Comentary
David Ogilvie explains the epidemiological terms used in this paper
On the face of it, this paper reports the straightforward
findings of a survey of current practice in nephrology
and transplant units. However, the paper also touches
on some wider issues in epidemiology and public
health. These include how you define the amount of a
given disease in a population, and how you decide
whether or not to screen people for it.
How big a problem is skin cancer in
people with renal transplants?
Epidemiologists have two main ways of answering this
question for a given condition. These are easily
confused.
The first measure is the incidence of the condition.
This is defined as the proportion of people who
develop the condition during a specified period of
time—usually one year. The authors tell us that
between 7.1% and 10.6% of people with renal
transplants develop skin cancer each year.
The second measure is the prevalence of the
condition. This is defined as the proportion of people
who have the condition at any given time. The authors
tell us that 16.5% of people with renal transplants have
skin cancer at any time.
Incidence and prevalence give us different infor?
mation. You might find it helpful to think of a
swimming pool where being in the water represents
having cancer. Incidence tells us how often people are
entering the water (developing cancer), so a high inci?
dence might imply that people are more vulnerable to
whatever is pushing them in (ultraviolet light).
Prevalence, on the other hand, tells us how many
people are swimming around in the water (have
cancer). This will depend not only on how often they
are being pushed in, but also on how long they stay in
the water before getting out (recovering) or dying.
Another point the authors make is that people with
renal transplants have a sevenfold increase in mortality
from skin cancer, although absolute death rates are low.
What would you make of this information if you had a
renal transplant? The sevenfold increase, which would
be referred to as the relative risk, sounds important.
However, before you had your transplant, your actual
chance of dying from skin cancer (the absolute risk) was
small. It would be more worrying to increase your risk of
a common condition (say, dying from a heart attack) by
seven times than to increase your risk of a rarer
condition (say, dying of skin cancer) by seven times.
Screening for skin cancer
However, this is not to say that skin cancer is not
important—the incidence and prevalence figures show
that it is common in this group of people. The authors
tell us that skin cancer surveillance is recommended
for renal transplant patients. But why? We do not
inspect everyone's skin every year for tumours.
The question of whether or not to screen for a con?
dition is a key public health conundrum. It is important
to recognise that just because you can screen for some?
thing does not necessarily mean that it is a good idea:
read about prostate cancer screening for a topical exam?
ple. Wilson and Jungner made the classic statement of
the principles for screening; the National Screening
Committee now offers a more contemporary
checklist.1 2
Some obvious points in favour of screening
for skin cancer in renal transplant patients might be:
- It is common
- Screening is painless and doesn't expose the person
to danger
- The patients are attending the clinic anyway
- If you find a tumour you can do something about it.
Might there be any problems? Look up the check?
list and see what you think.2
What about the survey?
The most obvious finding of the survey is that most
units do not screen their patients for skin cancer. But
even in units which do carry out screening, it is notice?
able that it tends to be carried out by people who have
not been trained to do it. The authors call for better
training, which we assume would lead to more accurate
screening.
Another point from the survey is that some
patients may not be getting enough advice about the
risk of skin cancer. This is important because if you
know the risks you can limit your exposure to the sun
and aim to avoid getting cancer in the first place
(primary prevention), rather than waiting to have your
cancer detected by screening (secondary prevention).
Screening may be useful, but there are usually other
things we should do as well.
David Ogilvie, specialist registrar in public health medicine, Hamilton, Lanarkshire
Email: david.ogilvie@lanarkshirehb.scot.nhs.uk
studentBMJ 2001;09:399-442 November ISSN 0966-6494
- Wilson JMG, Jungner G. Principles and practice of screening for disease.
Geneva: World Health Organization, 1968.
- National Screening Committee. Criteria for appraising the viability, effec?
tiveness and appropriateness of a screening programme. http://www.doh.gov.uk/nsc/pdfs/criteria.pdf (accessed 18 September 2001)
