29.11 1 = Y 39 1685 11 1735 “S 0.30 13.15 0.09 13.54 _ 130 1775 8 1913 8 " UM“ 1.01 13.85 0.06 14.93 _ 9 76 1 86 9 ‘ ”K 0.07 0.59 0.01 0.67 M. . 942 24 4385 5351 “mg 7.35 0.19 34.22 41.75 T H 1268 7120 4428 12816 ° 3 9.89 55.56 34.55 100.00 A personal or family history of Breast Cancer was documented for 13% of the active patients who received breast care. For 57% of cases, the family or personal history was either missing or undocumented. Breast Cancer History in Self: YSelf # % Cum. Freq. 0 = None 12559 98.06 12559 1 = Yes 238 1.86 12797 8 = Undoc 1 0.01 12798 9 = DK 10 0.08 12808 About 238 patients (2%) had a personal history of breast cancer documented in medical chart. 56 Age distribution of patients with self-history of Breast Cancer: Analysis Variable : AGE Mean Median 25th Pctl 50th Pctl 75th Pctl 52.71 57.00 51.00 57.00 64.00 SAS code (Proc means) was used to understand age distribution of patients with self- history of breast cancer. These patients had a mean age of 52.7 years and median age of 57 years. Ages 51 to 64 years covered 50% of patients with self-history of breast cancer. (Figure 16) Time Interval between Mammogram ordered and done: For patients who had mammogram ordered and done documented in medical charts, time interval was calculated. Overall, Mammogram were ordered and done within a month for 58% for overall study and 2 to 6 month was found in 29% of mammogram. (Figure 17) Overall Time Interval # % Same Day 1183 25.0 Within 1 Months 1547 32.7 2 to 6 Months 1380 29.1 7 to 12 Months 233 4.9 13 to 24 Months 394 8.3 4737 100.0 57 Screening rates for the two year study period: As explained in chapter 6 on the screening rate calculation, three different methods were applied to calculate two year screening rates. Screening rates were calculated for overall study, intervention sites and control sites. (Table 19) Overall two year CBE screening rate was 68% based on patient screening rate, 72% based on physician screening rate and 55% based on public health screening rate. As mentioned before in chapter 6, Physician based screening rate includes patients who get screened by other than FPC. Physician based rates are always higher than other methods of screening rate calculation. Two year screening rates were found higher in Intervention sites than control sites. However, for screening rates involving mammogram ordered numbers had higher rates for control sites. Later it was found out that there was a simultaneous study carried out during DOD study by the department of Family Practice at one site which was a part of control arm for DOD study. This study was focused on telephone intervention to improve mammogram screening rate at that particular site which was published in 1999 in Family Medicine. (36) This is one possible reason why these screening rates where mammogram order data included are higher in control sites. Mammogram either ordered or done screening rates for the two year study period provides highest screening rate values. Overall patient based screening rate for either ordered or done screening rate was 74%, 77% for physician based and 61% for public health or practice based screening rate. The lowest values were found for the mammogram done and ordered screening rate which were 42% Patient based, 49% Physician based and 39% Practice based respectively. 58 Chapter 8 Discussion 59 Discussion: Breast cancer is the most common malignancy among American women. (37) The American Cancer Society (ACS) estimates that in 2005, 269,730 new cases of breast cancer will be diagnosed among women in the United States: 211,240 invasive breast cancers and 58,490 cases of in situ breast cancer, of which, 85% will be ductal carcinoma in situ (DCIS). (2) Cancers of the breast will be the most frequently diagnosed cancers in American women, followed by lung cancers. (38) Due to increased screening, the majority of patients in the US present with early-stage breast cancer. Therefore it is essential to identify epidemiological and clinical issues important in breast care early detection. It is fi'uitless to screen for breast abnormalities if appropriate actions following detection are not followed. In the DOD study more than 450 variables related to epidemiological and clinical factors for breast care were collected. DOD data collected through medical chart abstraction provides a wealth of information about breast care and captures every aspect of it. Descriptive Analysis: The process of data cleaning in SAS version 8 was a very laborious and time consuming process as it included four different forms linked to each other with StudyIDs for more than 30,000 breast care visits. In the end, the total number of patients in the database is 10,101 for Year 1 and 12,816 for Year 2 with ahnost 30,000 breast care entries. At the end of the data cleaning process, there were 45 permanent files and 405 subsets created from the original dataset. Overall, 46% of patients in the study were in the age range 41 to 50 years. During the study period, only 42% of patients had breast care visits documented. Of total active patients, 25% patients were in the age range of 41 to 50 6O years. For age 41 to 70 years, about 58% patients were active and had some kind of breast care in that year. About 78% patients received breast care from the Family Physicians. Out of total visits recorded for purpose, 46% of those were Well Woman Exams and 24% was routine care visits. Family History of Breast Cancer: To date, the etiology of breast cancer is poorly understood with known breast cancer risk factors explaining only a small proportion of cases. (39) Family history has always traditionally been used to identify persons at high cancer risk and to target appropriate preventive and therapeutic measures. Claus et al in his study in 2003 concluded that a family history of breast cancer is associated with an increased risk of DCIS and LCIS, particularly among women with multiple relatives affected at early ages. (40) Study by Webers et al, showed that the first-degree female relatives of women with breast cancer were at increased risk for breast cancer (RR: 1.7, 95% CI: 1.4-1.9). (41) In the DOD study about 23% of the patients had either a personal or family history of breast cancer.. A study published in Lancet conducted collaborative reanalysis of 52 epidemiological studies and found that 12.9% women with breast cancer and 7.3% controls reported that one or more first-degree relatives had a history of breast cancer: 12% of women with breast cancer had one affected relative and 1% had 2 or more. (42) Time interval between Mammogram ordered and done: For patients who had a mammogram ordered and done documented in the medical charts, time interval was calculated. Overall, mammograms were ordered and done within a month for 58% for overall study and 2 to 6 month interval was found in 29% for overall study. 61 Breast Cancer Screening Guidelines: Most physicians agree that screening mammograms help detect breast cancer in its earliest stages, often several years before a lump can be felt. However, the debate over when women should begin receiving annual screening mammograms has been ongoing. Most physicians and cancer organizations believe that all women 50 years of age and older should have annual mammograms to help detect breast cancer. However, organizations including the American Cancer Society (ACS), the American College of Radiology (ACR), the American College of Surgeons, and the American Medical Association (AMA), recommend that women should begin receiving annual mammograms at age 40. However, National Cancer Institute (N CI) recommends annual screening beginning at age 50, while suggesting that women in their 40s have screenings every one or two years, depending on individual risk factors. In the DOD study, when recommending mammograms, physicians noted other guidelines in 17% of visits while only 5% of visits followed ACS guidelines. Screening rates for the two year study period: Overall two year CBE screening rate was 68% based on patient screening rate, 72% based on a physician screening rate and 55% based on public health screening rate. As mentioned before in chapter 6, Physician based screening rate includes patients who get screened by other than FPC. Physician based rates are always higher than other methods of screening rate calculations. Screening rates were found higher in intervention sites than control sites. However, for screening rates involving mammogram ordered numbers had higher rates for control sites. Later it was found that there was a simultaneous study carried out during the DOD 62 study by the department of Family Practice at that site. This study was focused on telephone intervention to improve mammogram screening rate at that particular site which was published in 1999 in Family Medicine. (36) This is one possible reason why the screening rates where mammogram order data included are higher in control sites. Having a mammogram either ordered or done screening rates for the two year period provide the highest screening rate values. Overall, the two year patient based screening rate for either ordered or done screening was 74%, 77% for physician based and 61% for public health or practice based screening rate. The lowest values were found for the two year mammogram done and ordered screening rate which were 42% Patient based, 49% Physician based and 39% Practice based respectively. Implications: Beginning at the age of 20, every woman should practice monthly breast self-exams and begin a routine program of breast health, including scheduling physician performed clinical breast exams at least every three years. As a woman ages, her risk of breast cancer also increases. Beginning at the age of 40, all women should have annual screening mammograms, receive clinical breast exams each year, and practice breast self- exams every month. 