Around the world, cervical cancer remains a leading cause of mortality among women. Each year, an estimated 604,000 women worldwide are diagnosed, with roughly 342,000 deaths from the disease.

In 23 countries, cervical cancer is the most commonly diagnosed cancer, and is the leading cause of cancer deaths in 36 countries. The majority of these nations are in sub-Saharan Africa, Melanesia, South America and South-Eastern Asia. In Ireland, 250 women are diagnosed with cervical cancer annually, with 90 deaths. It is the second most common cause of cancer mortality in women aged 25-40 in the country.

Unlike many cancers, cervical cancer is directly connected to infection with a common agent – the Human Papilloma Virus (HPV). This is an incredibly common family of over 170 strains, spread easily through skin-to-skin contact. Some of these strains are most readily transferred during sexual contact, and most sexually active adults are infected by at least one sexually transmitted strain of virus at some stage in their life, peaking between one’s teens and 30s. For most people infected, HPV causes no symptoms and is readily cleared by the immune system within two years. For a smaller subset, however, the virus persists. 

Long-term infections in women can induce abnormalities in cervical cells, collectively called Cervical intra-epithelial neoplasia (CIN), graded from CIN I (minor abnormalities) to CIN III (substantial abnormalities). While the majority of CIN cases regress to normal and eventually HPV-clear status, a proportion persist. Left unchecked, this can in turn progress into cervical cancer. This process is typically slow, usually taking more than 20 years to manifest.

Strictly speaking, no.

Due to the long sojourn time between HPV infection and cervical cancer and because cervical lesions can be readily treated, cervical screening is a highly effective method for preventing future cancers. Accordingly, cervical screening is not designed to catch cancers, but to detect lesions and persistent HPV infection that may one day become cancer. Screening works best when it is repeated at three or five yearly intervals, depending on age.

By detecting and treating these precursors, cervical screening ultimately saves lives. While there is great variation in methodology across the world, national screening programmes have reduced cancer mortality by greater than 80% in countries with established national programmes. 

Incidentally, while cervical screening is not designed to detect cancers, it can also detect very early-stage cancers which can also be readily treated. It is however neither designed as a cancer test, nor intended for this purpose. Like all screening programmes, cervical screening is a designed to reduce the burden of disease in a population. Screening is not designed as a diagnostic test for individuals, but a population-wide measure to reduce disease. This of course leads to direct benefit for many women, but the vital point is that the test is not designed or appropriate for individual diagnosis.  

Up until very recently, screening was typically performed with cytology-centric methods, such as liquid based cytology (LBC). This involves taking a sample of cervical cells from a patient (a smear) and then having two independent scientists who specialise in reading cells determine whether any abnormalities are present on the slide.

This process is difficult, and inherently imperfect, but still beneficial for a population. Cytology has a sensitivity of about 75%, meaning when the system works ideally, it will catch 3/4 of all abnormalities. Conversely, it has a specificity of about 90%, which means that when no disease is present, it will correctly determine this in 9 out
of 10 people.

Crucially, the test is not perfect and nor can be it be made so, but with populations it performs well enough to catch most abnormalities. More recently, Ireland has been a pioneer in adopting primary HPV screening with reflex cytology, which first tests patients for the presence of HPV before a smear is performed. This has the net-effect of boosting the efficacy of screening even further, though this was not available prior to 2018.

Controversy over the Irish programme exploded in 2018, when it was alleged by various politicians that the screening service had neglected to inform women of their cancer diagnoses following a slew of court cases against CervicalCheck and the HSE.

These stories of women facing cancer were emotive, and the political fallout severe, leading to the ousting of the head of the HSE, Tony O’Brien and the clinical director of CervicalCheck, Professor Grainne Flannelly.

The then-minister for health, Simon Harris, offered a free out-of-cycle test to any woman who was worried about her previous tests. This lead to 100,000 additional tests being processed with no additional screening laboratory facilities being made available to read these slides. Medical experts warned against doing this.

