Margaret McCartney's Blog, page 7

Todays’ ‘news’ on breast screening

mammography screening

This is the paper attracting the headlines.

It would make an ideal paper for students to analyse if they fancied doing it with How to Read a Paper by Trish Greenhalgh.

What we really have to ask is in what way this can be more authoritative than the Cochrane review on the subject.

It’s not an RCT, it relies on case study analysis (see blog post on the hazards of this below) it relies on expert estimation (and remember that expert opinion is far less reliable than RCTs) which is fraught with difficulties.

I have read a couple of media reports and am dismayed at the lack of journalistic critique of what’s been presented.

 

 

 

 

 

 

all BMJ stuff about Atos

For ease, here are all the articles I’ve written for the BMJ about Atos.

Well enough to work

http://www.bmj.com/content/342/bmj.d599

Atos and changes to disabled people’s benefits

http://www.bmj.com/content/344/bmj.e1114

The disturbing truth about disability assessments

http://bmj.com/cgi/content/full/bmj.e5347?ijkey=hCXfT1z84M6BopW&keytype=ref

with blogs on Atos adverts here

http://www.margaretmccartney.com/blog...

and on plans to remove GPs from long term sickness certification

http://www.margaretmccartney.com/blog...

Breast screening: case control vs RCT – the problems

The Daily Telegraph reported yesterday that “Breast cancer screening ‘works and we should move on” – “Women should undergo breast cancer screening because it halves the chance of them dying of the disease, according to a new study that claims to draw a line under the controversy”.

So simple? No!

Here’s the paper. It’s a case control study. So you look back at women who have died of breast cancer, and take a comparable group of women, but who didn’t die of breast cancer. Then you find out if one group went for breast screening more than the other. The researchers found that there was  ”an average 49% reduction in breast cancer mortality for women who are screened.”

I don’t think this means that we have proven whether breast screening works or not. The best way to do that is through a randomised controlled trial. In this type of trial, a comparable group of women are sorted into two random groups, one group is given breast screening, the other are not, and the end results are compared.

I think this is better, mainly because it reduces bias best. Bias – skewing of your results – can happen when, for example, the healthiest people (and least likely to die of breast cancer) attend most for screening; it will look as though screening made them live longer, but in fact, they would have lived longer anyway.

Many case control studies will try and account for this by controlling mathematically for these types of bias. But it may not be easy to do this fairly.

We can see this with this example, which I have taken from Karsten Jorgensen’s work on this. The Malmo trial (in Sweden) contained over 21,000 women in each of the screening and non screening groups, aged over 45, and followed for a mean of 8.8 years. The data were assessed as a RCT, and the difference in mortality between the two groups was negligible. But the same data was then assessed as though it was a look-back, case control study, and found an odds ratio of 0.42.

The Malmö mammographic screening trial assessed as a randomised trial (Janzon and Andersson, 1991)

Invited (n=21088)
Controls (n=21195)




Proportion
50%
50%


Breast cancer deaths
63 (31 were non- participants)
66


Mortality rate
0.299%
0.311%


Relative risk
0.96 (95% CI 0.68–1.35)

 

The Malmö trial assessed using a case-control design Janzon and Andersson, 1991)

Living controls (random sample)
Women dead from breast cancer
Total




Participation in screening


 Yes
229
36
265


 No
71
24
95


Total
300
360



Crude odds ratio
0.46


Adjusted (matching for age) odds ratio
0.42 (95% CI 0.22–0.78)




So in the case control design of the same Mamlo data, it seemed that you were much less likely to die of breast cancer if you had been screening. But we know when we analyse it as an RCT – as the trial was actually run –  that same large benefit does not appear.
I think this is very important because it demonstrates how we can be misled by supposed benefits from case control studies, but the RCT results are much less impressive in terms of a mortality difference.
This kind of problem isn’t new, but we seem to keep doing it. For example, a couple of years ago this study found that women over 80 who were diagnosed with early breast cancer were more likely to have had mammography. Given what we know about overdiagnosis (being diagnosed with a cancer at breast screening that was not going to interfere with your lifespan) or lead time bias (being diagnosed with a breast cancer earlier because of screening, but not in a way that would allow for better treatment) this wouldn’t be surprising.
In fact, the WHO have said  that we should not be using case control studies to assess the effect of screening; “observational studies based on individual screening history, no matter how well designed and conducted, should not be regarded as providing evidence of an effect of screening”.
We seem to be in a bit of a loop when it comes to breast cancer screening reporting. The bottom line is that RCTs are more reliable than case control studies, and it is either brave or daft to claim the difficulties with the evidence are resolved on lower quality data.