63 Limitations: Results presented in the thesis are preliminary findings from the data cleaning process carried out on the DOD dataset. These results should not be quoted or used as the final results. Further data cleaning is necessary and is in progress. Final results for the DOD study will be published in firture. 64 Appendix A Chart Abstraction Forms Used for DOD Study 65 Form I- Front-End Form M I Patient Name (Last): (First): Medical Record Number. Add New Patient Date Of Birth: Abstractor's ID: Lname Fname Data 1 1 11 1 1 “1,1900 Transfer 11 Eligibility Criteria:Check One Item For Each Statement (1 -5) 1. Patient gender is: Female 2. Patient has been seen in last three years Yes 3. Patient birthday is between August 1, Yes 1928 and July 1, 1959 4. Breast health care provided by FPC Provider 5. Active patient between 811/98-7l31199 Yes Click to Determine Eligibility Code: I Rules for Assigning Study ID: Meaning of Eligibility Code: For site number 1-5: 1= Eligible for abstract and insertion 2= Eligible for insertion only 3= Ineligible For site number 6-9: 1= Eligible for abstract 2 or 3= Ineligible Study ID is a 6-digit number. The first digit is your site number. The second digit is the Eligibility code shown in the box above. The rest four digits are consecutive numbers starting 0001. Please assign study ID: 100000 Today's Date: For your reference, please look in the box on the right, find out what was the last number assigned for that specific eligibility category, and use the next consecutive number. 11/11/2000 For eligibility code = For eligibility code = For eligibility code = Click here To Add New 13......“ 1 Continue , Chart Review Form (OnIv For Eligible Patient) Study ID: 100000 1. Date of Most Recent Office Visit (MMIDDIYY): 11/11/1911 2. Autocalculated Date For the Last Eligibiie Visit Within the Last 15 months (MM/DD/YY): 8/11/1910 3. Total Number of Visits Within 15 Months. Including The Most Recent Visit: 1 4. Was A Breast Care Performed During Any of The Visits Within The 15 Months Period: Yes 66 5. PersonaIIFamlly History Of Breast Cancer? Add New Patient I None Rule for filling In the age at diagnosis: 1) Fill in exact age when information is availabe; 2) Fill in '777‘ if only known Pre«menopausal equal to or less than 50 years ol 3) Fill in ‘888' if only known Post-menopausal or greater than 50 years old; 4) Fill in ’999' if no information is available. In Self? No Age: Surgery/Reconstruction: C] Complete Breast Removal (:1 Partial Breast Removal/Lumpectomy D Prophylactic Implant [:1 Autoiogous Reconstitution [:1 Other, specif [:J Undocumented Treatments (check all that apply) [:1 Chemotherapy [:J Radiation [:1 Tamoxifen/Nolvadex [:1 Alternative medicine(s), specify El Other, specif [:I Undocumented In Mother? No Age: In Sister? N° [3 Sister1 Age: [3 Sistem Age: In Daughtefl NO D Daughtefl A992 CI Daughter2 Age; In Other ROISIIVBS? NO Phase specify; BOX-A Record information for patients each visit when a breast care was performed. Start with the first visit when any breast care activity was Countinue to record recorded during that 15 months period. Click the button on the right to ‘ visit info. continue. Go To First Patient I Go To Previous Patient I Go To Next Patient] Go To Last Patien J (Click Any of the Buttons Above to Navigate the Record) 67 Form II- Visit Ent «mm! I ell-«I ........I a... command! Phase fill out Question 6 and Question 1 for every visit/call. 6. Date of Breast Care Activity Was Recor #Name? If this visit is about a test result, you can directly go Type of Cont to Test Result Form, without filling out CBE 7. Purpose of this Visit! documentation 8'" mama? r I "5‘3’. :N\"=3"v..»-.:;?V\’. "' .".-1‘ fits;- a. Who Perfomted Breast Care/Phone Consultation? (Check All That Apply) Resident Physician Faculty Physician Physicln Assistant Nurse Practitioner Undocunented 9. Patient Presenting Symptoms/Signs (Check All That Apply) Which breastIs) has presenting symptom? If you don't know which breast. please record information In “Left Breast“ category. Left Breast: Right Breast: None UndocunentediDon'tknow None Undocumntcdlbon'tltnow Lump(s)/Mass(es)/Asymmetrical thickening Lump(s)/Mass(es)/Asymmetrical thickening Nipple Discharge Nipple Discharge Skin/Nipple change (check all that apply) Skin/Nipple change (check all that apply) 25232 Sign Dangling; it.)y'§?=f:"=3‘.‘t‘:'.9i‘t:l‘. :‘ttir.'.5<:;‘.:::=tg,' Six-2.”. wimp-5:11 ¥;r.tiinl'r=.'s.'i’53-:r: 552:1: Emmi: Haggis-2 {titres-:1 :cr: reg-size? foresees; ‘1..2<£.s.-523:.t:0:a 5953558 :icitizrt Pain/Tendemess Painfl'endeme Occult Mammographlc Abnormality Occult Mammographic Abnormality {'L-ersssétyiz‘iccitsls: 112‘ Payer-1818513.? {35513113‘,*{3*\5<21§E:E-3 or 53.5-j,‘5535‘<'::2L-t1'}-‘L: retain-camim.aims. iy-ii:..‘.=.'i;:f.§i-§'-“§§§-'::,1‘.:.'::’:~'e- Other, specify. #Name? Other, specify: #Name? 10. CBE Documentation: 11. CBE Findings (Check All That Apply): Bilateral hrtplants Previous abnonnaiity resolved i.i.3i'3”:‘,:":‘t".35:$‘:$: resolves. Q's-s. terrific-'12:; M stitches. teeters-.2 discharge: see “sec. :‘:513.’:{3133'1€‘ NonnallSyrrInetricaI nodularltylSyrrsnetricaI fibrocystic (Fill Out Quality of CBE Documentation) I.) 68 Quality of Written Description of CBE Documentation (Check All That Apply): Nipple Change Scar 35?? Inspection, specify' : Paipation, specify: Fibrocystic Breast Mass(es) No specific docunentation besides normal Other, Specify: #Name? Abnormal: Which breastIs) has abnormal finding? its: Lymph node examination Adenopathy/Axiliary Nodes Breast Size/Shape Skin Change Nodularity Pain/tenderness If you don't know which breast, please record information in ”Left Breast” category. Left Breast: Location: #Name? Lump(s)lMass(es)/Asymmetric breast thickening/ Asymmetric Fibrocystic Lump size: #Name? Depth: Hardness: Mobility: Shape: Texture: Additional Findings With Lumps (check all that apply): Skin DimplinglRetraction Skin Erythema Skin Peau d‘orange or Skin Thickening Nipple Retraction Nipple Scaling Pain/Tendemess Fibrocystic Breast(s) Nipple Discharge Other, Specify: #Name? Nipple Discharge With No Lump Spontaneous? Color Unilateral or bilateral? Single or multiple ducts? Observational Findings With No Lump Right Breast: Location: #Name? Lump(s)/mass(es)/Asymmetric breast thickening! Asymmetric Fibrocystic Lump size: #Name? Depth: Hardness: Mobility: Shape: Texture: Additional Findings With Lumps (check all that apply): Skin Dimpiing/Retraction Skin Erythema Skin Peau d'orange or Skin Thickening Nipple Refraction Nipple Scaling Pain/Tendemess Fibrocystic Breast(s) Nipple Discharge Other, Specify: #Name? Nipple Discharge With No Lump Spontaneous? Color Unilateral or bilateral? Single or multiple ducts? Observational Findings With No Lump .‘~\.'. ..’I....€»v.~.;.-.~ .»... .- arm’. x1:“.::.-:::i'5,-:::i<:t.‘.:t::i . . .~ - a-' o ,._ ,...t . C:-§\:-:‘- {flavtftttrs‘ni‘ «;‘.‘.. '7‘. , 4' .. .V ., ~rf~: z,” ‘:'-‘..,:..:.. ._,.,.. ,_-.}.\-,3‘. 3' {39.1.1 5' "tx11-'{1;'\’~'x}f(t- '~ 25:42. . "3!- lv.> 3 s...)'... ‘35,; I “qa| -. } :‘-::..~?.n? -‘.:::-:.o. amt.- . . E Pain I a bitch: far-:83: Lhtsmsss’szni Other, specify: #Name? 1“,, ' . ..'.... u}\u- I \~lv ’ a - :‘-:::...~'? ~i::: -:l-. tit-z:- . . ‘.-.....‘v . . , . .2,... ‘ " ' 0‘ ‘ ' '. 1' :'{:;.}J'\- m watt: . . Pain > pass: Other, specify: #Name? Quality of Written Description of CBE Documentation For Abnormal Findings (Check All That Drawing of abnormal findings Inspection, specify: Sear WY): Nipple Change Breast Size/Shape Skin Change Palpation, specify Fibrocystic Breast Noduiarity Mass(es) Painltendemess Lymph node examination Adenopathy/Axillary Nodes Lymph Node Enlarged? 0mg, Smcily: #Name? 7O Form III-Test Result Entry Study ID: #Name? Date of the Visit: Mame? 12. Mammogram Documentation: 1. Ordered/RecommendedlEncouraged 2. Mammogram Performed 3. Results Obtained Stamped/Documented? 