Examination of Witnesses (Questions 115-163)
MIKE KELLY, ALAN BRENNAN, RICHARD ALLSOP AND DR HAMISH
MELDRUM
12 OCTOBER 2010
Q115 Chair:
Good morning, gentlemen. Welcome to the Transport Select Committee.
Could I ask you to identify yourselves, please, for our records?
Dr Hamish Meldrum:
I am Dr Hamish Meldrum. I am Chairman of Council at the British
Medical Association.
Professor Richard Allsop:
I am Richard Allsop. I am Emeritus Professor of Transport Studies
at University College London. I am a director of PACTS and of
the European Transport Safety Council.
Professor Alan Brennan:
I am Alan Brennan, Professor of Health Economics and Decision
Modelling at the University of Sheffield.
Professor Mike Kelly:
I am Professor Mike Kelly. I am the Director of the Centre for
Public Health Excellence at NICE.
Q116 Chair:
Thank you very much. The evidence that we have received from Professor
Allsop and from NICE shows very different estimates concerning
the number of casualties that would be reduced with a 50mg limit
on alcohol. What is the reason for this discrepancy, if it is
indeed a discrepancy? Professor Allsop, would you like to give
me your views? Why are your figures so different?
Professor Richard Allsop:
I think we were doing different things. First of all, my approach
was deliberately cautious. My approach was concerned simply with
the reduction of the limit without any consideration of effect
that associated increased enforcement and associated new public
information might have on people who are already way above the
existing limit, but I always recognised in what I wrote that there
would be additional savings at those higher levels. I was concerned
with the changes coming from the reduction itself. I would also
say that my approach is to be rather careful about the interpretation
of evidence from other countries in relation to our over-80mg
people - people breaking the existing limit - because we have
been so effective against those offenders now, and we have reduced
to about a fifth now of where we started the amount of offending
at that level, whereas some of the other countries that were reducing
their limits had been less tough on the people above the existing
limit.
Q117 Chair:
So is the issue then to do with comparable information on what's
happening in this country and what's happened elsewhere: what
the impact would be?
Professor Richard Allsop:
That is a question, I think, for my colleagues.
Professor Alan Brennan:
I think Professor Allsop's estimates were essentially making assumptions
about what would happen in this country to various different bands
of people. You have a set of people who are currently driving
just above the current limit, maybe 80mg to 110mg, and just below
the current limit, maybe 50mg to 80mg. In Professor Allsop's figures,
as I've read them, he made assumptions about the behavioural effect
of changing on those bands of people. He didn't make any assumption
at all about what would happen to people above 110mg, so those
people who were quite far above the limit. He assumed they would
be completely unaffected by the change.
What we did in our modelling was use two core pieces
of evidence from overseas: a big European study that looked at
15 countries, and another much more detailed Australian study.
In both of those sets of evidence there were behavioural effects
seen, not just in the people around the limit but higher above
the limit, and indeed those below the current limit. The main
difference between our two estimates is that we widened, in ours,
the number of people who were affected to the ones very high up
and the ones low down. The big issue in numbers terms in the difference
between the two estimates is the effects that we looked at for
those people above 110mg. So we were essentially using evidence
from Australia and 15 European countries that had implemented
moving from 0.08 to 0.05.
Q118 Chair:
Dr Meldrum, do you have any observations on this issue?
Dr Hamish Meldrum:
I think we could spend a long time arguing about the exact numbers,
and that is always going to be an estimate. I don't think, though,
what is in doubt is the effect on performance that various levels
of blood alcohol seem to have. There is a significant difference
between performance at 80mg and performance at 50mg. Your performance
is quite greatly impaired at 80mg compared to 50mg. It is still
impaired at 50mg but not nearly so much. On that evidence alone
one can assume that you would actually reduce the number of accidents
and reduce the number of fatalities. As the other speakers have
said, because we don't know exactly where we are starting from
it is difficult to put absolute figures on, but there is no doubt
that actually you will reduce the level of accidents if you stop
people who have an impaired performance driving.
Q119 Chair:
So you are looking at general impairment?
Dr Hamish Meldrum:
Yes.
Q120 Iain Stewart:
Just to follow up on that point - forgive me if this is a naïve
medical question - in my mind the intake of a particular quantity
of alcohol affects people in different ways at different times.
If I have a glass of wine on an empty stomach, it affects me more
than the same quantity of alcohol with an evening meal, for example.
Are there other factors that affect the quantity of alcohol in
the bloodstream and therefore the level of impairment?
Dr Hamish Meldrum:
How you consume alcohol, in what form and whether you are eating
with it affects how quickly it gets into your bloodstream and
therefore affects the level in your bloodstream. For instance,
if you drink sparkling wine on an empty stomach that will raise
your blood alcohol levels quite quickly. If you drink something
with a meal then it won't; but it is the level of alcohol in your
blood that affects your performance. It is not directly how much
you've drunk. How much you've drunk is related to that, but, as
you say, there are other factors and how quickly you absorb it.
What is crucial is the level of alcohol in your blood, and that
is what you are measuring when you breathalyse people or take
their blood alcohol. It is the level of alcohol in your blood
that affects your performance, not necessarily how you have taken
in that alcohol and what other factors apply.