No diagnoses were withheld from women. All women were told they had cancer once the cancer was diagnosed. Instead, the failure was one of communication and, arguably, policy.

Ireland, like all countries with a screening programme, performs regular audits of the system to ensure it is performing to standard. If a woman develops cervical cancer, her prior smears are reanalysed to determine whether there were any potentially detectable signs in previous smears. It is expected that abnormalities will be found in
40% of cancer cases.It is inevitable that since the test is only 75% sensitive and misses 1 in 4 abnormalities, these abnormalities can be visible at the time of review, as the reviewer is reviewing the slide in the knowledge that the woman went on to develop cancer.

The overwhelming majority of these arise from the fundamental limits of smear testing. Audits are important to ensure the system is performing to the expected standard.

Audits are exclusively performed on women who already have a cancer diagnosis;  in most counties in the world, these internal quality checks are not communicated to women. Ireland had a policy which seemed to state that the results should be communicated to women. In 221 cases, this did not occur, and several women with discordant results sued the State. However, neither diagnosis nor treatment was delayed in any case, as audits are only performed post-diagnosis.

This information could not and did not change clinical outcomes. Throughout the rest of the world, results of audits are not routinely communicated to women as it is considered a system rather than individual check.

The 221 cases where audit results were not communicated dominated headlines for many years. In some cases, audit results were communicated to a GP or gynaecologist, who declined to notify patients. In other cases, audit outcomes were not sent, or were sent to the wrong address on record.

The 221 figure also concealed important caveats; about 30% of this group had cancer types undetectable by screening, while a further 29% were screen-detected by the CervicalCheck programme (ie picked up at early stage before symptoms present). In 2019, 80% of the 221 patients were alive with no evidence of active disease, 10% in treatment and a similar number deceased. In none of these cases was treatment delayed or diagnosis withheld. 

The assertion of wrongdoing led to multiple reviews of the screening service by outside agencies and bodies, including the Scally Inquiry and an independent review by the Royal College of Obstetricians and Gynaecologists in the UK. These enquiries showed no negligence in the system, and in fact affirmed the Irish programme performed better than other comparable national programmes, including the UK.  

Ireland, in common with many small nations, subcontracts a proportion of its screening to major foreign labs in the USA and elsewhere. A frequently asserted soundbite during the height of coverage is that US-based labs were likely to do sloppier work, as women in the USA are screened annually, with the implication this was subpar for Irish screening.

This assertion, however, is wrong on both counts. Firstly, the labs involved are accredited to international standards which meet or exceed Irish quality, a fact born out by subsequent investigation showing the labs to be performing to standard. Secondly, annual screening is in fact not recommended, even in America, which does not have a national screening programme. In the United States, screening is in fact recommended once every five years for women over 25, which is also the policy of CervicalCheck in Ireland.

For screening to be effective, benefits must be balanced against harms. Retesting outside of screening windows is not helpful, because the actual prevalence of high-grade cervical lesions is relatively rare, typically 2% in a typical population, or 20 in 1000 women.

For an Irish population, smear tests will detect 15 of the real positives, and incorrectly flag 98 women in that 1000 without the disease as having it. All 113 of these women will require an invasive colposcopy to confirm the diagnosis, which aside from being uncomfortable comes with some risks of harm. Of these women, almost 87% are false positives, without the disease.

‍Retesting the entire population unacceptably increases the false positive rate higher still, to diminishing gains. As cervical cancer is slow to evolve, even a missed positive can be caught on the next screening window. Retesting, however, leads to unacceptably high levels of false positives and needless invasive procedures.  

Women facing cancer is an inherently emotive issue, and the high-profile cases of dying woman suing the state elicited sympathy and outrage. It did not help however that reporting was frequently poor – Dr Gabriel Scally commented on the febrile media environment, deeming it “a toxic mix of media, politicians and legal professionals”.

Politicians including Mary Lou MacDonald and Alan Kelly stated or implied that women were dying because of CervicalCheck, despite this being blatantly untrue.