A Doctor Writes

at | 21 Replies

Further to the ongoing CervicalCheck scandal.

Dr Primum Non Nocere  writes:

I wanted to add my thoughts on the CervicalCheck scandal from the perspective of a doctor working outside of Ireland.

Without wishing to state the obvious I think that the handing of the cervical cancer audit is very poor by both CervicalCheck and the Government. The Government is doing damage limitation but I’m not sure they really understand the following:

Medicine is inexact. Smear tests are a screening programme. There will be misses.

CervicalCheck are at fault for the following reasons.

Fault 1: They should have told women there was a miss rate. Informed consent is important. Also, knowledge of the miss rate is a crucial part of the cervical cancer prevention process.

Knowledge of the miss rate stops women from ignoring symptoms in the belief that nothing can be wrong as their last smear test raised no issues. It also gives women the option of opting for more frequent smear tests, free elective colposcopy or HPV testing as a back-up.

Fault 2:
In addition to informing women that there was a miss rate, Cervical Check should have been able to provide these women with an accurate miss rate.

The CervicalCheck website currently states:

“A cervical screening test is not a diagnostic test. As with all screening tests, cervical screening may not always be 100% accurate. There is a small risk that cell changes will not be picked up in a test. However, any cell changes will usually be picked up in future tests. This is why it is important to have regular cervical screening tests.”

This doesn’t cut it in terms of information which should be provided to women. The words “usually be picked up” do not reflect the normal miss rate for cervical smear tests worldwide, which is significantly higher than this. No wonder these women weren’t concerned.

Fault 3: CervicalCheck wasn’t honest. If they were going to do an audit, they should have decided in advance what they would do with the outcome.

If they didn’t know, they could have looked to their colleagues in the United Kingdom. When an audit into cervical screening practices in Leicestershire disclosed that fourteen patients had died after missed smear tests, the NHS held a press conference, and published the results. Publication of cervical smear test audits disclosing false negative smears is standard practice in the UK.

Not every country can afford yearly screening, HPV testing or free elective coloscopies. It is a cost-benefit issue.

However informing women that medical tests are not infallible, that there is a miss rate for smear tests and providing them with an accurate miss rate prevents them from disregarding symptoms and gives them the information that they need in order to decide whether they should voluntarily opt for additional testing.

It is a crucial part of the screening process and its goal of saving lives.

Dr Primum Non Nocere is a doctor qualified in Ireland, currently practising abroad.

Yesterday: When Sorry Rings Hollow

21 thoughts on “A Doctor Writes

    1. Anon

      But but but privatization is the answer to everything!
      That’s unfair, dont let Harney hog all the credit, surely we can blame Bertie too?

      Seriously if I recall the Irish lab had an accuracy rate of 80% and the cheaper alternative had a rate of 70%, there were always going to be a lot of misses. Cancer is awful.
      It was stupid and misleading to use such casual informal language as “screening may not always be 100% accurate” when it is known to be significantly less accurate than that.

      Reply
      1. Cian

        “Seriously if I recall the Irish lab had an accuracy rate of 80% and the cheaper alternative had a rate of 70%, there were always going to be a lot of misses.”
        source please?

        There has been a *lot* of misinformation over the last 3 weeks. I was talking to my folks over the weekend and they were convinced that Cervical Check withheld information from women who though they were healthy.

        Reply
  1. b

    Good article.

    “Government is doing damage limitation but I’m not sure they really understand the following:
    Medicine is inexact. Smear tests are a screening programme. There will be misses.”

    There is truth in that but it must be acknowledged that many including Varadkar and Harris have been pointing this out while many in the opposition have been making statements about ‘missed cancer diagnoses’ and needlessly scaremongering

    Reply
    1. frankie

      This government has shifted the outrage to something that it was not about
      A cunning plan to shift accountability
      Yes smear tests are not a proper test they do have human error
      This is not about smear testing
      Its about the HSE who decided not to inform victims
      Its about cover up and due to this women were not informed
      They were treated with utter contempt
      Heads must roll in both the department of health and the HSE
      The HSE have shown time and time again they are a rogue department and are not fit for purpose

      Reply
  2. Cian

    As resident defender of the defenceless I will start the ball rolling on this one.
    Firstly, this is a good clear article, and I applaud the tone and the overall direction. Secondly, if anyone hasn’t read the newspaper article attached – I strongly recommend you do so.

    Fault 1: They should have told women there was a miss rate. this is immediately contradicted by Fault 2. So Fault 1

    Fault 2: In addition to informing women that there was a miss rate, Cervical Check should have been able to provide these women with an accurate miss rate. I wholeheartedly agree with this. But when I went to the UK equivalent they say the same as Ireland “But cervical screening isn’t 100% accurate and doesn’t prevent all cases of cervical cancer.” and as a disadvantage “[there is] a very small chance of getting incorrect results, which could lead to abnormalities being missed or unnecessary distress and treatment ”
    What’s more, in their detailed PDF this isn’t mentioned at all.
    https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/598278/cervical_screening_leaflet.pdf

    Do any countries publish there false-negative rate? Is this a fault with all screening services worldwide?