4. Results Reviewed By FPCP Signed/Documented? Date: #Nama? Date: #Name? D‘t‘: #Name? Date: #Name? 13a. Mammogram Findings: Final Impressions Which Breast? If you don't know which breast, please record Information In "Left Breast" category. Left Breast: Normal/No Finding ldentifledICategory I [E NormaIIBenignnappearIng abnonnalltleategory II Probably benign/possibly malignant, inderterminate [Category III Suspicious for malignancy/Category IV Malignant until proven otherwise/Category V Right era NormalINo Finding IdentifiedICategory I E] NormallBenign-appearlng abnonnaIItyICateg Probably benign/possibly malignant, inderterminate lCategory Ill Suspicious for malignancy/Category Malignant until proven otherwise/Categor .Other: Specify: #Nama? Other". Sp #Name? 13b. Mammogram Findings: Description Which Breast? If you don't know which breast, please record Inforrnatlon In "Left Breast" category. LeftBreast: Right Breast: Asymmetric Breast: more in which breas Bilateral Implants Radiolucent Breasts E] Dense Breasts/Dense Nodular Breasts Rounded density(ies), most likely cyst or fibroadenom Irregular Density(ies) Benign Appearing Calcifications Suspicious Calcification Calcified Fibroadenoma Axillary Lymph Node Other, specify: #Name? 13c. Mammogram Findings: Location For Category II and Up Bilateral Implants [E] Radiolucent Breasts Dense Breasts/Dense Nodular Breasts Rounded densities, most likely cyst or fibroaden Irregular Density(ies {-2 Benign Appearing Calcifications [E] Suspicious Calcification Calcified Fibroadenoma Axillary Lymph Node Other, specify: #Name? Which Breast? If you don't know which breast, please record Information In "Left Breast" category. IF AREA NOT SPECIFIED, check SCATTERITHROUGHOUT Breast category Left Breast Location: 71 Right Breast Location: Upper Outer Quadrant Upper Inner Quadrant Lateral Breast Medial Breast Areolar/Nipple Area Deep Against Chest Wall Scattered/Throughout Breast Other, specifyfiName? Lower Outer Quadrant Lower Inner Quadrant Upper Outer Quadrant Upper Inner Quadrant Lateral Breast E] Medial Breast AreolarlNipple Area Deep Against Chest Wall Scattered/Throughout Breast Other, specify: #Name? Lower Outer Quadra Lower Inner Quadra 14. Patient Notified of the Mammogram Findings? Date of Notification: #Name? 15.Cyst-Fine Needle Aspiration (FNA) Done by: Date done: #Name? Mass resolved/fluid not bloody Fluid blood [El Residual Mass Other, specify: #Name? Sent Fluid to Cytology Results Obtained Stamped/Documented? Date: #Name? Results Reviewed By FPCP Signed/Documented? Date: #Name? Cytology Results: lnsuflicientlHypocalluIar/Apocrine Cells Malignant [E] Atypical cells Suspicious for malignancy Benign/Fibrocystic/Apocrine Cells Other, specify: #Name? 16. Patient Notified of the FNA Findings From Cytology? Date of Notification: #Name? 17. Solid Mass-Fine Needle Aspiration Biopsy (FNAB) Done by: Date done: #Name? Specimen Submitted For Analysis Results Obtained Stamped/Documented? Date: #Name? Results Reviewed By FPCP Signed/Documented? Date: #Name? Pathology Results: lnsufiicientlHypocellular Benign/Fibrocystic [Z] Atypical cells Suspicious for malignancy Malignant Other, specify: #Name? 18. Patient Notified of the FNAB Findings From Path Report? Date of Notification: #Name? 19. Ultrasound Findings: Ordered by: Date done: #Name? Results Obtained Stamped/Documented? Date: #Name? 72 Results Reviewed By FPCP Signed/Documented? Date: #Name? Negative finding [E] Simple cyst(s) Solid mass(es) or complex cyst(s) Other, specify: #Name? 20. Patient Notified of the Ultrasound Findings? Date of Notification: #Name? 21. Image-Guided Biopsy/Open Biopsy Results: Date done: #Name? Results Received Stamped/Documented? Date: #Name? Results Reviewed By FPCP Signed/Documented? Date: #Name? Open Biopsy Flndings(check all that apply): [E] Benign/No Evidence of Malignancy Ductal Carcinoma in situ Benign/Fibrocystic Changes Lobular Carcinoma in situ Benign/Fat Necrosis Atypical Hyperplasia Benign/Lipoma Invasive Ductal Carcinoma Benign/Fibroadenoma Tsii Invasive Lobular Carcinoma Other, specify: #Name? Go Back to Visit Form Go To Followup Form 73 Form lV-Follow-up Entry StudyID: #Name? Date of Visit: #Name? 23. Recommended Follow-Up(s) (Check All That Apply) Undocumented Follow-up for Normal CBE and Mammogram (or One of Them Undocumented): Routine Screening Following ACS Guidelines 12 Month cee Recommended by: Follow-up for Specific Abnormalities: Breast Mass/Asymetry Initial Approach: CBE at better phase cycle (3-10 days) Fine Needle Aspiration for Cyst If Known Breast Cyst: Send Fluid to Cytology Reaspiration #Name? (How many) month CBE If Known Solid Mass: 333353 Fine Needle Aspiration Biopsy Specimen Submitted for Analysis Repeat aspiration Clinical Followup Every 3 Months for1 Year For Nipple Discharge: Endocrine work-up For Skin/Nipple Changes on Observation: 2 weeks antibiotics Skin Biopsy 2 weeks topical hydrocortisone For Breast pain: Eliminate Caffeine 2:: Adjust Estrogen Dose Local Anesthetic Injection Primrose OlII, How Many Months? #Name? Reassurance and CBE within 3—6 months if pain persists 22:5: Sipportive Brassiere Over-the-counter Analgesics Danazol, Bromocriptine 74 12 Month Mammogram Following Other Guidelines specify: #Name? Comments: #Name? Follow-up Common To Any Abnorrnali Call if Problem Worsens Routine Screening Recom. by: Immediate Mammogram Workup: Regular Mammogram Extra Mammogram Views Cone or Spot Compression Magnification Views Recom. by. Interval Followup: an (How many) month mammo an (How many) month CBE Recom Ultrasound Recom. by: Surgical Referral Recom. by: Undocumented Other Reconmendations Or Comments Concerning Abnormality(les): #Name? For Occult Marunographic Abnomaiity: General Comments About This Visit: Radiologic Biopsy/lmage-Guided Biopsy Recommended by: #Name? Assessment/Recommended Follow-up From Surgeon's Letter 1 . Letter Written Date: #Name? 2. Letter Received Stamped/Documented? Date: #Name? 3. Letter Revimd by F PCP Signed/Documented? Date: #Name? Assessment Followup Referral Diagnosis Not Confirmed Referral Diagnosis Confirmed No Further Workup Required Additional/New findings Further Tests Recommended/Done By Surgeon, check all that apply Immediate Mammogram Interval Mammogram, how long? #Name? Followup I" Primary Care Office Interval CBE, how long? #Name? Ultrasound FNA FNAB RadiologicaVlmage Guided Biopsy Open Biopsy Evidence of Malignancy/I Previous Abnormality Resolved Current Abnomaiity Resolved Other Comments From Surgeon's Letter Followup In Surgeon's Office #Name? """" 5...... mm mm“ mm 75 Appendix B Data Dictionary used for DOD Datasets 76 Value Labels Variable Nalid Name Label Type Values Note Front-End Lname Patient Last Name Char Fname Patient First Name Char MRNum Medical Record Number Num DOB Patient DOB Date DOB stands for Date of Birth Able Abstractor’s ID Num Gender 1. Patient Gender is Char 1=Female 2: Male 9 = Undoc Active 2. Patient has been seen in Char 1=Yes O = No: last 3 years 9 = Undoc FirstVis 2a.Date of the very first Date Very first visit to the FPC visit to the FPC provider provider Age70 3 Patient birthday is Char 1=Yes 0 = No Patient birthday must be between August 1, 1928 9 = Undoc between 8/01/1928 and and July 31g 960 7/01/1959 Care 4. Breast health care Char 1=FPC Breast Care Health Provider provided by Provider 6 = Other 9 = Undoc Active1 5. Active patient between Char 1=Yes 0 = No Set dates are 08I01/98 - 8/1/99 - 7/31/00 9 = Undoc 07/31/99 E5a 5a. If there is Char 1=N/A if inactive patient documentation patient left 2=Death practice before 7/31/00 3=Transferre d ~4=Move out of town 8=Other, specify E5aspe Other, specify Char specify E5adate Date of Documentation: Date date of inactivity ECode Eligibility Code Char 1=Eligible 2 = Guide;ine Insertion 3 = Ineligible ECodeOld The old ECoded assigned Char 1=Eligible 2 = last year Guide;ine Insertion 3 = Inelilifle StudleOId Study ID Num Old Study ID Dateold Date the form was filled out Num Date the form was filled out old 77 Value Labels] Variable Nalid Name Label Type Values Note Front-End Date Today’s Date: Date the Num Date the form was filled out form was filled out year 2000 NewGuideI Guideline Inserted Char Yes/No NewGuide2 Guideline Not Found Char Yes/No NewGuide3 Summary Sheet Inserted Char Yes/No NewGuide4 Summary Sheet Not Found Char Yes/No NewGuide5 Additional Information on Char Yes/No Summary Sheet NewGuide6 No Additional Information Char Yes/No on Summary Sheet Stamp Are documents stamped? Char 1=Guideline Stamped 2=Summary Sheet Stamped 3=Both Stamped 4=Not Applicable StudyID Study ID Num StudyID year 2000 DMRVis 1. Date of most recent visit Char Date of most recent visit (MM/DD/YY) dclevis 2. Autocalculated Date For Char Calculated Last Eligibility Visit the Last Eligibile Visit Within the Last 15 months (MM/DD/YY): DateAdd 2a Overlap Period Num DCLEVisoId 2b. Last Years Num Calculated Last Eligibility Visit Autocalculated Date For the Last Eligibile Visit Within the Last 15 months: TNum 3. Total Number of \fisits Num Within 15 Months, Including The Most Recent Visit ActY 4. Was A Breast Care Char 0=No 1 = Yes Performed During Any of The Visits Within The 15 Months Period History 5. Personal/Family History Char 0=None, Of Breast Cancer? 