The Steeper Clinic and baby helmets

ASA judgement, all my points upheld.

http://www.asa.org.uk/ASA-action/Adjudications/2012/9/RSL-Steeper/SHP_ADJ_154343.aspx

It upsets me to see helmets for deformational plagiocephaly sold at huge cost to parents.

HIV testing and Inside Health

 

Here are some of the references I used for todays’  Inside Health on HIV testing.

One thing I didn’t get a chance to talk about was how prevalence – the number of cases within a community – influences the false positive rate in the people being tested.

The bottom line is that false positives in a low prevalence community are more common than in a high prevalence community.

So, for example, one 4th generation HIV test has a false positive rate of 3 per 1000. So for every 1000 HIV negative people, 3 will test as positive.

 

If you are running the test in a low prevalence area, where the rate – as in the UK – is 0.1%, 1 in 1000, this means that in 1000 people, 4 will test positive – 3 will be false positives, and one a true positive. So if you are in a low prevalence community and test positive, you only have a 1 in 4 chance of being a true positive.

 

If you run the test in a high prevalence area, say for example the rate of HIV infection in men who have sex with men in London is usually given at 1 in 10, it’s quite different. The false positive rate of the test is still 3 in 1000, but there will be 100 true positives in this group. So there will be 103 positive tests altogether. The chance of a false positive test is just under 3%, at 3/103. So 100 positive tests in this group will be true positives, 3 will be false positives.

 

This sort of knowledge I think very important. Here are some of the references I used.

 

 

http://www.aidsmap.com/HIV-home-tests-how-will-they-be-used/page/2421108/

http://i-base.info/guides/testing/test-accuracy-results-and-further-testing

http://www.nat.org.uk/media/Files/Policy/2011/Oct-2011-Facts-Types-of-HIV-test.pdf

http://www.bashh.org/documents/1176/1176.pdf

http://www.aidsmap.com/Rapid-tests/page/1323371/

http://www.hpa.org.uk/NewsCentre/NationalPressReleases/2011PressReleases/110606HIV/

http://www.hpa.org.uk/hpr/archives/2012/news1612.htm

http://www.hpa.org.uk/webw/HPAweb&HPAwebStandard/HPAweb_C/1317134411572

http://www.hpa.org.uk/hpr/archives/2012/hpr3012.pdf

 

http://www.hpa.org.uk/webc/HPAwebFile/HPAweb_C/1317131685847

The Paralympics and Atos.

Atos are one of the sponsors of the Paralympics.

I struggle to know why. Atos have the multimillion pound contract from the Department of Work and Pensions to assess people who are sick. There are multiple problems with their assessments, and they have never published the evidence, or audit standards for the work they do. They refer to people as ‘clients’, not patients. Their assessments for fitness to work includes things like pressing a button or ensuring that someone is not mentally unwell by not shaking or trembling. Nor do the assessments reflect someone with chronic disease; they are especially poor at assessing people with significant and severe mental illnesses. It’s not just me who thinks that. The BMA has called, via their Local Medical Committees, to scrap the assessments that are currently carried out. The National Audit Office also thinks that Atos are underperfoming – and indeed, Atos assessments are often wrong. A third of appeals are successful – even more if the person has advice from Citizens’ Advice - which is worrying, because it means that some people are likely to be further disadvantaged when not supported at appeal.

The human cost of this is huge. People are worried, anxious, uncertain, and the process is not transparent. GPs are often asked to write letters in support of patients – meaning that many people are being charged for reports – which should surely be paid for by Atos, as it is plain that they cannot assimilate the necessary evidence as they currently operate.

And yet Atos are sponsoring the Paralympics.