Q121 Iain Stewart:
I just ask to make sure that we are not failing to consider other
tests as well as purely the alcohol level in the bloodstream.
Dr Hamish Meldrum:
I think these other things are constant. If you basically said
to me, "You can't drive if you've drunk X glasses of wine",
although that's helpful for the public to know that, because of
the factors you've mentioned what is crucial about impairment
is the level of alcohol in your blood.
Iain Stewart: Thank you.
Q122 Kelvin Hopkins:
In a previous session, we touched on this, but I'm interested
in the impact of lowering the limit on behaviour. At 80mg many
people, even oneself, take a chance perhaps. You're not quite
sure whether one or two glasses of wine will get you over the
limit but you think it's probably okay. At 50mg the limit is so
low that, rather than take a chance, you don't drink at all. I
wondered if there is any evidence to support that theory.
Professor Mike Kelly:
Sorry, which theory?
Q123 Kelvin Hopkins:
That if the limit is reduced from 80mg to 50mg people don't take
a chance on how much they can drink: they don't drink at all before
they drive because they are not sure if even one glass of wine
might put them over the limit. Rather than take the chance, instead
of having a couple of drinks and thinking it will be okay, you
drink nothing.
Professor Mike Kelly:
Right.
Chair: So you are talking
about below 50mg?
Q124 Kelvin Hopkins:
Yes, below 50mg.
Professor Mike Kelly:
If you are talking about the psychological dynamics, there are
two parts of the process where a substance like alcohol confuses
the issue for the person consuming alcohol. Some of our behaviour
is automatic: you do it on the basis of not thinking through what
you're doing. It is an automatic response to a situation. The
other part of our behaviour is dictated by certain rational processes
where we make calculations. The difficulty with alcohol is that
it interferes with the second, and the former comes to predominate.
It is in an exponential relationship with behaviour. As the consumption
increases, the rational function in the mind decreases. So there
is a complex interaction going on. You will see behaviours being
changed as a consequence of consumption, even if it is only marginally.
But in terms of performance, as Hamish said, you also see this
question of impaired performance that moves up the scale as the
consumption of alcohol, or the volume of alcohol in the blood,
increases.
Q125 Kelvin Hopkins:
I am just thinking if I can rephrase the question. If one is going
out to a dinner party and one thinks, "Well, 80mg is enough;
I can have a couple of glasses of wine", at 50mg it is a
level where you think, "It's not worth having one glass because
it might be a problem, so I won't drink at all." I just wondered
if there is any evidence from abroad perhaps that people do change
their behaviour in that marked way.
Professor Mike Kelly:
I don't know what my colleagues would say, but I think in terms
of the cultural norms in relation to drinking, the not drinking
at all and driving seems to be much more common, particularly
among certain younger members of the population, than it used
to be. From the time when the breathalyser was introduced in 1967,
I think there has been a huge shift in cultural norms in relation
to drinking and driving. The kind of calculation you are making,
which is that it wouldn't be worth drinking at all, is probably
much more likely made now than it would have been in 1967 or 1968.
Q126 Chair:
Professor Allsop, I think you wanted to come in?
Professor Richard Allsop:
Yes. In this context I think it wouldn't be a new reaction. I
think we have benefited a great deal in terms of the effectiveness
of the existing limit from the fact that many people greatly overestimate
their risk of exceeding it with ordinary social drinking, like
a couple of glasses of wine with a meal over the evening. Of course,
that effect will be greater if the limit is reduced to 50mg. Cautious
people will be a bit more cautious. The fact that there will be
people who are being more cautious than they need be is already
a very widespread phenomenon.
Q127 Mr Harris:
The information we have is that, in 2008, 430 out of 2,538 road
accidents were related to drinking alcohol. That is about 17%.
From what Professor Brennan was saying, there is a group of people
who drink way above 80mg. Am I right in suggesting that this research
shows that those people wouldn't feel affected by a mandatory
lowering of the limit? Those are people who are problem drinkers
and problem drink drivers. What I am trying to work out is, of
those 430 and of that 17% of road accidents caused by alcohol,
what percentage of that are caused by people who frankly don't
care what the law is in terms of the alcohol limit? Those are
the people that a reduction won't affect.
Q128 Chair:
Professor Brennan, you look as if you want to answer that one.
Can I invite you to?
Professor Alan Brennan:
There are three or four things to unpack there in the 430 number.
It is a kind of Department for Transport definition which says
if the person has died while driving, and their blood alcohol
has been measured, they've been over 80mg at that point. That
isn't all of the people who die as a result of alcohol. There
are a whole gang of other people below the 80mg who are also dying
due to alcohol. It is a kind of arbitrary threshold -
Q129 Mr Harris:
Sorry to interrupt you, but does that figure also include, for
example, pedestrians who were not at all drunk but were hit by
drivers who had been drinking? It's not just the driver who dies
in the car; it is pedestrians also? I see your colleagues are
nodding.
Professor Alan Brennan:
I'm not sure. I think so, yes.
Professor Richard Allsop:
If I can clarify, it is an estimate from the Coroners and other
data of the number of people who died in accidents in which at
least one driver had an alcohol level over 80mg; but the people
who died may have had no alcohol at all.