1=Yes 8 = Not Apple 9 = Undoc 78 Value Labels Variable Nalid Name Label Type Values Note Front-End YSelf In Self? Char 0=None, History of breast cancer in self 1=Yes 8 = Not Applc 9 = Undoc Ages Age Num Age of self SeIfFuI Complete Breast Removal Char Yes/No In self Complete Breast Removal SelfPar Partial Breast Char Yes/No In self Partial Breast Removal/Lumpectomy Removal/Lumpectomy SelfProl Prophylactic Implants Char Yes/No In self Prophylactic implants SeIfAut Autologous Reconstitution Char Yes/No In self Autologous Reconstitution Selstth Other Char Yes/No In self Other Selstts Other, speciL Char In self other specify SelfU Undocumented Char Yes/No In self Undocumented SeIfT Che Chemotherapy Char Yes/No In self Chemotherapy SeIfT Rad Radiation Char Yes/No In self Radiation SeIfTT am Tamoxifen/Nolvadex Char Yes/No In self Tamoxifen Nolvadex SeIfT Alt Alternative medicinejs), Char Yes/No In self Alternative medicine SeIfT AIS Alternative medicine(s), Char In self specify specify SeIfTOth Other Char Yes/No In self Other SeIfTOts Other, specify Char In self Specify SelfTUn Undocumented Char Yes/No In self Undocumented YMother In Mother? Char 0=None, 1=Yes 8 = Not Apple 9 = Undoc _AgeM Ag; Num Amf the Mother Ysister In Sister Char 0=None, 1=Yes 8 = Not Apple 9 = Undoc YSister1 Sister1 Char Yes/No AgeS1 A e Num age of sister1 YSister2 Sister2 Char Yes/No AgeSZ A e Num age of sister2 YDaugh In Daughter Char 0=None, 1=Yes 8 = Not Apple 9 = Undoc ‘ YDaughI D_agghter1 Char Yes/No Agem Ag: Num age of dagghtefl YDauth Dagghel’z Char Yes/No AgeDZ Age Num age of daughterz 79 . Value Labels Variable Nalid Name Label Type Values Note Front-End YOther In Other Relatives Char 0=None, 1=Yes 8 = Not Apple 9 = Undoc SYOther specify: Char BOX-A Record information for Char patients each visit when a breast care was performed. 8O Variable Name Label Typo Value Labels Narid Values Note Form II Visit Entry Form DM RVis Last Eligible Visit Num Date of most recent visit (MM/DD/YY), carried over DVisit 6.Date of Breast Care Activity Was Recorded Num Date of visit texttel Type of Contact Char 1=Office Visit 2=Office Initiated Phone Consultation 14=Office Response Phone Consultation 3=Patient Initiated Phone Consultation 4=Screening/Routine/Re gular Mammogram 5=Diagnostic(ReguIar) Mammogram 6=Diagnostic/Cone Compression/Magnificati on Mammogram 7=Ultrasound Result 8=FNA for Cyst Result 9=FNAB Result 10=Pathology Report for Radiological/Image Guided Biopsy 11=Pathology Report for Open Biopsy 12=Surgeon's Letter 13=Other purpose 7.Purpose of this Visit/Call Char 1 =ScreeningNVeII Women Exam/Annual Exam 2=Presenting symptom(s) 3=Follow-up of a previous abnormality 4=Prompted by results of screening mammogram 5=Prompted by results of other test(s) 6=Routine care/Other health problems 8=Other text Specify Char Describe YRPN 8.Resident Physician Char Yes I No Who Performed Breast Care/Phone Consultation? 81 Variable Value Labels New Name Label Type Values Note Form II Visit Entry Form YFPN 8.Faculty Physician Char Yes I No ~ do ~ YPAN 8.Physician Assistant Char Yes / No ~ do ~ YNPN 8.Nurse Practitioner Char Yes I No ~ do ~ UndocN 8.Undocumented Char Yes / No undocumented ProviderI 8.Breast Care Char Yes I No ~ do ~ ProviderI Provider2 8.Breast Care Char Yes I No ~ do ~ Provider2 WbrePre 9.Which breast(s) Char 1=Left has presenting 2=Right symptom 3=Both 9=Don't Know PreNone None Char Yes I No Presenting symptoms, left breast Smendo Undocumented/Don't Char Yes I No ~ do ~ know SymLump Lump/masses Char Yes I No ~ do ~ SymDis Nipple Discharge Char Yes / No ~ do ~ SpoonDis Nipple Discharge Char 1=Spontaneous ~ do ~ specify 0=Non-Spontaneous 9=Undocumented SymCha Skin/Nipple change Char Yes I No Presenting symptoms, left breast SkinDim Skin Dimling Char Yes I No ~ do ~ Eryth Erythema/Skin Char Yes I No ~ do ~ Thickening NipRet Nipple Retraction Char Yes / No ~ do ~ NipSca Nipple Scaling Char Yes I No ~ do ~ SymPain Painfl'endemess Char Yes I No ~ do ~ SymWPain Painfl'endemess Char 1=Premenstrua|lMenstru ~ do ~ specify al 2 = Persistent 8 = Not Specified SymOcc Occult Char Yes I No Presenting symptoms, Mammographic left breast Abnormality symOcc1 Density(Nodule or Char Yes I No ~ do ~ Asymmetry) symOcc2 Microcalcifications Char Yes I No ~ do ~ SymOth Other Char Yes I No ~ do ~ Other Other, specify Char ~ do ~ PreNoneR None Char Yes I No ~ do ~ SmendoR Undocumented/Don’t Char Yes I No ~ do ~ know SymLumpR Lump(s)/Mass(es)IA Char Yes / No ~ do ~ symmetrical thickening SymDisR Nipple Discharge Char Yes I No ~ do ~ 82 Variable Value Labels NaIiTI Name Label Type Values Note Form Il Visit Entry Form SponDisR Nipple Discharge Char 1=Spontaneous 2 = Non~ ~ do ~ specify Spontaneous 9 = Undoc SymChaR Skin/Nipple change Char Yes I No Presenting symptoms, right breast SkinDimR Skin Dimpling Char Yes I No ~ do ~ ErythR Erythema/Skin Char Yes I No ~ do ~ thickening NipRetR Nipple Retraction Char Yes I No ~ do ~ NipScaR Nipple Scaling Char Yes I No ~ do ~ SymPainR Painfl'enderness Char Yes I No ~ do ~ SymWR Pain/Tendemess Char 1=Premenstrua|lMenstru ~ do ~ al 2 = Persistent 8 = Not Sgcified SymOccR Occult Char Yes I No ~ do ~ Mammographic Abnormality symOcc1 R Density(NoduIe or Char Yes I No ~ do ~ Asymmetry) symOcc2R Microcalcifications Char Yes / No ~ do ~ SymOthR Other Char Yes I No ~ do ~ OtherR Other, specify Char ~ do ~ CbeDoc 10. CBE Char 1=Documented 0 = Not Documentation Done Undoc BP 11. Bilateral Implants Char Yes I No 11. CBE Findings (Check All That Apply): Mas 11. Mastectomy, Char Yes / No ~ do ~ which breast? MasWhich 11. Mastectomy, Char 1=Left ~ do ~ which breast? 2=Right 3=Both 9=Don't Know AbnRes 11. Previous Char Yes I No ~ do ~ abnormality resolved LGone Lump/mass resolved Char Yes I No ~ do ~ OFGone Observational finding Char Yes I No ~ do ~ resolved NDGone Nipple discharge Char Yes / No ~ do ~ resolved PGone Pain fine Char Yes / No ~ do ~ CBEFN 11.NormaIISymmetri Char Yes I No Quality of Written cal Description of CBE nodularity/Symmetric Documentation al fibrocystic Insn Inspection, specify Char Yes I No 83 Variable Value Labels Nal'id Name Label Type Values Note Form II Visit Entry Form CBEInsNC Nipple Change Char 1=Yes O = No 9 = Undoc CBEInsSc Scar Char 1=Yes 0 = No 9 = Undoc CBEInsBs Breast Size/Shape Char 1=Mentioned O = Not mentioned 9 = Undoc CBEInsS Skin Change Char 1=Yes 0 = No 9 = Undoc Paln Palpation, specify Char Yes I No CBEPaIFC Fibrocystic Breast Char 1=Yes 0 = No 9 = Undoc CBEInsMa Mass(es) Char 1=Yes 0 = No 9 = Undoc CBEPaIND Nodularity Char 1=Yes 0 = No 9 = Undoc CBEPaIPa Pain/tendemess Char 1=Yes 0 = No 9 = Undoc Pain gone CBEPALMAS Masectomy site(s) Char 1=Yes 0 = No 9 = Undoc free of masses NodeN Lymph node Char Yes I No examination AdenoN Adenopathy/Axillary Char 1=Yes 0 = No 9 = Undoc Nodes UndocNR Undocumented NoDocl No specific Char Yes I No documentation besides normal CBEFinOl Other Char Yes I No OtherA1 Other, Specify Char Wbre Which breast(s) has Char 1=Left Abnormal abnormal finding? 2=Right 3=Both 9=Don't Know LClock Location Char Left Breast 0-12 for clock position LCBEFLu Lump(s)IMass(es)/A Char Yes I No CBE finding: symmetric breast lump/masses thickening/ Asymmetric Fibrocystic LLumSize Lump size Char Left breast lump size LLumDept Depth Char 1=Superficial, 2=Medium, 3=Deep LLumHard Hardness: Char 1=Hard, 2=Firm, 3=Soft LLumMobi Mobility Char 1=Mobile, 2=Fixed LLumShap Shape Char 1=Round, 2=Oblong 3 = Irregular 84 Variable Value Labels Nal'id Name Label Type Values Note Form II Visit Entry Form LLumText Texture: Char 1=Regular, 2=Irregular, 3=Smooth LLumSkiD Skin Char 1=Yes, 0=No 9 = Undoc ~ do ~ DimpfiLgIRetmction LLumEry Skin Erythema Char 1=Yes, 0=No 9 = Undoc ~ do ~ LIumPeau Skin Peau d'orange Char 1=Yes, 0=No 9 = Undoc ~ do ~ or Skin Thickening LLumNip Nipple Retraction Char 1=Yes, 0=No 9 = Undoc ~ do ~ LLumSca Nipple Sealing Char 1=Yes, 0=No 9 = Undoc ~ do ~ LLumpPa Pain/Tenderness Char 1=Yes, 0=No 9 = Undoc ~ do ~ LLumpFi Fibrocystic Breast(s) Char 1=Yes, 0=No 9 = Undoc ~ do ~ LLumpND Nipple Discharge Char 1=Yes, 0=No 9 = Undoc ~ do ~ Ilumoth Other Char Additional Findings With Lumps- left Ilumoths Other, Specify Char ~ do ~ LCBEF Dis Nipple Discharge Char Yes / No CBE finding: nipple With No Lump discharge - no lump LSpon Spontaneous? Char 1=Yes, 0=No, CBE finding: nipple 9=Undocumented discharge LCoIor Color Char 1=Milky, ~ do ~ 2=Green/BrownIYeIIow 3=Watery/SerouslBlood Y LLate Unilateral or Char 2=Bilateral, 1=Unilatera| ~ do ~ bilateral? 9 = Undoc LDuct Single or multiple Char 1=Single duct, ~ do ~ ducts? 