Screening Mammography Program of British Columbia M. Graham Clay, MD, FRCSC,Vancouver,BritishColumbia,Canada In an attempt to diagnose breast carcinoma as early as possible, a government-funded screening mammography program in British Columbia discovered 29 malignancies in the first 7 , 1 0 0 women screened. The program was carefully planned by a group consisting of representatives of all appropriate constituents.

The screenees have demonstrated the anticipated increased frequency with increasing age and risk factors; in addition, the frequency of diagnosis is markedly lower in screenees who have had mammograms within the 2 previous years. Exclusive of equipment costs, amortization, and data processing,

the cost per study is $33.81 (Canadian), resulting in a cost of $8,277.62 for each case diagnosed. This includes, for each study, an administrative fee of $ 1 . 0 0 as well as $1.00 devoted to continuing education for the reading radiologists. The success of the program has resulted in encouragement from the Health Ministry to expand rapidly throughout the Province, including use of a mobile van in more remote areas. Continued monitoring and peer review will ensure standard quality control in all of the screening facilities.

Encouraging results from major studies throughout Europe and North America stimulated initiation of a provincial breast screening program in British Columbia, the first such program in Canada [5,6]. The impetus for this development originated with the Breast Tumor Group of the Cancer Control Agency, and the opportunity to begin a pilot program was taken when facilities were vacated by the recently completed National Breast Screening Program. National statistics demonstrate that 10% of Canadian women will develop a breast malignancy during their lifetime. Our provincial studies do not yet include the stage of disease at time of diagnosis, but we do know that the frequency was 98.3/100,000 and the mortality was 30.3/100,000 in 1983, the last year for which complete figures are available [7]. Although the actual frequency is rising, the age-adjusted frequency is not. We expect more than 1,500 new cases each year [8]. Repeated studies have demonstrated that mortality expectations are dependent on the stage of the disease at diagnosis [2]. Therefore, one can anticipate a significant improvement in disease-free survival with diagnosis of tumors that are nonpalpable or of minimal size.

MATERIAL AND METHODS Two factors have assisted in the successful development of our program. The first is the long-term cooperation between the government-funded Cancer Control Agency and British Columbia physicians, The second is an established acceptance of sessional funding for physicians involved in screening programs that, while community-based, are organized cooperatively between the Agency and physicians in private practice. During develreast carcinoma is the most common cause of death opment of this cooperative effort, care was taken to indue to malignancy in Canadian women. Although elude all constituents during the organizational phase. there has been some improvement in survival rates be- Representatives from the Departments of Radiology, cause of advances in management, survival is best im- Epidemiology, Pathology, Radiation Oncology, and Surproved by treatment of the disease at the earliest stage gery of the Cancer Control Agency, together with adminpossible [1]. istrative personnel, were included. Members of the Breast Screening mammography represents an attempt to Tumor Group of the Agency, including both salaried lower mortality from this disease. Although some detrac- employee-physicians and Agency consultants, who are tors suggest that improved mortality rates from early private practitioners working voluntarily within the diagnosis are simply a result of lead-time bias, many Agency, gave significant advice, while the provincial orstudies confirm that early diagnosis by mammography ganization of physicians, the B.C. Medical Association, will co~tribute to a significant improvement in mortality sent representatives from both their Cancer Committee from thigdisease over the long term [1-4]. and the Family Practice section. The radiologists were Fromthe Divisionof GeneralSurgery,Departmentof Surgery,Univer- represented by the B.C. Radiological Society, while acasity of Bdti~,hColumbia,and the Cancer ControlAgencyof British demic input came from the membership of the Head of the Department of Radiology of the University of British Columbia,Vancouver,BritishColumbia,Canada. Requests for reprints should be addressed to M. Graham Clay, Columbia. The Canadian Cancer Society, a volunteer lay MD, 308-888 West 8th Avenue,Vancouver,BritishColumbia,Cana- organization, was included since it is directly involved da, V5Z 3Y1. Presentedat the 76thAnnualMeetingof the North PacificSurgi- throughout the province with both public relations and cal Association,Victoria,BritishColumbia,Canada,November10-11, funding of research. Of paramount importance was the 1989. enthusiastic involvement of the Ministry of Health of the

