BreastCheck and the fear of failure
An expectation that medical tests should be infallible is at odds with reality and raises serious concerns for cancer screening in Ireland
Cancer screening programmes are public health initiatives designed to reduce mortality by detecting the disease at an early stage when it may be more easily and effectively treated. The Irish National Screening Service currently operates three such programmes for breast, cervical and bowel cancer. These programmes are already contributing to improvements in cancer mortality.
The national breast screening programme, BreastCheck, was established with the stated aim of reducing the number of women dying from breast cancer by 20 per cent. BreastCheck was established to the highest international standards underpinned by a clear patient charter, quality standards and performance monitoring. The results are published annually.
The most recent external review was by Euref, the official European accreditation group, in 2017. Euref awarded BreastCheck its highest standard of accreditation, a designation which remains valid until 2021. Operating to these standards, the programme has diagnosed more than 12,000 cases of breast cancer. It now conducts screening mammograms on 165,000 Irish women each year.
It has been acknowledged from the outset that, while mammography is the best available test to screen for breast cancer, it is not perfect. It is neither 100 per cent sensitive nor 100 per cent specific. It will not detect all cancers and not all the lesions that are suspicious on a mammogram will turn out to be cancer.
As a result, mammographic screening does give rise to false positives and false negatives. Some healthy women with an abnormal mammogram are recalled. While the final result may be normal, such a recall is associated with significant anxiety for women and their families. Research has shown that it may take a woman up to one year to get back to baseline anxiety levels about breast cancer following a recall. We have also shown from our own BreastCheck research that women are significantly less likely to return for subsequent screening mammograms if they have previously been recalled.
All involved with breast screening programmes are acutely aware of the emotional and logistical challenges caused by recall. It is not possible, or acceptable, to recall every woman for more detailed testing to try to ensure that no cancer has been missed. As a consequence, there are published international standards designed to keep recall rates as low as possible without compromising cancer detection rates.
Screening mammography will diagnose more than 1,000 cases of breast cancer, but fail to diagnose approximately 320 cases every year in Ireland. These women develop signs or symptoms in the two years after a normal screening mammogram. Such tumours are termed interval cancers. They may have been missed due to factors including the type and rate of growth of the particular breast tumour, or be invisible on a mammogram. A small number can be attributed to human perceptual or interpretive error, but every effort is made to minimise such occurrences.
All mammograms are independently viewed by two consultant breast specialist radiologists, yet despite this and despite all the training and quality assurance efforts of the staff in the programme, interval cancers will occur. They cannot be eradicated. They are a predictable and inevitable component of mammographic breast screening programmes. Indeed, there are even published international standards for expected numbers of interval cancers.
In a programme operating to the highest standards, two interval cancers will still arise for every 1,000 women screened. The Irish interval cancer rates have been published as part of the BreastCheck annual report and are in fact lower than required by international standards.
The limitations of mammography are openly communicated to those who undergo screening. Women attending BreastCheck are informed at all stages that screening will not diagnose all cancers. The women sign a consent form acknowledging this limitation, prior to proceeding with a mammogram.
The recent judgment by High Court judge Kevin Cross in the Ruth Morrissey case raises serious concerns for all those involved in cancer screening in Ireland. The judgment relates to cervical screening. It states that screeners must have absolute confidence that a slide is normal if it is to be so deemed. The consequences for cervical screening are readily apparent. Its implications for breast and colon screening programmes are uncertain at this time.
However, there is no test in medicine that offers clinicians absolute confidence. Every test, whether it is performed in a laboratory, an X-ray department or on a ward, has an inherent margin of error and this is the case whether the test is performed by a nurse, doctor, scientist or other health professional. Screening mammography, as I have outlined, is no different.
A judgment that gives rise to a public expectation that medical tests should be infallible is at odds with reality and is a serious threat to the continued provision of this important health initiative.
The judgment also refers to the matter of false negatives and false positives, arguing that a significant increase in the rate of false positives is justified to reduce or eliminate false negatives and would be readily tolerated by the public. Again, such a position is at odds with established screening principles and, as I have outlined, does not tally with our experience of women recalled for assessment.
The WHO criteria, determined by Wilson and Jungner in 1968, set out key requirements of a population-based screening programme. There are ten such parameters, including that a disease should be a significant risk to public health, the test should be acceptable to the public and there should be an accessible intervention that changes the course of the disease. A further important requirement is that the screening should be economically justifiable and cost effective.
If the costs of the screening programme were to rise to accommodate the demands of an increased recall rate, with the increased numbers of tests and potentially unnecessary surgical interventions that would go along with it, the cost- effectiveness of the programme would be undermined. Should legal and financial settlements be added to that cost base, the programme could be unjustifiable and unsustainable.
A diversion of the funds from screening to symptomatic breast services for women presenting to their doctor with symptoms may then be considered more appropriate than maintaining an increasingly expensive screening programme.
Already the screening controversy and the environment around it has had an impact on BreastCheck, with the failure of recent efforts to recruit consultant radiologists.
While there are likely to be other factors at play for some Irish doctors working abroad, their decision to reject opportunities in Ireland is due in no small measure to the anxiety generated by working in an unacceptably demanding environment compared with other international programmes.
We must also remain vigilant to ensure that this increased pressure and fear of litigation does not translate into a defensive practice with a steady increase in recall rates and the harm potentially associated with that outcome.
Those delivering the programme must keep a constant focus on the fact that fewer women in Ireland will die from breast cancer due to the breast screening programme. Indeed, it is estimated that 150 lives are saved in this country every year due to breast screening. This has translated into improved overall survival rates from breast cancer in this country, a result that reflects not just breast screening, but also huge advances in medical oncology, surgery and radiation oncology. I would like BreastCheck to continue, and survival from breast cancer to continue to improve. Sadly, no screening test can detect all cancers, but the reduction in mortality through screening for the whole population is clearly welcome.
Unfortunately for individuals whose cancer is not diagnosed through screening, the consequences may be grave, and can give rise to the feeling that the programme has failed.
However, such failures are to be expected and are, at this point in time, largely unavoidable. Whether the Cross judgment signals the end of cancer screening programmes in Ireland is still uncertain. It is clearly time for an informed discussion between clinicians, the HSE, the Department of Health, our politicians, the judiciary and the Irish public on whether screening should continue in this country.
Professor Ann O’Doherty is a consultant radiologist and clinical director of BreastCheck