2=Multiple ducts 9 = Undoc LCBEFObs Observational Char Yes I No CBE finding: Findings With No observational finding Lump LSkiDim Skin Char Yes I No ~ do ~ dimpling/retraction LEry Skin Erythema Char Yes I No ~ do ~ LPeau Skin Peau Char Yes I No ~ do ~ d'orange/Skin ThickenirLL LNipRet Nipple retraction Char Yes I No ~ do ~ LNipSca Nipple scalim Char Yes I No ~ do ~ LCBEFP Pain Char Yes I No CBE finding: pain LBreast Breast pain Char Yes I No Pain: breast pain LCycIic Breast pain Char 1=Cyclic, 0=Noncyclic 9 Pain: breast pain = Undoc 85 Variable Value Labels Naiid Name Label Type Values Note Form II Visit Entry Form LChest Chest wall pain Char Yes I No Pain: chest pain LUnspe Unspecified Char Yes I No Pain: unspecified LCBEFinO Other Char Yes I No CBE finding other LOtherA Other, specify: Char CBE finding other, ' specify RCIock Location ' Char Location Right Breast Or 12 for clock position RCBEFLU Lump(s)/mass(es)/A Char Yes I No CBE finding: symmetric breast lump/masses right thickening! breast Asymmetric Fibrocystic RLumSize Lump size Char Lump size L'ght breast RLumDept Depth Char 1=Superficial, Depth lump right breast 2=Medium, 3=Deep RLumHard Hardness Char 1=Hard, 2=Firm, 3=Soft Hardness lump right breast RLumMobi Mobility Char 1=Mobile, 2=Fixed Mobility lump right breast RLumShap Shape Char 1=Round, 2=Oblong, Shape lump right breast 3=lrregular RLumText Texture: Char 1=Regular, 2=Irregular, Texture right breast 3=Smooth RLumSkiD Skin Char 1=Yes, 0=No, Additional Findings Dimpling/Retraction 9=Undocumented With Lumps- right breast RLumEry Skin Erythema Char 1=Yes, 0=No, ~ do ~ 9=Undocumented RLumPeau Skin Peau d’orange Char 1=Yes, 0=No, ~ do ~ or Skin Thickenipg 9=Undocumented RLumNip Nipple Retraction Char 1=Yes, 0=No, ~ do ~ 9=Undocumented RLumSca Nipple Scaling Char 1=Yes, 0=No, ~ do ~ 9=Undocumented RLumPa Pain/Tendemess Char 1=Yes, 0=No, ~ do ~ 9=Undocumented RLumFi Fibrocystic Breast(s) Char 1=Yes, 0=No, ~ do ~ 9=Undocumented RLumpPa Nipple Discharge Char 1=Yes, 0=No, ~ do ~ 9=Undocumented rIimoth Other Char Yes I No ~ do ~ rIimoths Other, Specify: Char ~ do ~ RCBEFDIS Nipple Discharge Char CBE finding: nipple With No Lump discharge right breast RLSpon Spontaneous? Char 1=Yes, 0=No, CBE finding: nipple 9=Undocumented 86 discharge right breast Variable Value Labels Nalid Name Label Type Values Note Form II Visit Entq Form RColor Color Char 1=Milky, ~ do ~ 2=Green/Brown/Yellow 3=Watery/SerouslBlood ' Y RLate Unilateral or Char 2=Bilateral, 1=Unilateral ~ do ~ bilateral? 9 = Undoc RLDuct Single or multiple Char 1=Single duct, ~ do ~ ducts? 2=Multiple ducts 9 = Undoc RCBEFObs Observational Char Yes I No CBE finding: Findings With No observational finding, Lump right breast RSkiDim Skin Char Yes I No dimpling/retraction REry Skin Erythema Char Yes I No RLPeau Skin Peau Char Yes I No d'orange/Skin Thickening RLNipRet Nipple retraction Char Yes / No RNipSca Nipple scaling Char Yes / No RCBEFP Pain Char Yes I No CBE finding: pain RBreast Breast pain Char Yes I No Pain: breast pain RCycIic Breast pain specify Char 1=Cyclic, 0=Noncyclic 9 Right breast pain = Undoc RChest Chest wall pain Char Yes I No Pain: chest pain RUnspe Unspecified Char Yesl No Pain: unspecified RCBEFinO Other Char Yes I No CBE finding other ROtherA Other, specify: Char CBEDDra Drawing of abnormal Char Yes / No Quality of Written findings Description of CBE Documentation For Abnormal Findings Ins Inspection, specify: Char Yes I No Written description CBEInsAC Nipple Change Char 1=Yes, 0=No, 9=Undocumented CBEInsas Scar Char 1=Yes, 0=No, 9=Undocumented CBEInsAB Breast Size/Shape Char 1=Mentioned 0 = Not mentioned CBEInsag Skin Change Char 1=Yes, 0=No, 9=Undocumented Pal Palpation, specify: Char Yes I No Palpation CBEPaIaf Fibrocystic Breast Char 1=Yes, 0=No, 9=Undocumented CBEInsAM Mass(es) Char 1=Yes, 0=No, 9=Undocumented CBEPaIan Nodularity Char 1=Yes, 0=No, 9=Undocumented 87 Variable Value Labels Nalid Name Label Type Values Note Form II Visit Entry Form CBEPaIPI Painltendemess Char 1=Yes, 0=No, 9=Undocumented Node Lymph node Char Yes I No examination Adeno Adenopathy/Axillary Char 1=Yes, 0=No, Nodes 9=Undocumented LNLar Lymph Node Char 1=Yes 0 = No Enlarged? cbeoth Other Char Yes I No other quality control cbeoths Other, Specify: Char Change Click here if you Char Yes I No Is this visit changed? changed anything about this visit entry, compared to last year's entry and briefly specify the changes COMMENT COMMENT Char 88 Variable Value Labels Nai‘id Name Label Type Values Note Form III -Test Result Entry StudyID Study ID Num Carried over study ID Date Date of the Wsit Date Carried over date of visit DMRWs Last Eligible Visit Date MamDoc 12. Mammogram Char 1=Documented Documentation 2=Previously Documented O=Not Done/Undocumented DiaOrd 1. Char 1=Yes Ordered/Recommended/Enco 0=No uraged 9=Don't Know/Undocumente d DDiaOrd Date Date DiaMam 2. Mammogram Performed Char 1=Yes 0=No 9=Don't Know/Undocumente d DDiaMam Date Date DiaOtht 3. Results Obtained Char 1=Yes Stamped/Documented 2=Yes, but can not read date 0=No DDiaObt Date Date DiaRevSt 4. Results Reviewed By FPCP Char 1=Yes Signed/Documented? 2=Yes, but can not read date 0=No DDiaRev Date Date bcsideC 13a. Mammogram Findings: Char 1=Left Final Impressions Which 2=Right Breast? 3=Both 9=Don't Know MamFinNL Left Breast: Normal/No Finding Char Yes I No Identified/Catqu I MamFinBL Left Breast: Normal/Benign- Char Yes I No appeanng abnormality/Category II MamFinPL Left Breast: Probably Char Yes I No benign/possibly malignant, inderterminate ICatggory III MamFinSL Left Breast: Suspicious for Char Yes I No majgnancy/Category IV MamFinML Left Breast: Malignant until Char Yes I No proven otherwise/Cateng 89 Variable Value Labels Nai'id Name Label Type Values Note Form III -Test Result Entry MamFinoL Left Breast: Other: Char Yes I No MamFinLS Left Breast: Other: Specify: Char MamFinNR Right Breast Normal/No Char Yes / No Finding Identified/Category I MamFinBR Right Breast NormaVBenign- Char Yes / No appeanng abnormality/Category" MamFinPR Right Breast Probably Char Yes I No benign/possibly malignant, inderterminate ICategory III MamFinSR Right Breast Suspicious Char Yes I No MamFinMR Right Breast Malignant until Char Yes / No proven otherwise/Categm V MamFinoR Right Breast, Other Char Yes I No MamFinRS Right BreastOther. Specify: Char Right Breast bcsideb 13b. Mammogram Findings: Char 1=Left, 2=Right, Description Which Breast? 3=Both 9 = OK MamDesA Asymmetric Breast: in Breast Char Yes I No Location MamDesAw Asymmetric Breast: more in Char 1=Right, 2=Left which breast: MamDesBL Left Breast Bilateral Implants Char Yesl No MamDesRL Left Breast Radiolucent Char Yes I No Breasts MamDesDL Left Breast Dense Char Yes I No Breasts/Dense Nodular Breasts MamDesL Left Breast Rounded Char Yes I No density(ies), most likely cyst or fibroadenoma MamDesIL Left Breast Irregular Char Yes I No Density(ies) MamDescL Left Breast Benign Appearing Char Yes I No Calcifications MamDesSL Left Breast Suspicious Char Yes I No Calcification MAMDECFL Left Breast Calcified Char Yes I No Fibroadenomas MamDesAR Left Breast Axillary Lymph Char Yes / No Nodes MamDesOL Left Breast Other Char Yes / No MamDesLS Left Breast Other, specify: Char MamDesBR Right Breast Bilateral Implants Char Yes I No MamDesBR Right Breast Radiolucent Char Yes I No Breasts 9O Variable Value Labels Nal‘id ’ Name Label Type Values Note Form III -Test Result Entry MamDesDR Right Breast Dense Char Yes I No Breasts/Dense Nodular Breasts MamDesr Right Breast Rounded Char Yes I No densities, most likely cyst or fibroadenoma MamDeis Right Breast Irregular Char Yes I No Density(ies) MamDesCR Right Breast Benign Appearing Char Yesl No Calcifications MamDesR1 Right Breast Suspicious Char Yes I No Calcification MamDecFR Right Breast Calcified Char Yes I No Fibroadenomas MamDesAl Right Breast Axillary Lymph Char Yes I No Nodes MamDesOR Right Breast Other Left Breast Locatio it: [No MamDesRS Right Breast Other, specify: Char bcsideA 13c. Mammogram Findings: Char 1=Left, 2=Right, Location For Category II and 3=Both 9 = OK Up Which Breast? MamLUppO Upper Outer Quadrant Char Yes I No Left Breast Location MamLLowO Lower Outer Quadrant Char Yes I No ~ do ~ MamLUppI Upper Inner Quadrant Char Yes I No ~ do ~ MamLLowI Lower Inner Quadrant Char Yes / No ~ do ~ Lmame Lateral Breast Char Yes I No ~ do ~ LMame Medial Breast Char Yes / No ~ do ~ MamLRetr Areolar/Nipple Area Char Yes I No ~ do ~ MamLDeep Deep Against Chest Wall Char Yes I No ~ do ~ MamLSca Scattered/'I’hrorghout Breast Char Yes / No ~ do ~ MamLPosO Other Char Yes I No ~ do ~ MamLPosS Other, specify: Char ~ do ~ MamRUppO Upper Outer Quadrant Char Yes I No Right Breast Location MamRLowO Lower Outer Quadrant Char Yes I No ~ do ~ MamRUppl Upper Inner Quadrant Char Yes I No ~ do ~ MamRLowI Lower Inner Quadrant Char Yes I No ~ do ~ Rmamlb Lateral Breast Char Yes I No ~ do ~ RMame Medial Breast Char Yes I No ~ do ~ MamRRetr Areolar/Nipple Area Char Yes I No ~ do ~ MamRDeep Deep Against Chest Wall Char Yes I No ~ do ~ MamRSca Scattered/“whom Breast Char Yes I No ~ do ~ MamRPosO Other Char Yes / No ~ do ~ MamRPosS Other, specify: Char ~ do ~ 91 Variable Value Labels Nalid Name Label Type Values Note Form III -Test Result Entry PM 14. Patient Notified of the Char 1=Yes, 0=No 9 = Mammogram Findings? Undoc I DK DPM Date of Notification: Char Date patient was notified WHoF NA 15.Cyst-Fine Needle AspirationiChar 1=FPCP, (FNA) Done by: 2=Surgeon, 8=Other 9 = Undoc I DK FNAD Date done: Date Date FNA was performed FnaFinM Mass resolved/fluid not bloody Char Yes / No Fine Needle Aspiration mass resolved/fluid not bloody FnaFinF Fluid bloody Char Yes / No Fine Needle Aspiration fluid Ifioody FnaFinR Residual Mass Char Yes I No Fine Needle