B

470

THE AMERICAN JOURNAL OF SURGERY VOLUME159 MAY 1990

SCREENING MAMMOGRAPHY IN BRITISH COLUMBIA

British Columbia Government, which not only funded the program but also encouraged us to develop facilities throughout the province at a speed that we had not thought possible. Our pilot study began in July 1988, with a start-up grant of $41,000 and an operating budget of $365,000. Although we had expected to screen 7,600 women during the first 9 months, we actually screened 7,100. The average number of women screened per day with our single machine was 43, but it was found that 50 could be accommodated without difficulty. Following each study, the results were reported to the family practitioner of the screenee. Screenees were notified of the results a few days after notification of the practitioner. RESULTS Throughout the 9-month period, 722 mammograms were reported as abnormal and further investigations were suggested. From these investigations, a total of 29 malignancies were diagnosed, giving a positive diagnosis rate of 4.1 / 1,000. Of the risk factors studied (Table I), family history and age of the patient showed definite trends. There has been a steady rise in incidence each decade from 1.4/ 1,000 in those under 50 years to 9.8/1,000 in those 70 years or older. Those with no family history have an incidence of 3.8/1,000 compared with a higher incidence for those at risk because of family history. Patients whose relatives had unilateral disease or disease that developed at or over 50 years had an incidence of 4.4,/1,000, whereas those with relatives developing bilateral disease or disease under the age of 50 had an incidence of 9.7/1,000. Of particular interest is the fact that women who had mammograms within 2 years prior to entering the program had an incidence of only 0.8/1,000, whereas those who either had no previous mammogram or had a mammogram prior to 2 years earlier had an incidence of 4.8/ 1,000. We may be 95% confident that the true number per 1,000 in the population from which we have a representative sample has been captured by each of the 95% confidence limits in Table I. The actual cost of each screening was $33.81 (Canadian) per patient, a cost that is lower than those reported from other North American centers. This includes all operating expenses, as well as the $7.00 professional fee, which includes $5.00 for the interpreting radiologist, a $1.00 payment to the Director, and $1.00 to underwrite costs of professional education. Equipment costs, amortization, and data collection are not included in the $33.81. For each of the 29 malignancies discovered, the cost of diagnosis by this method is calculated as $8,277.62. The expense of follow-up for those with abnormal mammograms [9] must be included in any calculation of the overall cost of a program such as ours (Table II). Seven hundred twenty-two of those screened were reported as having abnormalities that required investigation. Of these, 32 were not seen by their physician, suggesting that the mammograms were assessed by another radiologist and the findings considered benign, perhaps to be assessed at a later date by reexamination. However, the

TABLE I Risk

Factors

Screened (n)

Diagnosed (n)

n/1,000 *

2,906 2,196 1,545 408

4 9 12 4

1.4 (0.4-3.5) 4.1 (1.9-7.8) 7.8 (4-!3.5) 9.8 (2.7-24.7)

6,318 456 310

24 2 3

3.8 (2.5-5.6) 4.4 (0.5-15.7) 9.7 (2-27.7)

1,234 5,850

1 28

0.8 (0.01-4.5) 4.8 (3.1-6.9)

Age (yr) --70 Family history No Yes, low-risk Yes, high-risk Previous screening Within 2 years Not within 2 years

* Values in parentheses indicate 95 % confidence limits.

TABLE I1

Physician Visits

Not seen Family physician Diagnostic radiologist Surgeon

n

%

32 484 644 262

4 67 89 36

TABLE III

Diagnostic Investigations

Diagnostic mammography Ultrasound Fluid aspiration Tissue aspiration Localization biopsy Other biopsy

n

%

511 182 59 8 110 42

71 25 8 1 15 5

majority were seen by a physician. Our program, which involves only mammographic screening, is intended to be combined with a visit of the screenee to her family physician for physical examination, recognizing that all breast malignancies are not seen mammographically. We believe that 96% of those with abnormalities were seen, but we do not know how many of those with normal studies underwent the physical examination that was intended. Investigations were undertaken in most of those with abnormalities (Table III). The largest number of those investigated, 71%, underwent diagnostic mammography and 25% had ultrasound assessment. Aspiration of cysts was undertaken in 8% and aspiration biopsy in 1%. A total of 152 patients had an open biopsy, 110 by a localization technique. Of those who underwent biopsy, 29 were positive, giving a 19% positive biopsy rate. Although this is lower than we would accept in our own institution, the biopsies were being performed in many institutions throughout the province. Therefore, one cannot expect the very high

