Hard to believe that this appeared in a Med Journal

[Guest guest]

BMJ 2003;326 ( 26 April )

 

 

Editor's choice

 

 

 

The screening industry

 

 

 

Richard Smith, editor.

 

Have you thought of buying your significant other a whole body scan? It's

the most fashionable of gifts and perhaps particularly suitable for a 50th,

60th, or even 40th birthday. You might be giving your loved one the supreme

gift of extra years of life. Unfortunately, you may be more likely to give

him or her a lorry load of anxiety and a series of invasive, painful, and

unnecessary investigations.

 

Whole body scanning is currently being intensely marketed in the United

States (p 894) and the enthusiasm will surely spread. The " sell " is simple.

You might have something horrible lurking in your body. The scan will show

it and allow early treatment. Or the scan will give you the all clear,

providing the perfect excuse for an expensive dinner.

 

The problems lie in medicine's difficulties in defining normality, the

devil of " false positives, " and our limited understanding of the natural

history of disease. The commonest way of defining normal is that the

measure lies within two standard deviations of the mean. So in a set of

measurements from a normal population 5% will be classed as " abnormal. " The

whole body scan will produce hundreds of measurements. So you have almost

no chance of emerging as normal, but which of your abnormalities signify

serious disease?

 

Stephen Swensen describes his experiences of using computed tomography to

screen for lung cancer[---]within the context of a major trial. His team

has found 700 ancillary findings within its cohort, but most were false

positives and led to " adversely affected quality of life " and " unnecessary

diagnostic and interventional procedures. "

 

A E Raffle and others analyse the outcomes of a form of screening that we

understand much better, screening to prevent cervical cancer (p 904). They

show that 1000 women have to be screened for 35 years to prevent one death.

This means that one nurse performing 200 tests a year would prevent one

death in 38 years. " During this time she or he would care for over 152

women with abnormal results, over 79 women would be referred for

investigation, [and] over 53 would have abnormal biopsy results. " During

this time one woman would die of cervical cancer despite being screened.

These authors also point out that over 80% of cases of high grade

dyskaryosis and of high grade dysplasia do not progress to invasive cancer.

The same may well apply in other organs, and the prophylactic removal of

colons, ovaries, breasts, and gullets may be killing people without

benefit.

 

Simple minded enthusiasm for screening[---]combined with the industrial

opportunity to make fat profits[---]may mean that soon none of us will be

normal. We will be screened not only for cancers of the cervix and breast

but also of the prostate, colon, ovary, lung, stomach, and so on. It's

always hard to put the case for " not knowing, " but economists[---]cold

hearted beasts that they are[---]have a wonderful notion of " rational

ignorance. " It simply isn't sensible to try to know everything. Ignorance

can be bliss.

 

 

 

 

 

Footnotes

 

To receive Editor's choice by email each week via our website:

bmj.com/cgi/customalert

 

 

______________________________

© 2003 BMJ Publishing Group Ltd

 

 

 

 

 

To from or edit your subscriptions to any bmj.com Customised

Alert, point your browser at http://bmj.com/cgi/customalert .

 

 

------

© 2003 BMJ Publishing Group Ltd

 

---

Incoming mail is certified Virus Free.

Checked by AVG anti-virus system (http://www.grisoft.com).

Version: 6.0.473 / Virus Database: 271 - Release 4/17/03

---

Outgoing mail is certified Virus Free.

Checked by AVG anti-virus system (http://www.grisoft.com).

Version: 6.0.476 / Virus Database: 273 - Release 4/24/03