|
Home | Issues | eMJA shop | Classifieds | Contact | More... | Topics | Search | Login | Buy full access |
Paul S Craft, John G Primrose, Julie A Lindner and Peter R McManus
For editorial comment, see Collins
Abstract - Introduction - Methods - Data source - Study design - Analysis by demographic characteristics - Results - Discussion - Acknowledgements - References - Authors' details
Make a comment - Register to be notified of new articles by e-mail - Current contents list - ©MJA1997
Abstract |
Objective: To examine patterns of surgical
management of breast cancer among Australian women. Design: Retrospective survey of Medicare records (a national dataset of all services rendered on a "fee-for-service" basis for which a Medicare benefit has been paid). Patients: All Australian women (4683) who underwent surgery consistent with being for breast cancer in 1993 and for which Medicare benefits were paid. Main outcome measures: Proportions of women undergoing different forms of mastectomy, breast-conserving surgery and axillary surgery by patient age and State and region (urban or rural) of residence. Results: Modified radical mastectomy was the most common surgery, performed in 2097 of the 4683 women (44.8%), while 1868 (39.9%) had breast-conserving surgery. Frequency of breast conservation decreased significantly with age and varied significantly between States and region of residence. It ranged from 34% in Western Australia to 49% in South Australia and the Northern Territory, and from 34% among rural women to 42% among urban women. Axillary surgery was recorded for 83% of all women studied. Conclusions: There was substantial geographical variation in patterns of surgical management for breast cancer. The tendency for rural women to undergo mastectomy rather than breast-conserving surgery may reflect the relative lack of access to postoperative radiotherapy. We are unable to explain the variation between States. |
Introduction |
Breast cancer is the commonest malignancy among Australian women
after non-melanocytic skin cancer and is the commonest cause of
cancer deaths in women.1
Until recently, the optimal management of breast cancer that is
limited to the primary site (with or without axillary nodal
involvement) was unclear, with some advocating routine mastectomy
and others advocating a breast-conserving approach for selected
patients. After randomised controlled clinical trials, a consensus
has emerged that most patients with Stage I and II carcinoma of the
breast (comprising about 80% of patients presenting with breast
cancer) can be managed with a breast-conserving approach, with rates
of local and regional control and survival equivalent to those
obtained with mastectomy.2-6
Breast-conserving treatment also seems less disturbing to
perceived body image and sexuality than mastectomy.7
Surgical management of breast cancer has been found to vary significantly between different population groups in the United States.8,9 We investigated surgical management of Australian women treated for breast cancer by analysing Medicare records, and assessing variations in management by patient age and State and region (urban or rural) of residence. |
Methods | |
Data source | Medicare statistics were obtained from the Commonwealth Department of Health and Family Services. They contain details of services rendered on a "fee-for-service" basis for which Medicare benefits were paid. They exclude services to public patients in hospitals and to Veterans' Affairs patients, and services for injuries that are eligible for compensation, unless an interim Medicare benefit is paid. |
Study design |
All women who received at least one Medicare benefit for surgical
items on the Medicare Benefits Schedule (MBS) consistent with breast
cancer surgery during the 1993 calendar year were included. These
items comprised mastectomy -- simple, partial, extended simple,
subcutaneous, modified radical an d radical -- and excision of a
breast lump if followed by axillary dissection and/or radiotherapy.
Medicare records for the first two quarters of 1994 were also examined
to ensure that all subsequent episodes of breast surgery or
radiotherapy were captured. It was assumed that very few patients
would receive these treatments for non-malignant breast disease.
Women who had had only excision of a breast lump but no radiotherapy or
axillary dissection were excluded.