    Fault 3: CervicalCheck wasn’t honest. If they were going to do an audit, they should have decided in advance what they would do with the outcome. yeeeees, and no. When they were set up they had decided what to do when they got a positive test: Retest previous samples, and use this information to improve the training to improve the service. Oh, and not to tell the woman the earlier test was wrong (she would have just been informed of the 2nd positive test).
    But then in 2015 the HSE pushed voluntary Open Disclosure across all areas. This changed the way CervicalCheck worked. This is what triggered the compilation of all their false negatives (and is likely the reason the infamous memo was created), and what led to Cervical Check informing the women’s doctors of the mistakes.

    So, I think, Fault 3 should either be “When Cervical Check was set up it should have started as an open disclosure entity” or “When open disclosure was mandated in 2015 – they should have been completely open and immediately told the women”.

    Reply
  3. Alors

    Mm. But a doctor on the radio was saying that when the tests were done in Ireland, if a clear test came in & the doc was a bit worried, the doc could call up & say “Would you just mind taking another little look to be sure to be sure?” But when they moved to America, this option was no longer possible. There are times (many, many times) when local is advantageous.

    Reply
    1. Pip

      Repeating myself but we mustn’t let the clear, strong and prophetic words of Dr David Gibbons (as reported on BS) be forgotten. Very few, if indeed any, of the recent interviews, attacks and analyses all over the media address the simple points he made way back about the trouble coming down the line.

      Reply
  4. Fact Checker

    This is an excellent post. Every diagnostic test runs the risk of false positives and false negatives.

    Generally false negatives are more problematic. Although false positives are bad too as it can lead to interventions where none are necessary.

    There is often a trade-off between minimizing false positives or false negatives, all depending on how you calibrate the test. Most patients don’t realise this.

    Unfortunately some doctors are not always the best at understanding and communicating statistics: https://www.bmj.com/content/349/bmj.g5619.full

    Reply
    1. Cian

      The US is moving from an annual test to a 3-year test because of all the false positives. If you have 35-annual tests – then over their lives 70% of women would be (incorrectly) told they had abnormalities! This leads to stress for the woman and additional testing that is not required.

      The US Centre for Disease Control and Prevention recommends a 3-year gap for screening. https://www.cdc.gov/cancer/cervical/pdf/guidelines.pdf

      Reply
      1. Fact Checker

        Not correct. The chance of receiving a false positive is (at least somewhat) independent of having received one before.

        Assuming full independence and a 2% false positive rate, over the first five years 98 women in 1,000 will get one false positive result, and two women will get two false positive results.

        Over 35 years you will have some women who receive several false positives, so somewhat less than 70% will receive one at all.

        Like I said – statistics is very hard to communicate!

        Reply
        1. Cian

          I don’t know the false-positive rate – the site stated the 70% figure.

          But, lets crunch the numbers:
          If 2 in 100 giving false positive over 35 years you get 507 of the 1,000 women have at least one false positive. 1-(0.98^35)
          If it is 3-in-100 that number jumps to 656 of the 1,000.
          And at 4-in-100 then 760 of the 1000 women will get (at least one) false positives over the 35 years.

          Reply
      2. Nuala Mc Namara

        In Ireland cervical screening is every 3yrs while in US it was every year re standard screening if just standard screening done.CPL labs normal procedure for US women was imaging,HPV testing of smears as well as standard screening but were only contracted to do standard cervical screening for women in Ireland.
        In Australia they have a 5year gap because they use HPV screening of smears also.

        Reply
  5. GahBlahBlah

    I understand that smear tests are only 70% effective at detecting cancerous or precancerous cells, but I do not think that the case with these tests were that they were false negatives. I have heard that rechecking of the slides in question found that the slides were indeed positive for precancerous or cancerous cells, so the issue here was in fact that there was human error in not recognising these cell classifications. This is my understanding of the case and perhaps I have been misinformed, but I feel that perhaps the doctor has not read into the case enough and is letting the HSE off too lightly here.

    Reply
    1. b

      as is my understanding, the tests are not definitive and it relies on humans looking through a microscope to identify certain traits of cells, so it is down to human error and a false negative – some are difficult to spot and will be missed and some should have been spotted within reason. In vicky phelan’s case the lab was negligent over an above what could be explained by human error

      the audit gives high false negative results because when you are rechecking samples where you know cancer has developed your are more likely to spot the anomaly (i’m not sure what information the re-checker is given here, though i’m guessing if you’re asked to review a sample from a number a years ago, you’ve a better chance of spotting what you were looking for)

      Reply
    2. Cian

      In the Vicky Phelan case: US Labs settled for millions. We can ‘guess’ that this means that there was human error and her slide was very obviously positive, and this was missed, so they were liable.

      But we can’t say anything about any of the other cases because we simply do not know. It may have been the 30% you mention are missed. It may have been human error (misread slide, data entry error, mislabelled slide, computer error, …). It may have been a lack of audit/oversight by Clerical Check (i.e. they should be doing immediate re-tests on random samples to ensure the results are consistent, they should send know positive slides and check they are caught, anyone in training should be monitored, all samples should be checked independently by two people, or whatever), It may have been a failure in Cervical Check to ensure the tender to external parties was sufficiently robust/high enough standard. Cervical Check didn’t have sufficient risk mitigation for this.

      But until the report is done we can should only speculate.

      Reply

Leave a Reply

Your email address will not be published. Required fields are marked *