Aspiration residual mass FnaFinO Other Char Yes I No Fine Needle Aspiration other Otherz Other, specify: Char F NACyto Sent Fluid to Cytology Char Yes I No DiaOthtf Results Obtained Char 1=Yes 2 = Yes, but Stamped/Documented? can not read 0 = No DDiaObtf Date: Date DiaRevStf Results Reviewed By FPCP Char 1=Yes 2 = Yes, but can not read 0 = No DDiaRevf Date: Date Insuf1 Cytology Results: Char Yesl No Insufficient Insufficient/Hypocellular/Apocri ne Cells Benign1 Cytology Results: Char Yes I No Benign Benign/Fibrocystic/Apocrine Cells Acell1 Atypical cells Char Yes I No Susl Suspicious for malignancy Char Yes I No Malig1 Malignant Char Yes I No FnaRFinO Other Char Yes I No ~ do ~ ROtherB Other, specify: Char PF 16. Patient Notified of the FNA Char 1=Yes, 0=No 9 = Findings From Cytology? Undoc I DK DPF Date of Notification: Date Date patient was notified 92 Variable Value Labels ATai'id Name Label Type Values Note Form Ill -Test Result Entgr WHoFNAB 17. Solid Mass-Fine Needle Char 1=FPCP, Aspiration Biopsy (FNAB) 2=Surgeon, 8=Other Done by: 9 = Undoc/ DK FNABD Date done: Date Date FNA was performed F NABPath Specimen Submitted For Char Yes I No Analysis DiaOthtfb Results Obtained Char 1=Yes 2 = Yes, but Stamped/Documented? can not read 0 = No DDiaObtfb Date: Date DiaRevStfb Results Reviewed By FPCP Char 1=Yes 2 = Yes, but Signed/Documented? can not read 0 = No DDiaRevfb Date: Date lnsuf Pathology Results: Char Yes / No Insufficient Insufficient/Hypocellular Benign Pathology Results: Char Yes I No Benign/Fibrocystic Acell Pathology Results: Atypical Char Yes I No cells Sus Pathology Results: Suspicious Char Yes / No for malignancy Malig Pathology Results: Malignant Char Yes I No FnabFinO Other Char Yes I No ~ do ~ OtherB Other, specify: Char PF B 18. Patient Notified of the Char 1=Yes, 0=No 9 = FNAB Findings From Path Undoc/ DK Report? DPF B Date of Notification: Date Date patient was notified WhoOrdUl 19. Ultrasound Findings: Char 1=FPCP 2=Surgeon Date ultrasound Ordered by: 3=Radiologist was performed 8=Other 9=Undocumented/Do n't Know UItD Date done: Date DiaOthtuI Results Obtained Char 1=Yes 2 = Yes, but Stamped/Documented? can not read 0 = No DDiaObtuI Date: Date DiaRevStul Results Reviewed By F PCP Char 1=Yes 2 = Yes, but Signed/Documented? can not read 0 = No DDiaRevuI Date: Date UItFinNe Negative finding Char Yes I No ultrasound negative finding_ UltFinS Simple cyst(s) Char Yes I No ultrasound simple cyst 93 Variable Value Labels Nalid Name Label Type Values Note Form III -Test Result Entry UItFinC Solid mass(es) or complex Char Yes I No ultrasound solid cyst(s) or complex cyst UltFinO Other Char Yes / No ultrasound other Other3 Other, specify: Char PU 20. Patient Notified of the Char 1=Yes 0 = No 9 = Ultrasound Findings? DKI Undoc DPU Date of Notification: Date Date patient was notified OBRD 21. Image-Guided Date Biopsy/Open Biopsy Results: Date done: OBRRecS Results Received Char 1=Yes 2 = Yes, but Stamped/Documented? can not read 0 = No OBRRecD Date: Date OBRRevS Results Reviewed By FPCP Char 1=Yes 2 = Yes, but Signed/Documented? can not read 0 = No OBRRevD Date: Date OBRB Benign/No Evidence of Char Yes I No Open Biopsy Malignancy Findings OBRFC BthignIFibrocystic Changes Char Yes I No ~ do ~ OBRBenFi Benign/Fat Necrosis Char Yes I No ~ do ~ OBRBenLi Benign/Lipoma Char Yes / No ~ do ~ OBRBean Bengt/Fibroadenoma Char Yes / No ~ do ~ OBRDC Ductal Carcinoma in situ Char Yes I No ~ do ~ OBRIDS Lobular Carcinoma in situ Char Yes / No ~ do ~ OBRAH Atypical Hypemlasia Char Yes I No ~ do ~ OBRID Invasive Ductal Carcinoma Char Yes / No ~ do ~ OBRLC Invasive Lobular Carcinoma Char Yes / No ~ do ~ OBRO Other Char Yes I No ~ do ~ OBROS Other, specify: Char Change Click here if you changed Char Is this visit anything about this visit entry, changed? compared to last year's entry and briefly specify the changes COMMENT Char What kind of change? 94 Variable Value Labels Nafid Name Label Type Values Note Form IV - Follow-Up Form StudyID StudyID: Num Carried over study ID Date Date of Visit: Date Carried over date of visit DMRVis Last Eligible Visit: Date Undoc 23. Recommended Follow- char Yes I No Up(s) (Check All That Apply) Undocumented Rou Routine Screening char Yes I No Follow-up for Normal CBE and Mammogram (or One of Them Undocumented): CBE12 12 Month CBE char Yes I No ~ do ~ Mam12 12 Month Mammogram char Yes I No ~ do ~ ACS Following ACS Guidelines char Yes I No ~ do ~ OthGui Following Other Guidelines char Yes I No ~ do ~ OthGuiS specify: Char Yes I No WhoRou Recommended by: Char 1=Family Practice Doctor only(FPD) 2=Radiologist only 3=Both FPD and Radiologist 4=Surgeon 5=Nurse Practitioner 8=Other 9=Undocumented ncomment Comments: Char Yes I No CBEbet Breast Mass/Asymetry Initial Char Yes I No Approach:CBE at better phase Follow-up for Specific cycle (3-10 days) Abnormalities: FNAC Breast Mass/Asymetry Initial Char Yes I No Approacthine Needle . Aspiration for Cyst ~ do ~ cytology If Known Breast Cyst:Send Char Yes I No Fluid to Cytcfigy ~ do ~ Reasp If Known Breast Char Yes I No Cyst: Reaspiration ~ do ~ MlnterC If Known Breast Cyst: (How Char Yes I No manyymonth CBE ~ do ~ FNAB1 If Known Solid Mass: Fine Char Yes I No Needle Aspiration Biopsy ~ do ~ SendSpe If Known Solid Mass: Char Yes I No Specimen Submitted for Analysis ~ do ~ rasp3 If Known Solid Mass: Repeat Char Yes I No aspiration ~ do ~ 95 Variable Value Labels Nalid Name Label Type Values Note Form IV - Follow-Up Form CF If Known Solid Mass: Clinical Char Yes I No Followup Every 3 Months for 1 Year ~ do ~ Endo For Nipple Discharge: Char Yesl No Endocrine work-up ~ do ~ Antibio For Skin/Nipple Changes on Char Yes I No Observation: 2 weeks antibiotics ~ do ~ cortis For Skin/Nipple Changes on Char Yes I No Observation: 2 weeks topical hydrocortisone ~ do ~ SB For Skin/Nipple Changes on Char Yes I No Observation: Skin Biopsy ~ do ~ Caff For Breast pain: Eliminate Char Yes I No Caffeine ~ do ~ Estro For Breast pain: Adjust Char Yes I No Estrogen Dose ~ do ~ anes For Breast pain: Local Char Yes I No Anesthetic Injection ~ do ~ Oil For Breast pain: Primrose Oill, Char Yes I No as do as. Oilday For Breast pain: How Many Char Months Oil? ~ do ~ Reass For Breast pain: Reassurance Char Yes I No and CBE within 3-6 months if pain persists as do as. Brass For Breast pain: Supportive Char Yes I No Brassiere ~ do ~ Analg For Breast pain: Over-the- Char Yes I No counter Analgesics ~ do ~ dabrom For Breast pain: Danazol, Char Yes I No Bromocriptine ~ do ~ RB For Occult Mammographic Char Yes I No Abnomaiity: Radiologic Biopsy/Image-Guided Biopsy ~ do ~ WhoRB For Occult Mammographic Char 1=Family Practice Abnomaiity: Recommended Doctor only(FPD) by: 2=Radiologist only 3=Both F PD and Radiologist 4=Surgeon 5=Nurse Practitioner 8=Other 9=Undocumented as do an. Call Call if Problem Worsens Char Yes I No Follow-up Common To Any Abnor: 96 Doctor only(FPD) 2=Radiologist only 3=Both FPD and Radiologist 4=Surgeon 5=Nurse Practitioner 8=Other 9=Undocumented Variable Value Labels Naifi Name Label Type Values Note Form IV - Follow-Up Form Rou1 Routine Screening Char Yes I No Follow-up Common To Any Abnor. WHORS Recom. by: Char 1=Family Practice ~ do ~ Doctor only(FPD) 2=Radiologist only 3=Both FPD and Radiologist =Surgeon 5=Nurse Practitioner 8=Other 9=Undocumented RegMam Immediate Mammo Workup Char Yes I No Follow-up Common Regular Mammo To Any Abnor: EV Immediate Mammogram Char Yes I No Follow-up Common Workup: Extra Mammogram To Any Abnor: Views CC Immediate Mammogram Char Yes I No ~ do ~ Workup: Cone or Spot Compression MV Immediate Mammogram Char Yes I No ~ do ~ Workup: Magnification Views WhoEV Recom. by: Char ~ do ~ 1=Family Practice Doctor only(FPD) 2=Radiologist only 3=Both FPD and Radiologist 4=Surgeon 5=Nurse Practitioner 8=Other 9=Undocumented MInterM Interval Followup How many Char Follow-up Common month mammo Yes I No To Any Abnor. MInterC1 Interval Followup: (How many) Char Yes I No ~ do ~ month CBE Wholnt Recom. by: Char 1=Family Practice Follow-up Common To Any Abnormalities: 97 Variable Value Labels Nifid Name Label Type Values Note Form IV - Follow-Up Form ultra Ultrasound Char Yes / No Follow-up Common To Any Abnon WhoUIt Recom. by: Char 1=Family Practice Follow-up Common Doctor only(FPD) To Any Abnor: 2=Radiologist only 3=Both F PD and Radiologist 4=Surgeon 5=Nurse Practitioner 8=Other 9=Undocumented SR Surgical Referral Char Yes I No Follow-up Common To Any Abnon WhoSR Recom. by: Char 1=Family Practice Follow-up Common Doctor only(FPD) To Any Abnor: 2=Radiologist only 3=Both FPD and Radiologist 4=Surgeon 5=Nurse Practitioner 8=Other 9=Undocumented undoc1 Undocumented Char Yes / No Follow-up Common To AnyAbnor: Comments Other Recommendations Or Char Comments for follow- Comments Concerning up Abnormality(ies): Gcomment General Comments About This Char General comments Visit: about this case 98 Appendix C Tables 99 Table l: A Woman’s Chances of Breast Cancer Increases with Age Ag: Incidence Rate By age 30 1 out of 2,212 By age 40 1 out of 235 By age 50 1 out of 54 By age 60 1 out of 23 By age 70 1 out of 14 By age 80 1 out of 10 Ever 1 out of 8 Source: F euer EJ, Wun LM. DE V CAN: Probability of Developing or Dying of Cancer. Version 4.0. Bethesda MD: National Cancer Institute. I 999. Table 2: Five Year Survival Rate by Age Age Survival Rate Younger than 45 81% Ages 45-64 85% Ages 65 and older 86% Table 3: Survival vs. Treatment Cost Source: American Cancer Society Stpge Average cost in S Mammography 90 Diagnostic Workup 500 Biopsy 2,000 Early stage treatment 11,000 Late stage treatment 140,000 Table 4: Guidelines by various organization for Breast Cancer Screening in Women Organization Ages 40-49 yr Ages 50 yr and older American Cancer Society Annual mammogram, Annual CBE, Monthly BSE Annual Mammogram, Annual CBE, Monthly BSE National Cancer Institute Mammogram every 1-2 yr Mammogram every 1-2 yr US Preventive Services Mammogram every 1-2 yr, with or Mammogram every 1-2 yr, with Task Force without CBE or Without CBE . Inadequate evidence to , 2mg:?