Data recorded in addition to surgical procedure(s) were the date of service, and patient's personal identification number (PIN), date of birth, sex and postcode of residence. The identity of each person in the study was protected by a PIN which could not be decoded by the investigators. |
Analysis by demographic characteristics | Patient age in 1993 was determined from year of birth, and State and region of residence from postcode. Postcodes were classified into regional zones according to the Rural, remote and metropolitan areas classification 1991 census edition.10 When a postcode extended into two regions, a factor based on population density was used to determine in which region it would be placed. To provide sufficient numbers for analysis in each State group, data were combined for the Australian Capital Territory (ACT) and New South Wales (NSW) and for the Northern Territory (NT) and South Australia (SA). Proportions were compared with the chi-squared test. Logistic regression analysis was performed with SPSS version 6.1.11 |
Results |
We identified 4683 women who received at least one Medicare benefit
for surgical items consistent with breast cancer surgery in 1993.
These represented 58% of the 8100 expected incident cases of female
breast cancer in Australia in this year.1 Surgical procedures performed are
summarised in Box 1 (below). Some form of total mastectomy was performed in
2815 women (60.1%), while surgery consistent with breast
conservation was performed in 1868 (39.9%). The most frequently
identified definitive surgical procedure was radical or modified
radical mastectomy, with 2097 services recorded. It was the only
surgical therapy in 1265 women, but was accompanied by other surgical
items in 832.
Patterns of surgery in different patient age groups are summarised in Box 2 (below). Breast-conserving surgery was most frequent in women aged less than 60 years, with the frequency falling to 32% in women aged 70 to 79 years. Variation in type of surgery by age was highly significant (chi-squared for trend = 16.3; P < 0.001).
Frequency of breast-conserving surgery was significantly higher among women resident in urban regions than among those resident in rural regions (comprising large rural towns, rural areas and remote areas). Breast conservation was undertaken in 1462 of 3483 urban women (41.9%), but in only 397 of 1170 rural women (33.9%) (chi-squared = 23.3; P < 0.001). The frequency did not differ between women resident in large rural towns, in rural areas and in remote areas. The residential area of 30 women could not be classified. Patterns of surgery also varied significantly between States (Box 3), with breast-conserving surgery performed in 49.2% of those resident in SA and the NT, but in only 33.8% of those resident in Western Australia (WA). Multivariate analysis showed that this variation was not caused by State differences in population distribution between urban and rural areas.
The axilla was treated surgically, either as part of a mastectomy or under specific axillary dissection MBS items, in 3889 women (83.0%). Frequency of axillary surgery was similar in most age groups, except those aged 80 years or older (Box 2). In this group, only 62.8% of women underwent axillary dissection versus 84.1% of women aged less than 80 years (chi-squared = 70.0; P < 0.001). Frequency of axillary surgery varied slightly between States, with WA having the highest rate and Tasmania the lowest (Box 2), but did not differ between urban and rural residents (2892 of 3483 [83.0%] urban residents and 970 of 1170 [82.9%] rural residents). |
Discussion |
We found that breast-conserving surgery was performed in almost 40%
of Australian women undergoing breast cancer surgery reimbursed by
Medicare in 1993. However, the proportion who underwent
breast-conserving surgery tended to decrease with patient age and
varied significantly beween States and between rural and urban
women.