THE AMERICAN JOURNAL OF SURGERY

VOLUME 159

MAY 1990

471

CLAY

TABLE IV

Results by Radiologist Radiologist

n

Abnormal Findings

%

1 2 3 4 5

1,395 1,427 1,372 1,365 1,541

123 138 137 149 175

9 10 10 11 11

To~l

7,100

722

10

positive biopsy results achieved in highly specialized centers [4], and we consider the 19% satisfactory for this situation. At our own agency, we have achieved a positive rate of 33% for fine-wire localization biopsies [10]. Positive biopsy rates as high as 60% have been reported from centers with specialized protocols. The biopsies in our patients are being done by many surgeons with variable experience and in many different institutions. The success of our program depends on careful training of our screening radiologists and constant peer review. The number of positive reports by the five radiologists varies from 9% to 11% (Table IV), and attempts are being made, by monthly reviews of randomly selected "abnormal" films, to standardize reporting. In addition, films, both normal and abnormal, are selected at random for assessment by an outside reviewer. With our perceived success, the Health Ministry has requested a more rapid expansion of the program than we would prefer in our attempts to maintain high standards. Before our pilot program was completed, we were asked to establish a second site, and the heavily populated area of Surrey, south of Vancouver, was selected. To serve the more remote areas of the province, a mobile van is under construction, funded and equipped by the Canadian Cancer Society, British Columbia and Yukon Division, at a cost approximating $250,000. Although neither of these is yet completed, we have been asked to expand further. After careful assessment of population bases, our Steering Committee has agreed to develop two more facilities in the Greater Vancouver area, one in a major shopping mall. Another facility will be located in Victoria to service the southern portion of Vancouver Island, The first interior clinic will be developed in the Okanagan Valley, a growing region in the interior of the province. All clinics will have a computer linkage to the central office in Vancouver and all screening radiologists will be involved in ongoing training as well as peer review. We have received great cooperation from radiologists in private facilities during these expansion programs. COMMENTS The Screening Mammography Program, begun with the 9-month pilot program discussed herein, has met with

472

wide local acceptance, and The Ministry Of Health of the Province of British Columbia has encouraged rapid expansion so that facilities will be available to all appropriate women throughout the province. A major factor in our success has been involvement of all constituent members of the medical community affected by the program. Care has been taken from the outset to ensure that quality controls were built into the process and will continue with the expansion. In addition, ongoing data processing will be a function of the central office and assist in ensuring high standards throughout all facilities. At this stage in our development, we are uncertain of the extent that patients will comply with annual return visits. The cost of the program, at $33.81 (Canadian) per study exclusive of equipment costs, amortization, and data processing, compares favorably with similar programs throughout Europe and North America and resuits in a cost of $8,277.62 per case diagnosed. Such an expense is considered to represent a justifiable investment, considering the fact that diagnosis of breast cancer at a stage where it is either nonpalpable Orof minimal size will result in a very high cure rate. Later diagnosis affects survival unfavorably, with the extremely high monetary and social costs resultant from early death from this disease. The investment is small considering the potential returns. Our program, the first government-funded mammography screening program in North America, has shown in its first 9 months of operation that a screening program can be conducted at a unit cost that the province can accept. REFERENCES 1. Tabar L, Dean PB. The control of breast cancer through mammography screening, Radiol Clin North Am 1987; 25: 993-1005. 2. Feig SA. Decreased breast cancer mortality through mammographic screening: results of clinical trials. Radiology 1988; 167: 659-65. 3. McLelland R. Low-cost mass screening with mammography as a means of reducing overall mortality from breast cancer. Radiol Clin North Am 1987; 25: 1007-13. 4. Tabar L, Dean PB. The present state of screening for breast cancer. Semin Surg Oncol 1989; 5: 94-101. 5. Basco V. Mammography. Br Columbia Med J 1988; 30: 87-8. 6. Warren-Burhenne L. Early detection of breast cancer. Br Columbia Med J 1988; 30: 84-6. 7. McBride M, Gallagher R. Cancer in British Columbia: incidence and mortality, 1974-1983. Monograph No. 1, Divisionof Epidemiology, Biometry and Occupational Oncology, Cancer Control Agency of British Columbia, 1987. 8. McBride M, Ma B, Band PR. Projected cancer frequency in British Columbia and regions, 1986, 1991, 1996, 2001. Monograph No. 2, Division of Epidemiology, Biometry and Occupational Ontology, Cancer Control Agency of British Columbia, 1989. 9. Cyrlak D. Induced costs of low-cost screening mammography. Radiology 1988; 168: 661-3. 10. Clay MG. Results of fine wire localization breast biopsiesat the Cancer Control Agency of British Columbia. Can J Surg 1990; 33: 17-9.

THE A M E R I C A N J O U R N A L OFSURGERY VOLUME 159 MAY 1990

Screening mammography program of British Columbia.

In an attempt to diagnose breast carcinoma as early as possible, a government-funded screening mammography program in British Columbia discovered 29 m...
344KB Sizes 0 Downloads 0 Views