Q130 Chair:
What would the impact be of a 50mg limit on those drivers who
take far higher levels of alcohol than the ones that Mr Harris
is talking about? There seems to be an assumption somewhere here
that lowering the limit would have an impact on those drivers.
Why is that?
Dr Hamish Meldrum:
I think it is quite difficult to quantify. There are, in a way,
two things that help stop people driving. One is the sense of
social responsibility to stop people drink-driving; and the other
is the fear of getting caught. Whatever system you have, you have
to have reliable and effective enforcement. That applies whether
your alcohol level is 80mg, 50gm or whatever, particularly I think
if you are going to get that category of people who, in my definition,
really don't have the same sense of social responsibility about
drink-driving. How much more reducing it is going to help that
group I would accept is quite difficult to say. I think for the
vast majority who have a sense of social responsibility, to get
back to Mr Hopkins' point, it will have an effect because they
will be even more cautious than they have been. What we are actually
trying to aim at is getting people not to drink at all when they're
driving, but you have to have an arbitrary level to allow for
some of the practicalities of enforceability. It will make them
more likely not to drink at all, and that in itself will reduce
the number of accidents.
Q131 Mr Harris:
Can I just pursue this, Chair? This 430 figure may be something
of a shibboleth - I am not sure - but we are asked to believe
that according to the European experience that number of fatalities
could be reduced by between 77 and 168, which is a massive percentage
of that 430; or 144 based on the Australian experience, which
is also a very large proportion of that total of 430. That is
between a third and a half of all fatalities. That is a huge reduction.
If we are to believe that the main effect is going to be on those
law-abiding, conscientious, responsible citizens who already respect
the drink-driving limits, and yet they don't seem to be responsible
for the vast majority of the deaths that already happen, if a
change in the law is only really going to affect the law-abiding
- the people who already respect the law - how can we expect to
see a reduction of 144 deaths a year when, at the moment, 430
deaths are caused by nutcases who just don't care about the law
anyway?
Dr Hamish Meldrum:
I didn't say it would only affect them.
Q132 Mr Harris:
It would largely affect them.
Dr Hamish Meldrum:
That might affect them more. I think it still will, and its enforcement
as well will help to reduce that.
Q133 Mr Harris:
But are you saying enforcement on its own would work without a
reduction in the legal limit?
Dr Hamish Meldrum:
Enforcement would help with that hard group. As I say, we don't
know exactly what the percentage and the make-up of that is, but
what I am also saying is there is absolutely clear evidence that
the difference in performance between 80mg and 50mg is significant.
Therefore, even outside that hard group, you will make a significant
difference. We keep getting back to this business of exact figures,
but then it is not an exact science because you are comparing
figures from abroad. We are not always starting from the same
place; we don't have the same enforcement regimes. We don't even
have the same culture. I get back to the point that you will have
a significant effect on road accidents and deaths if you reduced
from 80mg to 50mg. I think we could spend all morning talking
about how that is going to be at 10%, 20% or 30%, but to some
extent that is a guesstimate then.
Q134 Chair:
Professor Brennan, I think this was from your studies, wasn't
it? Can you help us on this?
Professor Alan Brennan:
Yes; can I have a go at explaining? The 68 to 144 numbers and
the 303 number at six years - all of those are seen in the context
of the whole 2,800-odd people who are dying. It is not just the
ones over 80mg that we have been looking at: it is everybody.
Not all of the effect is in the 430 people over 80mg. There is
a lot of effect in the people under 80mg as well, even though
they are at much lower risk. If you are at 80mg you are at 11
times more risk of having a fatality than when you're not drinking.
There are a lot of people in those groups, so even though they
are at a lot lower risk than someone at 150mg or 200mg, because
it is exponential, still affecting them a little bit is having
the effect because there are a lot of people in those groups of
having a large number of road accidents reduced. You are not really
talking about the 144 as a proportion of 430. The 144 is more
like a 6% reduction in fatalities in the country rather than 30%
or 40% or 50%.
Q135 Lilian Greenwood:
I think the point is that 430 - correct me if I'm wrong - are
the people who had a blood alcohol level of 80mg or greater, and
within that 2,538 there are people who would fall in the 50mg
to 80mg but we just don't have the number because there is no
recording of that. Is that right?
Professor Alan Brennan:
That is right. There has not been a roadside survey to get those
numbers since the late 90s, so we don't have the accurate numbers
about exactly how many people are driving round in England and
Wales today with what levels of blood alcohol. For our modelling
we have had to estimate those numbers by working back to those
numbers from looking at 18,000 fatalities and the blood alcohol
content for them, and the mathematics of the risk curve and how
many people must have been driving in those bands to get that
kind of number.
Q136 Chair:
Professor Allsop, do you want to add something?
Professor Richard Allsop:
What Professor Brennan has said is quite right. I would like to
make it clear that I have no problem at all with the modelling
and estimation that Professor Brennan has done. I think it is
the fact that we are talking about deaths in accidents in which
a driver has particular alcohol levels. We have a good estimate
which the Department for Transport relies on from year to year,
and most of us broadly believe. We have not only the number in
which a driver was over 80mg; we have the distribution of the
alcohol levels of the drivers over 80mg. We also have the numbers
of which drivers had amounts of alcohol less than 80mg, so we
have that full distribution, and both Professor Brennan and I
have used that. That is one of the common things about our work.
