Supplementary memorandum by Dr Petra Meier
(AL 62A)
You have asked me for some observations to explain
how it can be that both industry and health professionals cite
the ScHARR report as supporting their case, and I will try to
do so below.
1. The focus of our attention on both adults
and youngsters, rather than just young people. This is in recognition
that advertising may play a larger role in influencing the continuation
of drinking behaviours in existing consumers than in the inception
of new drinker groups. From a population harm perspective, this
distinction is crucially important. However the evidence base
on advertising effects on adults is both smaller and weaker than
for underage drinkers. No longitudinal cohort studies covering
the older age groups have been identified. What there is tends
to be econometric studies linking changes in advertising expenditure
to changes in population level consumption, finding only tiny
effects. However, advertising expenditure is fairly stable at
a high level in most developed countries so there is limited variability,
and studies also do not differentiate types of advertising or
target audiences.
In terms of interventions, codes and bans are
typically designed to protect young people and any effects of
bans on adults remain largely unknown. In practice, only France
has introduced a comprehensive ban but no convincing evaluation
was carried out.
Finally, we were tasked with finding evidence
on the differential effects of advertising restrictions on moderate,
hazardous and harmful drinkers, as our review is public health
focused and thus recognises that alcohol causes harm not only
or even primarily to those who are dependent or drink very heavily.
However, again the advertising literature does not give us information
on whether heavy drinkers are more affected by advertising than
moderate drinkers, although general advertising theory would suggest
this to be likely.
I think this explains why our evidence statements
are overall more cautious than that of authors who focused on
under-18s only.
We have tried to make this clear in the full
report, but realise that these points were all but lost in the
executive summary. Thus, evidence statement 6 should be seen
in this context (Evidence statement 6: There is an ongoing methodological
debate on how advertising effects can and should be investigated
and further research and methodological developments for establishing
a definite causal relationship is required.)
2. We investigated the link between advertising
and consumption only. We did not cover the many studies on the
degree of advertising young people or adults are exposed to, which
show convincingly that a programme does not have to be targeted
at young people to attract very significant numbers of young viewers
(eg soap operas, sport, talent shows etc). We also did not cover
the large body of literature showing that advertising codes do
not cover content features that are particular appealing to young
people (especially humour, music, animals etc).
3. There is more evidence on the link between
alcohol advertising and consumption than there is on the effectiveness
of codes or bans. However, few countries have changed their policies
in recent times thus there were limited opportunities for evaluations.
Also, countries where research exists have different drinking
contexts (drinking preferences, existing regulatory framework)
to that of the UK, so we found relatively few relevant evaluations.
It is however not the case, as some witnesses implied, that we
found that codes or bans have been shown to be ineffective.
We do point to the need for further research.
However, this is only possible if there is an actual policy change
that can be evaluated. Therefore, witnesses' suggestions that
policy changes should wait until further research becomes available
are not helpful. Instead, where policy changes are planned, it
would be important that policy makers engage with scientists so
that appropriate evaluations can be put in place.
4. Some witnesses referred to us saying
that advertising may actually increase consumption, without further
qualifying this. The context for this observation is that one
article by Nelson et al speculated that, in the absence of price
controls, restrictions on advertising could lead to price
wars. Since the effect for price appears stronger than for advertising,
this could have detrimental effects. No evidence is available
and this is only mentioned in order to point to the need for integrated
policy decision making.
5. In terms of new media, I'd like to refer
to the following section in our full systematic review report:
"There is a large evidence base […]
around established channels such as the mass media but a shortage
of studies evaluating newer media such as the internet and mobile
phones. Generally, the vast array of channels and of types of
promotional activity (Jernigan and O'Hara 2004) make it difficult
to isolate individual effects, and thus target individual strategies,
even though they consistently demonstrate an aggregative effect.
Policy options should therefore recognise where a common underpinning
mechanism exists and apply general principles to target such a
mechanism in anticipation of new channels rather than continually
attempt to respond to specific evidence on every new medium."
(p.80)
6. For completeness, we wish to point to
several evidence statements that some witnesses appear to have
missed:
Evidence statement 5: There is conclusive
evidence of a small but consistent association of advertising
with consumption at a population level. There is also evidence
of small but consistent effects of advertising on consumption
of alcohol by young people at an individual level.
Evidence statement 8: There is consistent
evidence to suggest that exposure to outdoor advertising, or advertisements
in magazines and newspapers may increase the likelihood of young
people starting to drink, the amount they drink, and the amount
they drink on any one occasion. Further research is required on
whether what young people say they are going to do at a particular
point in time translates into actual subsequent behaviour.
Evidence statement 10: There is consistent
evidence from longitudinal studies that exposure to TV and other
broadcast media is associated with inception of and levels of
drinking. Evidence for the effect of watching videos is equivocal.
Evidence statement 11: There is some inconclusive
evidence that suggests that advertising bans have a positive effect
in reducing consumption. Differences in contextual factors are
a likely explanation for these differences. It is methodologically
challenging to control for all possible confounding factors.
Evidence statement 12: There is some evidence
to suggest that bans have an additive effect when accompanied
by other measures within a general environment of restrictive
measures.
7. Dr Taylor and Ms Atkins asked witnesses
repeatedly for information on how overall consumption/sales would
be affected if everyone drank within the moderate drinking guidelines
but did not get a response (Q564, Q785, Q803). Whilst you did
not specifically ask me to comment on this, I have recently done
a calculation using General Household Survey data, so I thought
I will let you have the figures in case they are still of interest:
We used the weekly guidelines of not exceeding
21 units for men and 14 units for women. We assumed
that everyone who currently drinks more than 21/14 units
would shift to become only just compliant with the limits, so
now drinks exactly 21/14 units.
According to the GHS, 42,693,731 adults
(>16) currently consume 577,667,488 units per week (an
average of 13.5 per person—note: includes abstainers).
After the cap, this would fall to 347,462,198 units (8.5 per
person). Thus, if everyone who currently drinks over the limit
became just compliant with moderate drinking guidelines, the total
alcohol consumption would drop by 40%. This is explained by
the distribution of consumption in the population, see below a
Table from a forthcoming article based on our work, due to be
published in the journal Addiction.
Table 1
PROPORTION OF ALCOHOL CONSUMED BY PERCENTILES
OF THE POPULATION AND ITS VARIATION ACROSS AGE/SEX GROUPS
| Consumption decile
| Total population | Men
| Women | Young men
| Young women |
| 10th decile (10% heaviest drinkers) | 44.9%
| 52.0% | 32.4% | 59.0%
| 37.3% |
| 9th decile | 20.7% | 20.7%
| 20.7% | 17.5% | 20.3%
|
| 8th decile | 13.3% | 12.2%
| 15.2% | 10.3% | 15.4%
|
| 7th decile | 9.8% | 7.8%
| 13.3% | 6.2% | 10.2%
|
| 6th decile | 5.9% | 3.8%
| 9.4% | 3.7% | 8.5%
|
| 5th decile | 3.4% | 2.2%
| 5.4% | 2.1% | 5.4%
|
| 4th decile | 1.6% | 1.0%
| 2.7% | 0.9% | 2.2%
|
| 3rd decile | 0.4% | 0.2%
| 0.8% | 0.2% | 0.7%
|
| 2nd decile | 0.0% | 0.0%
| 0.1% | 0.0% | 0.1%
|
| 1st decile (10% lightest drinkers) | 0.0%
| 0.0% | 0.0% | 0.0%
| 0.0% |
August 2009
| | | |
| |
|
