Diabetes Screening

[finneyb]

BMJ Reasearch article http://www.bmj.com/content/356/bmj.i6538

Conclusion 3rd sentence ' As screening is inaccurate, many people will receives an incorrect diagnosis and be referred on for interventions while others will be falsely reassured and not offered the intervention. '


No surprise there IMO - the body is simply not black and white it is a highly complex interactive process so a simple test cannot predict - I suggest that even genomics will not predict

BBC Radio 4 Inside Health last week looked at Prostate Cancer it demonstrated that we are again working in the dark http://www.bbc.co.uk/programmes/b086s7jr This week it is treatment including a new one that looks to be an improvement

Brian

 

[bugbear]

BMJ Reasearch article http://www.bmj.com/content/356/bmj.i6538

Conclusion 3rd sentence ' As screening is inaccurate, many people will receives an incorrect diagnosis and be referred on for interventions while others will be falsely reassured and not offered the intervention. '


No surprise there IMO - the body is simply not black and white it is a highly complex interactive process so a simple test cannot predict - I suggest that even genomics will not predict

Even if the test is good (for some meaning of "good") there's this to contend with, before deciding to screen the nation:

https://en.wikipedia.org/wiki/False_positive_paradox

One might find it more than a little disturbing to know that even some doctors don't understand this;

https://www.sciencenews.org/blog/contex ... -test-math

I guess we can't expect doctors and surgeons to be mathematicians too.

BugBear

 

[Phil Pascoe]

A huge number of them will have A level maths ...

 

[backinthesaddle]

I would imagine they all would!

 

[Kalimna]

Theres quite a difference between a-level maths completed when 18, and the application of that when a specialist doctor age 30+.
The pros and cons of health screening is a huge and complex area, and not one that benefits from personal anecdote. When you are dealing with many thousands/millions of people the vast majority will adhere to the norm, but there will always be false positives and negatives. The real trick is balancing the risk and benefit of *any* intervention.

Adam

 

[Blockplane]

I would imagine they all would!

As medical schools generally insist on 3 sciences at A level for admission, I doubt that many doctors do have A level maths

 

[Phil Pascoe]

I don't know, I suspect physics/chemistry/maths is probably more common than physics/chemistry/biology. It was in my day, as we were discouraged from taking biology - it was seen as being second rate.

 

[Keithie]

If there was a gold medal for spouting utter twoddle about diabetes then the nhs would win it.

The NHS is Uk. Uk is a small part of the world. The World Health Organisation says this about diabetes :

http://www.who.int/mediacentre/factsheets/fs312/en/

Why does the nhs not pay attention to globally accepted definitions? Money ?

(did you or your doctor know there is type 3 diabetes?)

 

[finneyb]

If there was a gold medal for spouting utter twoddle about diabetes then the nhs would win it.

The NHS is Uk. Uk is a small part of the world. The World Health Organisation says this about diabetes :

http://www.who.int/mediacentre/factsheets/fs312/en/

Why does the nhs not pay attention to globally accepted definitions? Money ?

(did you or your doctor know there is type 3 diabetes?)

The NHS's role is to treat people efficiently and effectively. The two organisations cannot be compared they have different roles. The WHO are looking to the future with all its uncertainty; the NHS is dealing with real life.

In the case of diabetes the WHO is claiming that there will be a diabetes epidemic - they are right to high light the possibility but have no more knowledge of the future than you or I. Predictions of the numbers with HIV are now shown to be overly pessimistic and are lower than predictions.

If you repeatedly lower the diagnosis criteria, as has happened, the prediction becomes self-fulfilling - there are no evidence based criteria for the onset of Type 2 diabetes, unlike Type 1, the targets are just guesses - and there is financial gain (not NHS) to have the targets lower eg pharmco.

The concept that treatment is good is flawed - the treatment is only good if its essential treatment, otherwise it is over treatment will all its associated costs and risks to the patient; all treatment carries risk even the essential treatment.