Coming back to Mr Harris's question, I think, notwithstanding
what Professor Brennan has said, a very large amount of the reductions
that he estimates come from the accidents currently in which a
driver has more than 80mg. I do feel that we should be very cautious
about the transfer of experience from South Australia and the
European study to our situation. I am sceptical about these large
numbers because the starting point in South Australia was one
where 2.5% of people over the whole week were driving at over
80mg. That emerges from the monitoring work. We are starting from
a situation where, over the whole week, only 0.3% of the driving
that is being done is estimated by Professor Brennan, and I agree
with the estimate, to be done by people who are over 80mg. That
means there was a lot more scope for a reduction at those levels
in South Australia than there is here.
I have to say that in my judgment - and it is only
a judgment - there has been no study which has looked at a lot
of people who are driving around at very high alcohol levels and
tried to assess, in relation to those real people, what their
response might be to the lowering from 80mg to 50mg. If we had
such a study, we would all be better off. Really effective enforcement
of the 80mg limit began in the early 1980s. In those two and a
half decades, or a bit more, we have reduced from more than 2,000
deaths a year in that category to less than 400. One of the results
of that is that we have only 0.3% of our driving being done at
that level.
When you have done all that, I am very sceptical
that one will get the same proportion of further reduction by
lowering the limit as in a situation in South Australia where
you were starting with 2.5% of people driving around over 80mg.
The European figures are estimates from modelling over experience
in a large number of countries. They pin down effects on people
of different ages and gender, but they don't get to grips with
how much of it is coming from how far up the alcohol scale under
the existing limit. I feel your scepticism about the size of these
numbers. I'm not saying that there will be none. I deliberately
said I am not going to count those, but I said there would be
some. I am sceptical that they are as large as our colleagues'
estimates make them.[1]
Chair: I think that ultimately
we have to make a judgment on what all of this means. What is
important for us today is that we do draw out from you the basis
of the differences and the implications of information from other
countries as well.
Q137 Mr Leech:
There is obviously a big discrepancy in the number of proposed
saved lives, but is there anyone dissenting from the view that
a reduction to 50mg would save at least some lives: one or more?
Professor Mike Kelly:
No.
Professor Richard Allsop:
No dissent at all.
Dr Hamish Meldrum:
No dissent.
Chair: Can you just say
that again louder for our records?
Professor Richard Allsop:
No dissent at all.
Dr Hamish Meldrum:
No dissent whatsoever.
Q138 Mr Leech:
Has anyone come up with any study that suggests that no lives
would be saved?
Professor Alan Brennan:
No
Professor Richard Allsop:
No.
Chair: Shaking the head
doesn't go down.
Dr Hamish Meldrum:
Sorry. I'm not aware of any study that suggests that no lives
or serious injuries would be saved.
Q139 Mr Leech:
So, in your view, implementation of the North Review would save
people's lives?
Professor Alan Brennan:
No
Professor Richard Allsop:
Yes.
Dr Hamish Meldrum:
Yes.
Q140 Mr Leech:
Can I ask whether you take the view that a reduction to 20mg would
save more lives, or do you feel that it could be the case that
the difference between 50mg and 20mg wouldn't necessarily save
any more lives?
Chair: Dr Meldrum, do
you have a view?
Dr Hamish Meldrum:
That is a bit more difficult. You do get the law of diminishing
returns the further down you go. Also you get into areas of public
acceptability and at 20mg just how genuine the effect is. Therefore,
you may get a bit of public resistance. On purely performance
levels, yes, people's performance improves between 50mg and 20mg,
but not by anything like the same degree as it improves between
80mg and 50mg, so you are into the law of diminishing returns.
I think, certainly from the BMA, we would argue that moving down
to 50mg will create a substantial improvement: moving down to
20mg brings with it rather practical issues and the amount of
improvement would be much less. Therefore, we would advocate at
this stage a lowering to 50mg.
Professor Richard Allsop:
I share that assessment.
Q141 Mr Leech:
Is there any evidence that reducing the limit to 50mg would increase
the number of people being caught over the limit: the morning
after drinking?
Dr Hamish Meldrum:
Common sense says that that is the case. Getting back to Mr Stewart's
point, obviously how your blood alcohol changes relies on a whole
lot of factors; first of all, what level it was; how quickly people
metabolise it and get it out of their system - some people do
it more rapidly than others. The point is that your level of impairment
relates to the level of alcohol in your blood at that time. If
you have had a heavy night drinking and have gone on to the early
hours, and then drive the next morning before that level has reduced,
your driving ability is impaired. Therefore, you will catch more
people, but I would argue that it is quite right that you do catch
these people because they are more of a danger.
Q142 Mr Leech:
I completely agree with that, but if we are assuming that a reduction
in the limit from 80mg to 50mg would change some people's behaviour
- notwithstanding the idiots who will always drink and drive over
the limit - shouldn't that change in behaviour also have an impact
in the behaviour of people drinking to excess the night before
and then driving in the morning? Surely, there will be a subsequent
change in behaviour there as well.
Dr Hamish Meldrum:
Yes, I would certainly hope that it would. Certainly that would
be part of the education to even the responsible drinkers who
have taken a taxi to their function, saying "If you're going
to be driving at 7 o'clock the following morning, not only do
you need to watch your intake during the evening but you probably
need to stop drinking at a relatively early hour if you're going
to drive the following day." If it does that, then it will
reduce the accident rate.
