Examination of Witness (Questions 420-422)|
Professor Sir Richard Peto
14 FEBRUARY 2006
Q420 Lord Macdonald of Tradeston: In
order to inform decisions about the allocation of limited resources,
the Department of Health and NICE employed "quality adjusted
life years". In doing that, the gain to a 75-year
old is treated the same as the benefit to, say, a 20-year old:
giving an extra year to a patient in very poor health is seen
as yielding the same benefit as giving an extra year to a person
with normal health. Does that seem logical and appropriate?
Professor Sir Richard Peto: That is exactly
what it avoids doing. The use of quality adjusted life years tries
to give years different values, depending on how healthy the person
is. If a person is not in good health, then a year of life not
in good health is given somewhat less value than a year in good
health. There is a discount, in that benefit in the distant future
gets weighted somewhat less than benefit in the immediate future.
That is arguable: you get much the same conclusions, however,
with or without discounting. I think it is reasonable to try to
say, "How many extra years are you gaining by this treatment,
and, of those years, how many will be of a reasonable quality
of life?". The use of quality adjusted life years tries to
avoid exactly the problem that you raised. Avoiding the death
of a 20-year old not only gains a lot more years of life than
avoiding the death of an 80-year old, but in general those years
will tend to be of better quality, human life being what it is.
Attempting to quantify this is a sensible way of trying to proceed.
Q421 Lord Skidelsky: As a supplementary,
one does not want to carry that to any extreme, like weighting
IQ, for example? I just wonder what the measures of quality are
or are they mainly physical attributes and some alertness?
Professor Sir Richard Peto: You can do the calculations
with and without allowance for whether people are desperately
miserable. If you just try to look overall, either globally or
in particular countries, then in general the treatments that avoid
death are those that produce big gains in terms of quality adjusted
life years. The one mis-match is that if you do not put in some
measure of the quality of life, then you underrate the relevance
of musculo-skeletal disease and you underrate the relevance of
various mental conditions. Roughly speaking perspectives based
on what is most important in terms of quality adjusted life years,
either globally or in particular countries, come out roughly like
perspectives based on what is most important in terms of prevention
of premature death. These two approaches match pretty well the
things to which you would give priority. There are two things
that do come out differently, however: mental illness gets seen
as very much more important if you allow quality and also musculo-skeletal
diseases get made relatively very much more importantalthough,
actually, mental illness does cause quite a number of deaths worldwide.
There are about one million deaths from suicide each year. In
this country, suicide deaths outnumber traffic deaths. Suicide
is a somewhat preventable cause of death: there are things one
will do to reduce the likelihood of people dying by suicide. Returning
to the main question, you can assess cost-effectiveness in various
ways. It is surprising how robust the conclusions are. The approximate
conclusion is that, apart from musculo-skeletal disease and mental
illness, the things that are cost-effective for avoiding premature
death are in general, much the same as the things that are cost-effective
for gaining quality adjusted life years. Any such measure is imperfect.
It has to depend on human judgment about the things you really
value and the things you do not. But, use of quality adjusted
life years does not seem to produce conclusions that seem to me
widely wrong, in general, about the relative importance of things.
Q422 Chairman: I think we ought to bring
it to a close there. On behalf of the Committee, may I thank you
very much indeed for coming along, and, if I may say so, setting
a good example by indicating there was some degree of uncertainty
in some of the things we are looking into. That is helpful to
us in our general inquiry. We are very grateful to you for what
you have said and the way in which you answered the questions.
Thank you very much indeed.
Professor Sir Richard Peto: Thank you for the