Talking of coloured dots, anyone apart from me remember the old Rail adverts with the red and black dots? “The red dosts are the trains that didn’t run on time. The black dots…”

According to Dr Ray, who is a real radiologist and thus well clued up on these things, AAA screening and surgical repair broadly works out as “cost-viable” (i.e. cost is acceptable by comparison with other stuff NICE accepts as cost acceptable for the NHS) if mortality from AAA repair (by open surgery or stenting) is less than 5%-ish. As the mortality gets higher, the overall cost of screening per life-year saved increases, getting into the area where it is “too expensive to be cost-effective” in NICE terms.

This therefore suggests that the viability of the screening strategy will also depend how hot-shot the UK’s vascular surgeons are (which clearly affects the peri-operative mortality)… and whether there are enough of them.

The last figure I saw for mortality with endovascular graft AAA repair was 2-2.5%… which I guess you could say was safe-ish for a major operation usually on an older person, but still between 1 in 50 and 1 in 40 peri-operative deaths.

A final snag with the graft (keyhole) repairs is that from what I have read all the people who have this need to get an abdominal CT scan yearly thereafter to check for leaks… so that might well add to the queue for the CT scanners, see sideshowjim’s comment above, not to mention giving people a good deal of radiation.

So we’re back to “trade-off of different risks” again.

Gah! All a bit too complicated really… which is why the NICE folk are important, for (at least theoretically) taking these decisions “out of the political arena” … if only it were true.

PS For real tech-y surgery freaks with Athens passwords, more on repairing AAAs by endovascular graft here.

‘Breast Cancer Incidence Trends in Deprived and Affluent Scottish Women’, Brown SBF, Hole D, Cooke TG, Breast Cancer Research and Treatment, 2007, 103:233-238

First 6 references of this paper are other studies showing the trend also. Not open access i’m afraid.

(Guess who had their cardivascular radiography exam last week?).

Another possible problem could be the use of contrast agent (an Iodine based goo injected into the body to help visualise blood vessels on x-ray) and allergic reactions/renal disease. Since the introduction of iso-osmolar agents, the rate of complications has dropped masively, but there’s still gonna be someone out there who doesn’t know they’re allergic to iodine, or develops CIN and ends up on dialasys or worse…

And finally, most CT depts in big hospitals are manic production lines already, so where are the scans going to take place, and what poor sod is going to spend the rest of their life reporting thousands of near-identical C.A.P. scans???

Me, I vote for doppler ultrasound if it’s gonna happen. And aortic aneurysm stenting is a pretty safe procedure (there’s risk in everything, yeah, but I reckon there’s less risk in a stent insertion than there is walking around with a bulging water balloon of an aorta constantly swelling in yer abdo…).

Hope someone else can find the statistics and references for the above, cos at the moment, I’m far too knackered.

That’s interesting; I didn’t know that. Do you have a reference?

(Look at me, asking for a reference like a real scientist and that.)

Interesting thoughts – they certainly don’t call you superburger for nothing!

I can smell ‘bonanza’ in the air: I expect a clever (and good) scientist (like Dr Ben) will discover that a newer drug – a novel variant on the statin theme – will prevent millions of lives being lost to breast cancer – as well as that concrete cardiovascular end point that we all so dread.

I can smell a scam coming on – and I’ll bet Dr Ben can too.

Although it was my understanding that the aetiology of breast cancer is somewhat a can of worms. Interestingly for example it is one of the very few illnesses that don’t obey the usual social class distribution.

It would be an interesting debate if one could show that the money is better spent elsewhere and that those women who present with breast cancer that would have been picked up by screening have more chance of surviving due to the better care, or fewer women get cancer due to improved education / advice.

More recently, Biocard has been promoted as a form of rapid-testing for Coeliac Disease; however, although these tests have been validated, and are comparatively easy for an expert to interpret, it is not necessarily straightforward for the general consumer (see also, Update 2). It is possible for a test to have very high specificity and sensitivity when used by appropriately trained and experienced personnel but to have very different accuracy when used by the general consumer who is (presumably) seeing and using the test for the first time.

I hadn’t realised that a home test could be approved without extensive consumer testing to establish whether or not it has the same reliability as it does when used by trained personnel. What does the MHRA do?

BTW, in case anyone doesn’t know already, the standard line (which appears in all pharmacology textbooks) as to how deadly nightshade got the name Belladonna is as follows.

Deadly nightshade contains a chemical called atropine, which is a blocker of receptors (and hence of nerve signals being received) in some kinds of nerve endings. This includes the nerve endings in the eye controlling the muscles that determine the diameter of your pupils. Deadly Nightshade thus became an early cosmetic, since dropping a tincture of the plant in your eye would widen your pupils, which was supposed to make a woman look more attractive… hence “bella donna”. The scientific name of the plant is Atropa belladonna.

Sorry, I’m getting like the guys in Private Eye’s pedant’s corner… occupational hazard of spending too much time working in a University, I fear.

And belladonna is also the poisonous … deadly nightshade … and not such a pretty flower as you!

let’s be clear. i wrote a piece explaining that screening is not always definitely a good thing, and it often involves close risk benefit analyses; and secondly that the risks and benefits of screening can be misrepresented or brushed over by those running the programs.

i’m always very eager to hear critical discussion and if you disagree it would be great if could say something useful on those points rather than “i dont trust the organ from which your references come”.

the discussions here have generally been very interesting in the past, i think it’s a function of more traffic that they’re going a bit downhill. it’s hard to know what to do about it, but i guess i’ll do something. the trouble is if you spam/delete dreary and distracting abuse from people like “CelticLeopard” then they get all huffy about the fact that you CENSORED their REALLY IMPORTANT point. in fact i see he has already expressed that exciting thought. even re-reading his posts to check that bored me.

“Let’s imagine we start with two groups of 1000 patients (or 10,000, depending on the prevalence – occurrence – of the underlying condition, e.g. AAA). One group will all get screened, the other group won’t. Can you work out how many deaths the stats say there will be in each group (i.e. deaths from AAA w no screening vs. peri-operative deaths if we screen and then operate). So how many lives are saved? And what does it cost roughly per live saved if we assume the following costs?”

And so on.

Makes a useful exercise in stats , maths and reality.

It is a widespread view among scientists who use maths and stats, but who are not mathematicians or statisticians, that you have to put some “real” numbers in equations and calculations (even if they are example / hypothetical numbers) to make the whole thing comprehensible for the “consumer”. So with screening, seeing estimates from a mathematical model is better than no numbers at all, but trying a back-of-the-envelope calculation yourself is better still as it shows you in basic terms how the models will have been constructed.

If you’re not already snorting the belladonna … then you should be.