Q143 Mr Leech:
Just one more question. You said, and I completely agree, that
it would hopefully then catch more people the following morning
who were over the 50mg limit. We've had some conflicting evidence,
if it can be called evidence, that if the alcohol in your bloodstream
is decreasing rather than increasing, the impairment that creates
may be less if the blood alcohol level is going down. Is there
any real medical evidence that that is the case?
Dr Hamish Meldrum:
Yes. If you measure your blood alcohol at one particular point
in time, if it is a time when it is on an upward course, then
obviously your impairment is going to get worse over the next
hour or two. If it is on a downward course, it is going to get
better over the next hour or two. Over the period of time following
that period of measurement, then you will become less risky in
the morning if your blood alcohol level is dropping, and you will
become more risky in the evening if it is still going up. The
degree of impairment at that point in time when you measure the
blood alcohol is the same, whether it is going up or coming down.
Q144 Iain Stewart:
Following on from these points, I have a concern that there is
a considerable degree of ignorance in the public about what they
can consume in alcohol within all these limits. My further concern
is that if we reduce from 80mg to 50mg you are going to increase
that uncertainty. Should the message therefore be that you just
don't drink and drive at all and you take it down to effectively
zero - to understand the higher level of 20mg. I am coming to
the view that you might have to argue for either keeping the limit
as it is with much stricter enforcement or saying nothing, rather
than have this uncertainty in the middle.
Dr Hamish Meldrum:
It is always a balance between what might be nice in an ideal
world and what is practically achievable. As I said earlier, I
would argue that with some of the problems in bringing it down
very low, whereas it may be perfectly reasonable and logical in
terms of actually trying to reduce the number of alcohol-related
accidents to an absolute minimum, you get into these practical
issues and public acceptability issues, which might mean that
people are less happy with the type of legislation that is being
imposed upon them. You need to have, in a way, public ownership
of legislation to help to get the degree of co-operation you want.
I would take it contrary to you: I think reducing from 80mg to
50mg would have a significant impact both on the number of accidents
and on people's behaviour as well. Yes, the message should still
go out "Do not drink and drive" and that is the only
way to be absolutely safe in terms of both having accidents and
getting caught; but I think in terms of enforcement then you get
into other areas of difficulty which might cloud the issue.
Q145 Chair:
Professor Kelly, you wanted to make a point?
Professor Mike Kelly:
Yes. I simply wanted to say, of course, that you're absolutely
right. There is a degree of uncertainty - and it is in the various
reports that have come forward from NICE and others - among large
sections of the population. If the limit were to be lowered this
would be an ideal moment to educate the population again, as happened
in 1967. Some of the rules of thumb, although they are not scientifically
precise but they are close enough, could be very helpful with
an education campaign to go with it. I think it is quite clear
too, in the evidence that has come to the Committee from us and
from others, that this is not just a legislative matter. It is
legislation; it is education; it is enforcement. It is a range
of things being done together which will produce the effects that
we are looking for rather than one single thing.
Q146 Kwasi Kwarteng:
I just wanted to reinforce my colleague's point over there. Clearly,
yes, you are right to say that education is very important and
you mentioned the fact that in 1967 there was a campaign.
Professor Mike Kelly:
There certainly was.
Q147 Kwasi Kwarteng:
But in 1967 people's drinking habits were completely different
to today. We have a much wider range of alcoholic substances.
We have alcopops; we've got a huge range of things that people
consume. To educate people on that broad range of things that
are out there, in terms of one pint of beer equals whatever it
might be, I think is perhaps unrealistic. I mean, I remember those
posters from 30 years ago, where you were essentially assuming
that people either drank wine or beer, and possibly whisky. It
was a much more limited range of alcohol that was on offer, so
to expect the public to know - given the wide range of things
they can drink - what the limits are I think is unrealistic, even
after education.
Professor Mike Kelly:
I think we have a problem more broadly and it goes well beyond
the question of drinking and driving. The levels of alcohol consumption
since 1967 have gone up enormously in this country across all
sectors of the population with, as you rightly say, a broader
range of products and so on. Indeed, an alcohol strategy would
have to embrace dealing with that as well as these other things.
Q148 Mr Harris:
I am going to have one more shot at this. Professor Allsop, you
mentioned that you have breakdowns of the blood alcohol content
figures for a range of people contained within those 430 accidents.
I don't know if the Committee already has that information. It
would be extremely useful if we could have that because it is
a missing piece of the jigsaw. At the moment we have a very blunt
instrument. We have the figures for people who were over the legal
limit, but that doesn't really tell us very much unless we can
get the breakdown of what proportion of those people were over,
say, 100mg to 110mg.
Chair: Could you send
the Committee that information?
Professor Richard Allsop:
Can I refer you to Chart 3.5, not necessarily now but when you
have time? Chart 3.5 on page 59 gives a breakdown of all killed
drivers over the alcohol intervals. I beg your pardon; that is
an age diagram.
Chair: We will find it.
We know it is there, so we will find it.
Professor Richard Allsop:
Let me take time and then come back to you with the right number.