On their website they say “The Paralympic Games are elite sport events for athletes from different disability groups. However, they clearly emphasize the participants’ athletic achievements, not their disability.”

I’m worried about Atos view of ‘disability’. I’m worried that they somehow think that all people with disabilities somehow have an  elite athlete inside them. The truth is that some people are physically different and it is society that disables them; there are some disabled people who are phenomenal athletes, and there are some people who are so sick or disabled that they can’t work. Atos has made many disabled peoples’ lives miserable, and it would have been far better had they put their sponsorship money into making their system fairer. If there is a legacy from the Paralympics, I hope a small part of that will be considering how fairly we treat sick people when they are most vulnerable – currently, Atos is failing us all.

Overdiagnosis – Radio 4 Inside Health

Inside Health, all about overdiagnosis. Hope you like

http://www.bbc.co.uk/programmes/b01m0pq6

Here are some of the references I used :

High blood pressure

http://www.ncbi.nlm.nih.gov/pubmedhealth/PMH0012720/

 

http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD006742.pub2/abstract

 

http://www.hypertensionfoundation.org/PDInfo/JC5790-Moser.pdf

 

http://assign-score.com/estimate-the-risk/simd/

 

http://qintervention.org/

 

Pre diabetes

 

http://summaries.cochrane.org/CD005270/long-term-non-pharmacological-weight-loss-interventions-for-adults-with-prediabetes

 

http://www.bmj.com/content/344/bmj.e3564

 

http://www2.wpro.who.int/wpdd/downloads/diabetes_book.pdf

 

http://whqlibdoc.who.int/publications/2006/9241594934_eng.pdf

 

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1123928/

 

http://jama.jamanetwork.com/article.aspx?articleid=193772

 

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1123928/

 

http://www.eurekalert.org/pub_releases/2012-08/aha-rie081612.php

 

http://www.diabetesresearchclinicalpractice.com/article/S0168-8227(07)00309-9/abstract

 

http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(12)60525-X/fulltext

 

Osteopenia

 

http://www.thennt.com/bisphosphonates-for-fracture-prevention-in-post-menopausal-women-without-prior-fractures/

 

http://content.healthaffairs.org/content/26/6/1702.full

 

http://www.bmj.com/content/336/7636/126.full

 

http://www.mhra.gov.uk/Safetyinformation/Generalsafetyinformationandadvice/Product-specificinformationandadvice/Product-specificinformationandadvice-A-F/Bisphosphonates/index.htm

GPs and insurance survey

If any GPs would be free to visit this survey page;

http://www.zoomerang.com/Survey/WEB22GH2CDF7BU

I’d much appreciate your views on the insurance industry and paperwork.

I’m giving a talk to a lot of people in the medical insurance industry at the end of the month and it would be useful to have your views

 

(the fee will go to Mary’s Meals.)

Over diagnosis and CKD

This is a hugely important subject – overdiagnosis. Tonight’s Inside Health is all about CKD, chronic kidney disease. Basically the measurements we use to measure kidney function changed in 2006 and millions of people have been given a ‘disease’ label – and the vast majority will not benefit from this.

Here are some of the references I used. There is also a bit in The Patient Paradox about CKD.

 

http://jasn.asnjournals.org/content/19/5/844.full

http://www.kdigo.org/meetings_events/pdf/KDOQI-KDIGO_Editorial_on_CKD_Classification.pdf

http://www.renalandurologynews.com/ckd-staging-criteria-debated/article/119553/

http://www.bmj.com/content/333/7577/1047

http://www.nhsemployers.org/SiteCollectionDocuments/Chronic_Kidney_Disease_FAQs_cd_110209.pdf

http://jech.bmj.com/content/65/Suppl_2/A12.1.abstract

http://www.ncbi.nlm.nih.gov/pubmed/18092634

http://www.ncbi.nlm.nih.gov/pubmed/18083723

http://www.sciencedirect.com/science/article/pii/S0140673611601785

http://www.nhsemployers.org/SiteCollectionDocuments/Chronic_kidney_disease_FAQs%20-%20ja040711.pdf

http://www.renalreg.com/Report-Area/Report%202010/Chap01_Renal10_web.pdf

 

The ongoing Atos farce

piece in BMJ, here.