Our results are similar to those of other studies of surgical care for breast cancer in Australia. In WA, Byrne et al. found that breast-conserving surgery was used in 29.4% of all women treated for breast cancer in 1989, and in 31.3% of those with "potentially eligible" tumours,12 compared with a figure of 33.7% for WA in our study. In NSW in 1988-1991, breast-conserving surgery was used to treat 40% of a series of 105 mammographically detected tumours,13 while a six-month survey of Victorian surgeons in 1990 found that breast conserving surgery was performed in 42% of cases of operable breast cancer.14 We found significant geographical variation in the proportion of women receiving breast-conserving surgery reimbursed by Medicare. Women living in rural or remote locations were more likely to undergo mastectomy. Breast conservation usually necessitates postoperative radiotherapy, and, as radiotherapy services are often not conveniently located for rural populations, this treatment can involve considerable social and financial costs to patients, which may influence the decision to undergo mastectomy. The proportion of women undergoing breast-conserving surgery also varied between States. This variation was not explained by differences between populations, such as proportions of women living in rural and urban regions. The reasons for this are unclear, but preference for mastectomy versus breast-conserving surgery among surgeons and women may vary between States, explaining the observed differences. Older women were less likely to have breast-conserving treatment. Similar patterns have been observed in surveys of breast cancer management.7 Some older women might reasonably prefer to avoid radiotherapy and be relatively less concerned about mastectomy. Alternatively, as the study could not detect women with breast cancer who had had only excision of the primary tumour but neither axillary dissection nor radiotherapy, if this treatment was more common in older patients, then the frequency of breast conservation may have been underestimated. The rates of axillary surgery were generally high and relatively uniform across age groups and between States and regions. Treatment of the axilla is not recommended for in-situ disease15 and may reasonably be withheld for a small invasive cancer, particularly in older women, possibly explaining the omission of axillary dissection for a proportion of study patients.16 Our study, using data collected routinely for administrative purposes, has several advantages over special purpose surveys. These include timeliness, economy, objectivity and standardisation of data recording. In addition, such a study offers the possibility of examining trends in service delivery over time by means of repeated reviews. Special purpose surveys usually rely on healthcare providers to donate their own, or staff, time to complete the survey instrument, which can be a problem for both researchers and healthcare providers. However, a review based on administrative datasets has some limitations. It can provide only descriptive information and does not give insight into the reasons for disease management decisions. As subjects were identified by interpreting MBS items from Medicare records, the presence of an underlying diagnosis of breast cancer was inferred and not independently verified -- for example, by data matching. Therefore, no information was available about clinical or pathological features of the tumours (e.g., stage and size of primary tumour), which would have influenced the choice of surgical management. In addition, some women who underwent surgery for a breast neoplasm other than carcinoma, or for benign disease, may have been included as subjects. We believe that these instances would be rare, and that the vast majority of cases represented women with primary carcinoma of the breast or carcinoma-in-situ. It was also assumed that patients who had one procedure performed on a private "fee-for-service" basis would have subsequent related services provided on the same basis. However, if substantial numbers of women with breast cancer moved to the public hospital system for part of their surgical management, then this study would have underestimated the frequency of subsequent, more extensive, surgery. This might have led to underestimation of the frequency of mastectomy and axillary dissection, and could have had a variable effect between regions because of differences in private health insurance rates. Guidelines for the treatment of early breast cancer have been published recently and are expected to improve standards of care and treatment,6 and to reduce variability in clinical practice. However, the optimal ratio of mastectomies to breast conserving-surgery in Australia remains unknown. Ultimately, the choice of treatment is a decision taken by each individual, informed by advice from her surgeon and information from other sources. Secondary analyses of data collected for reimbursement purposes, such as in our study, can shed no light on such decisions. Detailed surveys of patterns of care should be undertaken and deserve the support of the medical community. |
Acknowledgements | We wish to thank the Medicare Statistics Section of the Department of Health and Family Services for the data supplied within this report. Special thanks to David Wong for his assistance in supplying data. |
References |
(Received 6 Sep 1996, accepted 3 Mar 1997) |
Commonwealth Department of Health and Family Services, Woden, ACT.
John G Primrose, FRACR, Senior Medical Advisor;
Julie A
Lindner, Computer Analyst, Drug Utilization Sub-Committee
Secretariat;
Peter R McManus, BPharm, MMedSc, Secretary,
Drug Utilization Sub-Committee of the Pharmaceutical Benefits
Advisory Committee.
No reprints will be available. Correspondence: Dr P S Craft, Medical
Oncology Unit, Canberra Hospital, Canberra, ACT 2607.
E-mail:
PAUL_CRAFT@dpa.act.gov.au
<URL: http://www.mja.com.au/>
© 1997 Medical Journal of Australia.
We appreciate
your comments.