Q149 Mr Harris:
The second point I was going to raise, Chair, is: what do you
think - all four of you - about the possible consequences of reducing
the minimum? The legal maximum blood alcohol content is one measure
of getting these road accidents down. Another way would be better
enforcement. My understanding is that the roadside tests have
reduced by 25% - 800,000 to 600,000 - in recent years. Do you
think that roadside enforcement could have an equal or even a
greater effect than reducing the legal upper limit?
Chair: Who would like
to give a view on that? Dr Meldrum.
Dr Hamish Meldrum:
I think it is difficult to say, because of all the reasons we
have talked about earlier about the figures. In my opinion, it
might be that you need to do both. It is not just the level. We
know the scientific evidence about the degree of impairment at
the various levels, and that is incontrovertible. But also you
have to have effective enforcement and I would actively accept
that you need to do both. Which would be more effective I don't
think I can answer, and I'm not sure there are figures out there
that would actually answer that. Both would have an effect, and
much more than just either one on its own.
Chair: Ms Greenwood, did
you want to say something?
Lilian Greenwood: That
was precisely the question I was going to ask.
Professor Richard Allsop:
May I answer what I wasn't able to? I will still refer Mr Harris
to Chart 3.5 because that has four alcohol categories: No alcohol
present; Alcohol present but not over the limit; Over the limit;
and Over twice the limit. So there is some breakdown there, but
I'm sure that Mr Harris has access - or would have friends who
have access - to Professor Brennan's own paper. If he will look
at Table 2 in Professor Brennan's paper afterwards then the breakdown
is there. It is very detailed indeed, at intervals of 10mg right
the way from zero up to greater than 300mg.
Q150 Mr Leech:
Professor Meldrum, in your -
Dr Hamish Meldrum:
I am afraid I'm only a simple doctor.
Mr Leech: Sorry; Dr Meldrum.
Chair: He has been impressing
us a lot today.
Mr Leech: In your professional
experience dealing with patients, how big a problem would you
say is drug-driving?
Dr Hamish Meldrum:
I think it is still significant. I am a GP and I still practise
one day a week despite my BMA duties. I think, although, as others
have said, we have improved considerably over the last 30 years,
there are still a number of people in all walks of life who drink
above an advisable level for driving and drive with blood alcohol
levels that are too high. I would say I think there is, in some
sections of society, a degree of complacency compared with the
initial reaction we got in the late '60s and early '70s, following
the first drink-drive legislation. Therefore, I think, leaving
aside the scientific evidence, if you did agree to a reduction
this would have a new impact on that and with the education we've
talked about would help to improve that behaviour.
Q151 Mr Leech:
Sorry, I think you misheard me. I was actually referring to drug-driving
rather than drink-driving.
Dr Hamish Meldrum:
I am terribly sorry. There are two categories. There are obviously
the illegal or controlled drugs and even taking them of itself
is an offence; added to that, driving would impair that too. Although
I don't have so much experience in terms of hard drugs like heroin
or cocaine, certainly in terms of cannabis, I am sure there are
quite a lot of people who drive while taking that, and that does
impair their driving. Once it comes to prescribed drugs, I think
we really get into a bit of a minefield. We have talked about
some of the individual variations in terms of alcohol. They are
probably much greater with prescribed drugs. You have to ask:
would you rather have somebody who is severely depressed but untreated
driving, or would you rather have somebody being adequately treated
even though the drug might cause a minimal impairment of that
driving ability? The same might apply to somebody who has a very
heavy, streaming cold and is sneezing all the time and takes a
cold remedy. You do get into much more difficult areas and there
is not an easy test. You can't just say that because somebody
has got a level of a certain drug in their blood, even if you
could measure it, that will cause X degree of impairment. You
are much more then into actual judgments about whether, in that
individual, their ability has been impaired either through prescribed
drugs, controlled drugs or a mixture of drugs and alcohol.
Q152 Mr Leech:
Do you think then it is unrealistic to get a medical consensus
on impairment through drugs?
Dr Hamish Meldrum:
If you are saying, "If you take drugs X and Y, you mustn't
drive, but if you take drugs A and B, you can drive", I don't
think you would get a medical consensus on that. With controlled
drugs -
Q153 Mr Leech:
What, now or ever?
Dr Hamish Meldrum:
Certainly not with the knowledge and the testing abilities we
have at the moment. I don't think you would get an adequate consensus
on that. In terms of controlled drugs, I take the view that you
are already committing an offence taking them; you are probably
adding to the severity of that offence taking them and also driving.
Q154 Mr Leech:
But haven't the BMA taken a zero tolerance attitude in relation
to drug-driving?
Dr Hamish Meldrum:
If you are talking about controlled drugs, yes.
Mr Leech: Just controlled
drugs?
Dr Hamish Meldrum:
You can't take zero tolerance in terms of prescribed drugs, because
I'm afraid the evidence is not there to allow you to make those
sorts of accurate judgments. Yes, we know that there are certain
drugs that are more likely to impair your ability, but how much
they will impair that ability in any individual and in any given
set of circumstances - and of course it depends what combination
of drugs they are taking and a whole lot of other factors - I
think to get a degree of consensus on that, you are almost back
to the warning they put on drugs at the moment: "If your
ability is impaired, don't drive". But how you make, and
who makes, that judgment is much more difficult.