£eco'b?cfi:ef re 00 nd or not recommend Ages 50-69.1rri2a;r;rgpgram every "lad mammography 100 Table 5: Staging and Survival Rates Stage 5-year Relative Survival Rate 0 100% I 98% IIA 88% IIB - 76% IIIA 56% IIIB 49% IV 16% Source: American Cancer Society Table 6: Overall Survival Rate After 5 years 85% After 10 years 71% After 15 years 57% After 20 years 52% Source: American Cancer Society Table 7: American women who have had a Mammogram within past 2 Years All Women Over 40 66.9% White, Non-Hispanic Women Over 40 68% Black, Non-Hispanic Women Over 40 66% Hispanic Women Over 40 60.2% Women Over 40 Below Poverty Level 50.5% Women Over 40 Above Poverty Level 69.3% 1998, Source: National Center for Health Statistics Table 8: List of intervention and control sites Intervention Sites: Control Sites: Sparrow/MSU Genesys Health Systems, Flint St. Lawrance/MSU McLaren Regional Medical Center, Flint Kalamazoo Center for Medical Studies Munson Medical Center, Traverse City Mid-Michigan Regional Medical Providence Hospital, Southfield Center - Midland Saginaw Cooperative Hospitals, Inc. 101 Table 9: List of site and assigned site numbers Site Site Number SPARROW 1 STLAWERENCE 2 KALAMAZOO 3 MIDLAND 4 SAGINAW 5 GENESYS 6 MCLAREN 7 TRAVERSE CITY 8 PROVIDENCE 9 Table 10: List of number of subjects per site Sr. No. Site # of Subjects 1 SPARROW 1886 2 STLAWERENCE 953 3 KALAMAZOO 1228 4 MIDLAND 2237 5 SAGINAW 1512 6 GENESYS 1276 7 MCLAREN 781 8 TRAVERSE CITY 1321 9 PROVIDENCE 2036 TOTAL 13230 Table 11: Cleaning of Duplicate Data Entries Sr. . Original After Data Number Of No. Site Numbers Cleaning Duplicate records 1 SPARROW 1886 1624 262 2 STLAWERENCE 953 953 0 3 KALAMAZOO 1228 1222 6 4 MIDLAND 2237 2178 59 5 SAGINAW 1512 1486 26 6 GENESYS 1276 1247 29 7 MCLAREN 781 773 8 8 TRAVERSE CITY 1321 1302 19 9 PROVIDENCE 2036 2032 4 TOTAL 13230 12817 413 102 Table 12: Symptom Codes Code Assimd Description of Abnormality detected by Symptoms presented 1. Lump 2. Nipple Discharge 3. Skin/Nipple Changes 4. Breast Pain 5. Occult Mammographic Abnormality 6. Rash under breasts Qntertri go) 7. Heavy, full breasts 8. Boils, pus 9. Prickly, itchy nipple 10. Auxiliary lump 11. Mole, pigmented lesion 12. Macromastia 13. Breast swelling on HRT 14. Increased breast size 15. Cyclic breast enlargement 16. Bruising, abrasion 17. Rash on skin of breast 18. Nipple bump 19. Skin bump 20. Leaky implant 21. Greenish-yellow m'pple discharge 22. Breast abscess 23. Pain under arm 24. Sore on breast 25. Chemical burn on breast 26. Auxiliary pain 103 Table 13: Formula for CBE screening rate calculation FORMULA CBE SCREENING RATE 1 SYMPTOM WITHIN 30 DAYS OF CBE 2 ABNORMAL MAMMOGRAM BEFORE CBE 3 CBE DOCUMENTED 4 AVAILABLE SCREENING ELIGIBLE PATIENTS 5 E-CODE ONE AND BREAST CARE 6 E-CODE 1 7 E-CODE 2 8 E-CODE 3 AND CARE 6 NUMERATOR FOR SCREENING C1 CBE DOCUMENTED C2 CBE DOCUMENTED + E-CODE 3 AND CARE 6 DENOMINATOR FOR SCREENING D1= 6 - g+2) E-CODE 1 - (1+2) D2: (6 + 8) - (1+2) (E-CODE 1 & 3*6) - (1+2) D3= (6 + 7 + 8) - (1+2) (E-CODE 1, 2 & 3*6) - (1+2) SCREENING RATE PATIENT C1 / D1 PHYSICIAN C2 / D2 PUBLIC HEALTH C2 / D3 Table 14: Formula for Mammogram Ordered FORMULA MAMMOGRAM ORDERED 1 DIA ORD 3 E-CODE ONE 4 E-CODE TWO 5 E-CODE THREE CARE SIX NUMERATOR FOR MAM ORDER SCREENING RATE Nl=l+2 DIA ORD + TEXTTEL 13 (REMINDERS) N2=l+2+5 DIA ORD + TEXTTEL 13 + TEXTTEL 3‘6 DENOMINATOR FOR MAM ORDER SCREENING RATE D1 = 3 + 2 E-CODE 1 + TEX'ITEL 13 (REMINDERS) D2=3 +5+2 E-CODE1+3*6+TEXT1‘EL 13 D3=3+4+5+2 E-CODE1+2+3*6+TEX'ITEL13 SCREENING RATE N1 / D1 BASED ON E-CODE ONE ONLY N2 / D2 BASED ON E-CODE ONE AND ECODE 3*6 N2 / D3 BASED ON E-CODE ONE, TWO AND 3*6 104 Table 15: Formula for Mammogram Done FORMULA MAMMOGRAM DONE 1 DIA DONE 2 E-CODE ONE 3 E-CODE TWO 4 E-CODE THREE CARE SIX NUMERATOR FOR MAM DONE SCREENING RATE N1 = l + 2 DIA DONE N2=1+2+5 DIADONE+ECODE3*6 DENOMNATOR FOR MAM DONE SCREENING RATE D1 = 2 E-CODE 1 D2=2+4 E~CODE1+3*6 D3=2+3+4 _ E-CODE1+2+3*6 SCREENING RATE N1 / D1 BASED ON E-CODE ONE ONLY N2 / D2 BASED ON E-CODE ONE AND ECODE 3*6 N2 / D3 BASED ON E-CODE ONE, TWO AND 3*6 Table 16: Formula for Mammogram Ordered and Done FORMULA MAMMOGRAM ORDERED AND DONE 1 DIA DONE & OR ORDERED 2 E-CODE ONE 3 E-CODE TWO 4 E-CODE THREE CARE SD( NUMERATOR FOR MAM DONE & OR ORDERED SCREENING RATE N1=1+2 DIADONE&ORORDERED N2=1+2+5 DIADONE+ECODE3*6 DENOMINATOR FOR MAM ORDER SCREENING RATE D1 = 2 E-CODE 1 D2=2+4 E-CODE1+3*6 D3=2+3+4 E-CODE1+2+3*6 SCREENING RATE N1 / D1 BASED ON E-CODE ONE ONLY N2 / D2 BASED ON E-CODE ONE AND ECODE 3*6 N2 / D3 BASED ON E-CODE ONE, TWO AND 3*6 105 Table 17: Formula for Mammogram either Ordered or Done FORMULA I MAMMOGRAM EITHER ORDERED OR DONE NUMERATOR FOR MAM EITHER ORDERED OR DONE N1= 1+2 DIA DONE&OR ORDERED N2=1+2+5 DIADONE+ECODE3*6 DENOMINATOR FOR MAM EITHER ORDERED OR DONE D1 = 2 E-CODE l D2=2+4 E-CODE1+3*6 D3=2+3+4 E-CODE1+2+3*6 SCREENING RATE N1 / D1 BASED ON E-CODE ONE ONLY N2 / D2 BASED ON E-CODE ONE AND ECODE 3*6 N2 / D3 BASED ON E-CODE ONE, TWO AND 3*6 Table 18: Formula for Combined Screening Rate FORMULA COMBINED SCREENING RATE 1 CBEDOC AND MAMDOC 2 E-CODE ONE 3 E-CODE TWO 4 E-CODE THREE CARE SIX NUMERATOR FOR MAM DONE & OR ORDERED SCREENING RATE N1=1 +2 DIA DONE&OR ORDERED N2=1+2+4 DIADONE+ECODE3*6 DENOMINATOR FOR MAM DONE & OR ORDER SCREENING RATE D1 = 2 E-CODE l D2=2+4 E-CODE1+3*6 D3=2+3+4 E-CODE1+2+3*6 SCREENING RATE N1 / D1 BASED ON E-CODE ONE ONLY N2 / D2 BASED ON E-CODE ONE AND ECODE 3*6 N2 / D3 BASED ON E-CODE ONE, TWO AND 3*6 106 Table 19: Two year screening rates for the DOD study Intervention Screening Rates ‘ Overall Sites Control Sites CBE Done Patient Based I 68.4 71.8 63.8 Physician Based I 72.2 75.9 66.8 Public Health I 55.5 60.5 48.7 Mammoggm Ordered Patient Based 61.7 57.9 66.9 Physician Based 65.7 63.3 69.3 Public Health 52.0 51.7 52.3 Mammogram Done Patient Based I 56.8 59.2 53.5 Physician Based I 61.5 64.6 57.0 Public Health I 48.3 52.5 42.7 Mammogam Ordered and Done Patient Based 42.8 43.6 41.6 Physician Based 49.0 51.1 45.9 Public Health 38.5 41.5 34.4 Mammogam either Ordered or Done Patient Based I 74.3 72.0 77.6 Physician Based 77.1 75.7 79.2 Public Health I 60.6 61.5 59.3 CBE and Mammogglm Combined Rate Patient Based I 54.2 52.8 56.3 Physician Based I 59.2 59.0 59.5 Public Health I 46.5 48.0 44.6 107 Appendix D Figures 108 Figure 1: Cancer Incidence Rates (Age-adjusted rate per 100,000 to the 2000 US standard population) for Women, US, 1975-2000 150‘ at 100i¥ Colon&Rectum >6 Uterinecorpus _ __ ' Ovary q “.32:22832333823383383383338 Source: Surveillance, Epidemiology, and End Results Program, 1975-2000, Division of Cancer Control and Population Sciences, National Cancer Institute, 2003. Figure 2: Cancer Death Rates (Age-adjusted to the 2000 US standard population) for Women, US, 1930-2000 60 4 Lung 40 ‘ Uterus w ‘ Breast 4 IF// \ Ovary ,....-..........v... .. ___v ”.22.... A v 0 ”m Pancreas / In ID 0 In O IO 0 In 0 ID is S 3 E a .. 3; a :c: E a a 3.3 Source: US Mortality Public Use Data Tapes 1960-2000, US Mortality Volumes [930-1959, National Center for Health Statistics, CDC, 2003. 109 Figure 3: Mammography 2000 Source: Behavioral Risk Factor System, CDC MICHIGAN 82% _ IT.“ :‘ ‘ I 84 fl . ,. ' 711 § 79 Age-adjundeemmageotwomenagedzwyeam : ,“ ",9 _ withinthepaaiZyem D eo-u 75-78 I 77.79 I 80-81 UnledStetesrata: 71; mmmlomebm Figure 4: Breast Cancer Deaths 1999 Source: National Vital Statistics System, NCHS, CDC 5a2§§§882 8.3888938?” mmunoauo. 20.9 r W death rates per 1oo.ooo mi. ventilation [I 20.5-252 ass-26.5 I 20.6-28.2 I 28.3-35.0 United States rate: 27.0. Hum mama W: as 110 State Specific Annual Breast Exam Guidelines Figure 5: MICHIGAN: Annual Exam age 40 + D an» “Mumfi bmlznumo-OJ‘nr-narmwo - hm:mi.u“.nlmmnleoo non- Sm' mumwummqw Figure 6: Triad of Breast Cancer Screening: Mammography Clinical Breast Self Breast Exam Examination 111 Continuum of Breast Care Figure 7 3-30:0". Leno... ncoEEouom a 3:63". $8. 