Brian

 

[Keithie]

Hi Brian... I think we agree on the point you are making re: shifting of criteria and over treating...both typically a bad idea.

My point wasnt very clear perhaps....I meant that WHO asserts that ongoing microvacular damage constitutes diabetes whereas NHS seems to define inrelation to blood glucose levels.

So, for example, a simple urine dipstick test can look for the presence of microalbuminuria and potentially find it long before there's glucosuria or elevated blood glucose levels. Some pancreatic conditions (type 3) can also be detected before NHS glucose testing would reveal type 2 or type1.

Early detection is key to informing whether there is a need for treatment. NHS dont, as far as I'm aware, look for microvascular damage as such...they seem to wait till things have progressed further. This means that in some cases patients may not be informed as far in advance of a looming problem as they could be ..and as a result dont have as much time to modify diet/lifestyle as they possibly could. Whether all patients would listen to and act on the adviceis of course matter!

Obviously in practice things are far more complex than this with possible false positves, human error, folk not necessarily wanting their medical notes to show diabetes, costs of regular finger prick testing, the way gp teams are funded etc etc...but it still seems to me that if a possible problem can be identified earlier at low cost (urine dipstick) then that might bea good thing overall.

 

[finneyb]

but it still seems to me that if a possible problem can be identified earlier at low cost (urine dipstick) then that might bea good thing overall.

I agree, but as the research shows we can't predict with any certainty - eg breast screening 9 out of 10 patients initially shown as a having a problem will on further investigation show to be clear. And who is not to say that the one that is breast cancer would not be found by the symptoms. I know of three women in their 40s with breast cancer all picked up by symptoms.

A bad screening result, that is later proved to be incorrect causes a lot of unnecessary psychological damage and produces no benefit. There are times when it is best not to open the can of worms if you can't control the worms when they are out. What this does is lead to the over medicalisation of the population and lead them to believe that the NHS has their health risk covered - it hasn't and it can't, although some in Public Health England with their screening programmes would have us believe that they can foretell the individual's future.

Brian

 

[Jacob]

BMJ Reasearch article http://www.bmj.com/content/356/bmj.i6538

Conclusion 3rd sentence ' As screening is inaccurate, many people will receives an incorrect diagnosis and be referred on for interventions while others will be falsely reassured and not offered the intervention. '


No surprise there IMO - the body is simply not black and white it is a highly complex interactive process so a simple test cannot predict - I suggest that even genomics will not predict

Even if the test is good (for some meaning of "good") there's this to contend with, before deciding to screen the nation:

https://en.wikipedia.org/wiki/False_positive_paradox

One might find it more than a little disturbing to know that even some doctors don't understand this;

https://www.sciencenews.org/blog/contex ... -test-math

I guess we can't expect doctors and surgeons to be mathematicians too.

BugBear

I'm not bad at maths but I couldn't work it out as it's a stats question and I haven't done much along those lines. So it's not about maths ability in general it's about learning a specific area of maths.

Re screening; it's not possible to know if screening is effective unless it is done and the results analysed over time. So it's no surprise that some screening is eventually shown to be ineffective and vice versa - some is very effective. It's a continuous process of review.

 

[bugbear]

Sadly, since the https://en.wikipedia.org/wiki/False_positive_paradox is not obvious, people cleave to what is obvious (but wrong) - the idea that screening can only be good.

This prejudice is exploited by companies to sell screening services.

http://www.lifelinescreening.com/

Parasitic vermin preying on people worries IMHO.

BugBear

PS The "worked example" is fairly easy to follow

 

[finneyb]

Re screening; it's not possible to know if screening is effective unless it is done and the results analysed over time. So it's no surprise that some screening is eventually shown to be ineffective and vice versa - some is very effective. It's a continuous process of review.

Agreed, but the problem is that if the NHS try and take the screening that has been shown to be ineffective away it will be seen as a cut in services - I can see the Daily Mail headline now.

Brian