Q155 Mr Leech:
Professor Allsop, you suggested that impairment, not just presence,
needs to be established as far as drugs are concerned. Why do
you think that is so important?
Professor Richard Allsop:
Mainly because, as I understand it - and I am on the fringe of
my expertise here - with recreational drugs there are some of
those which stay in the body for a long time after they have ceased
to impair. Therefore, if it were going to be an offence to have
it in the body at all, then people could of course give up their
recreational drugs but, having taken them, they would face really
quite a long time when they shouldn't drive. It depends on voluntary
choice or drug enforcement, but it concerns me that people would
be committing an offence in a state in which they are not impaired
in their driving and they may not be in a position, unless they
are quite well educated, to know that they've still got the stuff
in them. That is the big difference between alcohol, which I regard
as the simple drug in terms of having a driving law, and quite
a few of the recreational drugs. With alcohol, the impairment
is very closely in time and level matching the presence in the
bloodstream, whereas in the case of the other drugs it is not
sufficient just to test for presence. You have to test for presence
in a way that is impaired. I believe that this is a scientific
challenge, but I am not expert on it.
Q156 Mr Leech:
Dr Meldrum, when it comes to tests being carried out in custody
the BMA have suggested that nurses don't have the relevant training,
experience and expertise to undertake this role. Why do you think
that, and is there no way that they could get that relevant experience
and training?
Dr Hamish Meldrum:
I think the main reason we have that view is because it is an
area of judgment. It is quite a skilled area of judgment and an
imperfect area of judgment too. In terms of actually taking blood
samples and things, that is not a problem; but if you are actually
making a judgment about whether somebody is impaired, then I think
that is quite a skilled practice, and even I, as a GP, wouldn't
feel qualified to do that. Of course you could train a nurse to
the level of a forensic physician, but if you'd done that I think
I would call her or him a forensic physician rather than a nurse.
Q157 Julian Sturdy:
I think Mr Leech has covered a lot of the points, but could I
just follow up on some of the points he made earlier on. I very
much support what North is saying on drug-driving but there is
a fundamental problem here, which has been highlighted I think
with what the panel has said. I would be interested in everyone's
comments on this. North is recommending that the Government should
actively pursue research to determine proscribed levels of controlled
drugs, similar to the current drink-drive offence. The problem
is when we are talking about drink-driving, we are talking about
one drug: when we are talking about drugs, we are talking about
a huge spectrum of illegal and permitted. The problem here is
going to be the medical consensus in the whole process if we take
it forward. How do you think we can come to some sort of consensus
across the board? I put that to everyone because this is going
to be the big issue for drug-driving.
Dr Hamish Meldrum:
I have probably said enough, but the way you get consensus in
the medical profession is good evidence, and as you've suggested,
there isn't good evidence at the moment. I would agree with Sir
Peter North that you need to do more research to try to get better
evidence. If you can get better evidence then you're more likely
to get consensus.
Q158 Julian Sturdy:
Do you think that evidence is going to be easy to get?
Dr Hamish Meldrum:
No.
Julian Sturdy: That's
the problem, isn't it?
Chair: Does anybody else
want to comment on whether there can be a consensus? No; there
doesn't seem to be. Anything else, Mr Sturdy?
Julian Sturdy: No.
Q159 Iain Stewart:
One area that we touched on in a previous session was what we
called the combination of drink and drug-driving. I would be interested
in your comments on two very separate scenarios. One is what you
might term an innocent one. Someone has a very small quantity
of alcohol and takes, for example, a couple of antihistamine tablets.
Based on your comments, does that combination put them in an impaired
driving situation? At the other end of the scale, one of our previous
witnesses mentioned that there is evidence that people who take
recreational drugs in a nightclub and then driving home will consume
a very small amount of alcohol so that if they're stopped and
the police test them for drink, they will be under the limit but
actually their impairment is considerable because of the drugs.
I would just be interested: should we be looking at that combination
effect as well as the separate drink-driving and drug-driving?
Dr Hamish Meldrum:
You absolutely should because the combination can make a difference.
The problem is that unlike alcohol, which in this sense is a fairly
simple thing both to test or to assess the degree of impairment,
there is much more individual variation. For instance, in your
first scenario of the antihistamines plus a small amount of alcohol,
antihistamines on their own cause a very variable reaction in
people. Some people react quite badly to them; with some people
there is very little impairment at all. If you add into that the
combination of alcohol, then that variation is still there.
On your point about illegal drugs, again, getting
back to your colleague, there is the problem of lack of research
and not knowing exactly what levels and what degree of impairment
relates to those levels; and also what the effect is of the combination
of alcohol. I take your point that some people may well use alcohol
to say, "Well, I'll get tested for that and they will forget
to test me for drugs". Although you can try and do a fairly
broad screen, it is pretty complex and quite time-consuming. If
you want to screen for all sorts of drugs, including prescribed
drugs, it is very expensive, it is very involved and it is very
time-consuming. As I said earlier, the evidence there for what
impact that has had in that individual is pretty scanty.
Q160 Mr Harris:
It is quite difficult, isn't it, to get a handle on exactly how
big this drug-driving problem is because there is not a consensus
on testing, or whether it is prescription or controlled drugs?