05 Co 2533. .2030 2m we Egan. Em: no 3030M EFOLMU Emu... .650 no .5508 En! 03.3:qu SEQ 5595 30:50 112 Figure 8: Early Diagnosis by Mammogram Source: Osuch JR, Pathak DR, Barry HC, Zuber TJ, Slide 66 Number of Cdl Doublings 0 5 10 15 20 25 30 35 40 Latent Phase 0 5 10 Years of Growth Figure 9a: DOD Study Forms used for Breast Care data abstraction: Study Dan base consists of four forms ,- “Form N Follow-up “Form twiclt “Faun Ill Test Entry" records result entry many: each breast care term" comists record the received of the breast follow-tn that care relabd test results. occuned or was recommende - O '113 Figure 9b: DOD Study and Breast Care: m Sagan? .. Bank 650 5 322.582 EEuoEEm! . ominouutaooo 250.520 . as 3.32.3 3-30:3. Lanai ncoEEooom d on. 382 2 uuucofi 3381 «map 05 Co 395m .30an n5: .Eon \ >anuoEEm! 3E0 mem 535 325.0 a .2 we 322.8 LES Dace «=68 «so... 5 atoms gnu «macaw L8 :m_> mem 53.5 tom 114 Figure 10: Permanent Dataset Sub Setting Strategy mumwnsm 8v moi “accustom mv miiv muse macaw nisv Tau Gav Tao mteO mtg... ..\. moo 358:5 5]. satin—.835... “flip-0:2 0+0 fie mite ntsv mite A... 8 has. a. . sue}: a to; 2. r 3 M. n...» .J tail nu.‘ \4 fl“ 28> 8 t 8 15!; Kn}?! ,1? A... An»; 23> 8 51...”... ntrtv L 115 Figure 11: Method of Screening Rate Calculation Patient Based (Potential to Screen who vldbd the clinic) Patient Screening Rate Physician Based Physician Screening Rate Public Health . Screening Ra ‘ ‘ ' ‘ Figure 12: Method of Screening Rate Calculation Patient Screening Rate Physician Screening Rate 116 Figure 13: Criteria for Screening Rate Calculation: I Dataset I Remove Symptomatic Patients T {Asymptomatic PatientsJ T Remove Remove Abnormal Mammograms Abnormal CBE Prior to CBE Prior to Mammogram CBE S-Rate ( , .1 , . Eligible for Eligible for v3 CBE Screening Screening Mammogram Figure 14: Type of Mammogram Screening Rates: MAMMOGRAM ORDERED MAMMOGRAM ORDERED AND OR DONE 117 Figure 15: Age Distribution of Eligible Patients in the study 800 2:: [k / \A- .\ ‘\ l Frequency 8 c u 6 Q \ oIlljiITTjIIIIITIIIIIIIIIIIITTIIITHII s‘s‘b‘bb“PSPSPPBSBWRBSPSOOBSG Age of Patient in Years I 100 Figure 16: Age distribution of the patients documented with self history of Breast Cancer 60 50 CI- Q Frequency 0) c 20 ' lo . 0 ;.;.;,;;,... .. .1 1,915.}... fl 7 I _ I -:~;- ~ . 1. - «I LT 45 46 to 50 51 to 55 56 to 60 61 to 65 66 to 70 Age group 118 Figure 17: Time Interval between Mammogram Ordered and Mammogram Done 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Overall intervention Control llSame Day EVifithin 1 Months E2 to 6 Months D7 to 12 Months 3 13 to 24 Months 119 References References: 1. WHO. Global cancer rates could increase by 50% to 15 million by 2020. 3 April 2003. IQ://www.who.int/mediacentre/news/releases/2003/pr27/en/ 2003. 2. ACS. American Cancer Society, Breast Cancer Facts and Figures 2003-2004. Atlanta, GA, 2003. 2003. 3. Jatoi 1, Miller AB. Why is breast-cancer mortality declining? Lancet Oncol 2003; 4(4):251-4. 4. NCI. Early Detection - Breast Cancer Screening. http://progressreport.c2_mcegov/ 2005. 5. Vacek PM, Mickey RM, Worden JK. Reliability of self-reported breast screening information in a survey of lower income women. Prev Med 1997; 26(3):287-91. 6. NCI. NCI Website: What you need to know about - Breast Cancer. httg://www.cgmcer.gov/cancerinfo/wyntlg/breafl 2003. 7. Walker RA, Lees E, Webb MB, Dearing SJ. Breast carcinomas occurring in young women (< 35 years) are different. Br J Cancer 1996; 74(11):1796-800. 8. Nixon AI, Neuberg D, Hayes DF, Gelman R, Connolly JL, Schnitt S, et :11. Relationship of patient age to pathologic features of the tumor and prognosis for patients with stage I or II breast cancer. J Clin Oncol 1994; 12(5):888-94. 9. ACS. Cancer Facts & Figures - 1997, American Cancer Society, 1997. 1997. 10. Barlow WE, Taplin SH, Yoshida CK, Buist DS, Seger D, Brown M. Cost comparison of mastectomy versus breast-conserving therapy for early-stage breast cancer. J Natl Cancer Inst 2001; 93(6):447-55. 11. Hensrud DD. Clinical preventive medicine in primary care: background and practice: 1. Rationale and current preventive practices. Mayo Clin Proc 2000; 75(2):165-72. 12. Lynge E. [Screening for cancer. International knowledge and Danish practice]. Ugeskr Laeger 2002; 164(22):2892-7. 121 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. Bassett LW HR, Bassford TL. High-Quality Mammography: Information for Referring Providers. October 1994. Quick Reference Guide for Clinicians No. 13.AHCPR Publication No. 95-0633.Rockville, MD Agency for Health Care Policy and Research, Public Health Service, US. 1994. Overgaard M, Hansen PS, Overgaard J, Rose C, Andersson M, Bach F, et a1. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med 1997; 337(14):949-55. Fletcher SW, Black W, Harris R, Rimer BK, Shapiro S. Report of the International Workshop on Screening for Breast Cancer. J Natl Cancer Inst 1993; 85(20): 1644-56. J atoi 1. Breast cancer screening. Am J Surg 1999; 177(6):518-24. Elmore JG, Armstrong K, Lehman CD, Fletcher SW. Screening for breast cancer. Jama 2005; 293(10):]245-56. de Koning HJ. Mammographic screening: evidence from randomised controlled trials. Ann Oncol 2003; 14(8):1185-9. Timins J K. Controversies in mammography. N J Med 2005; 102(1-2):45-9. Brekelrnans CT, Peeters PH, Faber JA, Deurenberg JJ, Collette HJ. The epidemiological profile of women with an interval cancer in the DOM screening programme. Breast Cancer Res Treat 1994; 30(3):223-32. Tabar L, Fagerberg G, Chen HH, Duffy SW, Smart CR, Gad A, et al. Efficacy of breast cancer screening by age. New results from the Swedish T wo-County Trial. Cancer 1995; 75(10):2507-17. Humphrey LL, Helfand M, Chan BK, Woolf SH. Breast cancer screening: a summary of the evidence for the US. Preventive Services Task Force. Ann Intern Med 2002; 137(5 Part 1):347-60. Buist DS, Porter PL, Lehman C, Taplin SH, White B. Factors contributing to mammography failure in women aged 40-49 years. J Natl Cancer Inst 2004; 96(19): 1432-40. 122 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. Aubard Y, Genet D, Eyraud JL, Clavere P, Tubiana-Mathieu N, Philippe HJ. Impact of screening on breast cancer detection. Retrospective comparative study of two periods ten years apart. Eur J Gynaecol Oncol 2002; 23(1):37-41. Olsen AH, Njor SH, Vejborg I, Schwartz W, Dalgaard P, Jensen MB, et a1. Breast cancer mortality in Copenhagen afier introduction of mammography screening: cohort study. ij 2005; 330(7485):220. Zwahlen M, Bopp M, Probst-Hensch NM. Mammography screening in Switzerland: limited evidence from limited data. Swiss Med Wkly 2004; 134(21-22):295-306. Jonsson H, Nystrom L, Tomberg S, Lundgren B, Lenner P. Service screening with mammography. Long-term effects on breast cancer mortality in the county of Gavleborg, Sweden. Breast 2003; 12(3):]83-93. Tabar L, Duffy SW, Yen MF, Warwick J, Vitak B, Chen HH, et a1. All-cause mortality among breast cancer patients in a screening trial: support for breast cancer mortality as an end point. J Med Screen 2002; 9(4):]59-62. Allweis TM, Nissan A, Spira RM, Sklair-Levy M, Freund HR, Peretz T. [Screening mammography for early diagnosis of breast cancer: facts, controversies, and the implementation in Israel]. Harefuah 2003; 142(4):281-6, 317. Fei g SA. Decreased breast cancer mortality through mammographic screening: results of clinical trials. Radiology 1988; 167(3):659-65. Blanks RG, Moss SM, McGahan CE, Quinn MJ, Babb PJ. Effect of NHS breast screening programme on mortality from breast cancer in England and Wales, 1990-8: comparison of observed with predicted mortality. ij 2000; 321(7262):665-9. Harris R, Leininger L. Clinical strategies for breast cancer screening: weighing and using the evidence. Ann Intern Med 1995; 122(7):539-47. Syrnons AB, Mahoney MC, Englert J, Mirand AL. Variations in approaches to breast cancer screening among primary care physicians. J Cancer Educ 2002; 17(4):205-10. Ahmad F, Stewart DE, Cameron JI, Hyman 1. Rural physicians' perspectives on cervical and breast cancer screening: a gender-based analysis. J Womens Health Gend Based Med 2001; 10(2):201-8. 123 35. 36. 37. 38. 39. 40. 41. 42. NCCS. ABC Glossary of cancer terms National Cancer Center Singapore. http://www.nccs.com.sg[. Bodiya A, Vorias D, Dickson HA. Does telephone contact with a physician's office staff improve mammogram screening rates? Fam Med 1999; 31(5):324-6. Cianfrocca M, Goldstein LJ. Prognostic and predictive factors in early-stage breast cancer. Oncologist 2004; 9(6):606-16. ACS. American Cancer Society Breast Cancer Facts & Figures: http://www.ca_1ncer.org[. 2004. Dumitrescu RG, Cotarla 1. Understanding breast cancer risk -- where do we stand in 2005? J Cell Mol Med 2005;9(1):208-21. Claus EB, Stowe M, Carter D. Family history of breast and ovarian cancer and the risk of breast carcinoma in situ. Breast Cancer Res Treat 2003; 78(1):7-15. Weber W. Cancer control by family history. Anticancer Res 1993; 13(4):1197-201. Familial breast cancer: collaborative reanalysis of individual data from 52 epidemiological studies including 58,209 women with breast cancer and 101,986 women without the disease. Lancet 2001; 358(9291):1389-99. 124 IIII‘IIIIIIIIIIIIII