Do any of you have even anecdotal evidence about the actual percentage
or the size or scale of the problem that we can actually use as
part of our deliberations? Unless we have a figure on a bit of
paper saying that X percentage of accidents is caused by drug
taking, and of that X%, Y% is controlled and Z% is prescribed
- do we have any evidence or do we have to rely on anecdotal evidence?
Chair: Does anybody want
to answer that?
Mr Harris: I think that
answers the question, actually.
Dr Hamish Meldrum:
I think it is mainly anecdotal. We know that there is an increase
in the amount of recreational drug use - that is not in doubt
- but how much that feeds through then into those people who drive
and also what impact that has on their driving, I think, as I
have said earlier, we need much more research and much more evidence.
Only when you have done that are you likely to get more of a consensus.
Q161 Mr Harris:
So you are not recommending changes in legislation in this respect?
Presumably you wouldn't if there is so little evidence?
Dr Hamish Meldrum:
I think, in terms of controlled drugs, I take the point that it
is already committing a crime to consume them or to use them.
I would suggest, though it is not for me, if that is the law that
it might be considered that you've added to the severity of that
by actually driving as well as having taken a controlled drug.
In terms of prescribed drugs, I think we need an awful lot more
research, because otherwise I think it would be impossible to
create legislation given our present degree of knowledge.
Q162 Kelvin Hopkins:
Just before you leave us, I wondered if you could reinforce your
message. We have talked about a lot of other factors in deaths
from drink-driving and drugs, comparisons within different countries,
from different starting points and so on. I am concerned that
those who would oppose a reduction in the limit might take some
comfort or clutch at straws. Could you say, keeping all other
factors constant so that you just look at alcohol consumption
and accidents, that if you reduce the level from 80mg to 50mg
many lives will be saved, and it is inevitable as a demand curve
in economics that if you either reduce the price more will be
sold or if you increase the price less will be sold?
Professor Mike Kelly:
Yes, if I may. When NICE sets about doing an investigation of
anything, it seeks to appraise the best available evidence that
we can find, to assess its quality and then to build our recommendations
on that. We are not making recommendations here but we've certainly
assessed the evidence. I actually have two things in mind. First
of all, in terms of the quality of the studies from both Europe
and Australia, they are very good. This is good science. The second
question: can you apply that in the British context? Well, you
can, but there is some variation in terms of how one might predict
what will happen on the basis of the different modelling that
we have heard about. There is nothing unusual about that in science.
Science is about handling uncertainty, and what you have seen
here is quite a classic example of the way that building different
assumptions into what you do you will get different results but
- and this is very important - the direction of travel of this
evidence is one way. It is very clear; and it is that there is
a direct relationship between the number of accidents, the number
of fatalities and the volume of alcohol consumed. Therefore, other
things being equal, if you were to reduce that, the number of
accidents and the number of fatalities would decline. In terms
of the sorts of evidence that I look at in public health, this
is pretty compelling stuff, it has to be said. This is not a chance
association; it is an association that we've seen over 40 years
of research, pored over by very clever statisticians. It only
goes one way. That for me is both a practically plausible as well
as a scientifically robust argument.
Dr Hamish Meldrum:
Can I just add to that? I think you are quite right to ask that
question. As scientists we always try to be on one hand and the
other hand, but I am in no doubt - and the BMA is in no doubt
- that reducing the level from 80mg to 50mg would save many lives.
How many I can't say, but it would save many lives.
Q163 Angela Smith:
I just wanted to explore briefly the difficulties that could be
faced by drivers who take prescribed drugs, particularly antihistamines,
which are probably one of the most commonly prescribed drugs.
It seems to me that a great deal is left to the judgment of the
individual when it comes to antihistamines. You get warnings,
for instance, about potential drowsiness when taking these drugs
and you are told not to drive if you feel drowsy. That is a particularly
broad piece of advice when it comes to something as critical as
getting behind the wheel of a car. In addition to that, the advice
given on things like antihistamines is pretty broad when it comes
to drinking alcohol in combination with them nowadays. I would
just like the comments particularly of the BMA on that issue,
because I think a lot is left to the judgment of the individual.
Dr Hamish Meldrum:
I am afraid it will always be that. I would imagine more antihistamines
are bought over-the-counter than are probably prescribed. They
are freely available over-the-counter. Therefore, you are relying
on either the chemist or people reading the packet. I would say
the same about the judgment that an individual has to exercise
when they have had a few late nights and they are driving when
tired. Falling asleep at the wheel is dangerous. That requires
your judgment as to whether or not you feel that you should stop,
take a rest, take a break. I am not in any way going to absolve
the individual from making a judgment. I think when it comes to
these sort of drugs, because they have such an idiosyncratic effect
and it varies so much from person to person, then it is quite
right and in fact it's the only thing we can really rely on to
say, "If you feel you are affected and you feel your ability
is impaired, don't do it, and if you want to be absolutely safe
don't take them and drive". I think you have to rely on people's
common sense and judgment.
Angela Smith: And it will
ever be thus.
Dr Hamish Meldrum:
And it will ever be thus. We can't devise a system where you take
personal responsibility and personal judgment out of it altogether.
Angela Smith: That is
helpful.
Chair: Thank you very
much for coming and answering all our questions.
1 See supplementary written evidence from NICE (Ev
96). Back
| 
