The publication Alcohol and health in Wales 2014 consists of:
- a Wales profile
- seven health board summary documents
- interactive data files with the main indicators
- a technical guide describing the methods, data sources and caveats
These are available on www.publichealthwalesobservatory.wales.nhs.uk/alcohol
Main project team (analysis and writing): Andrea Gartner, Ioan Francis, Dee Hickey,
Rhian Hughes, Leon May
Additional analysis work: James Allen, Hugo Cosh, Arthur Duncan-Jones, Rhys Powell,
Holly Walsh, Mari Ann Jones
Project manager: Andrea Gartner
Project director: Nathan Lester
Acknowledgements:
Thanks to the following people for their help with this publication: Professor Mark Bellis, Helen
Crowther, Dr Gareth Davies, Ruth Davies, Andrew Dring, Cath Roberts (WG), Isabel Puscas,
Mark Robinson (PHE), Sacha Wyke (PHE)
Contact:
Public Health Wales Observatory
14 Cathedral Road
Cardiff CF11 9LJ
Email: [email protected]
Website: www.publichealthwalesobservatory.wales.nhs.uk
© 2014 Public Health Wales NHS Trust
All tables, charts and maps were produced by the Public Health Wales Observatory and the data
sources are shown under each graphic. Material contained in this profile may be reproduced without
prior permission provided it is done so accurately and is not used in a misleading context.
Acknowledgement to Public Health Wales NHS Trust to be stated. Typographical copyright lies with
Public Health Wales NHS Trust.
Alcohol and health in Wales 2014: Wales profile
Table of contents
FIGURES & TABLES ................................................................................. 2
KEY MESSAGES ........................................................................................ 5
1
INTRODUCTION................................................................................ 6
2
DEFINITION OF UNITS AND DRINKING PATTERNS .......................... 7
3
PATTERNS OF DRINKING.................................................................. 9
3.1
Drinking in children and young people .......................................... 9
3.2
Consumption in adults in Wales .................................................. 10
3.2.1 Frequency of alcohol consumption in Wales .................................11
3.2.2 Alcohol consumption trend in Wales ............................................11
3.2.3 Abstainers ............................................................................... 12
3.2.4 Drinking patterns by age group and over time ............................. 13
3.2.5 Geographical variation in very heavy drinking .............................. 16
3.3
Drinking patterns and trends in Great Britain ................................ 18
3.4
Comparison of survey results and sales data ................................ 18
3.5
International comparison of consumption .....................................20
4
HOSPITAL ADMISSIONS ................................................................. 21
4.1
The scale and breakdown by condition .........................................22
4.2
Hospital admission trends over time ............................................24
4.3
Geographical variation in hospital admissions ............................... 26
4.4
Comparison to England and English regions..................................28
5
MORTALITY .................................................................................... 32
5.1
Alcohol-specific and alcohol-attributable mortality ......................... 32
5.2
Chronic liver disease and cirrhosis ............................................... 34
5.3
Months of life lost .....................................................................35
5.4
Geographical variation in mortality ..............................................37
5.5
Comparison to England and English regions..................................38
6
SOCIO-ECONOMIC PATTERNS......................................................... 41
7
REFERENCES................................................................................... 47
1
Alcohol and health in Wales 2014: Wales profile
Figures & Tables
Table 1
Number of units in common alcoholic drinks
7
Table 2
Overview of definitions used
8
Figure 1
Percentage of respondents who reported drinking alcohol at least once a week,
boys and girls aged 11-16, 2009/10
10
Figure 2
Reported average frequency of drinking alcohol, percentages, males and
females aged 16+, Wales, 2012
11
Figure 3
Percentage of adults who reported drinking on the heaviest day in the past
week, age-standardised percentage, males and females aged 16+, Wales, 2008
-2013
12
Figure 4
Percentage of adults who reported abstaining from alcohol, males and
females aged 16+, Wales, 2008-09 and 2011-12
13
Figure 5
Percentage of adults who reported drinking above guidelines, males and females
aged 16+, Wales, 2008-09 and 2011-12
14
Figure 6
Percentage of adults who reported heavy (binge) drinking, males and females
aged 16+, Wales, 2008-09 and 2011-12
15
Figure 7
Percentage of adults who reported very heavy drinking, males and females aged
16+, Wales, 2008-09 and 2011-12
15
Figure 8
Percentage of adults reporting very heavy drinking (males over 12 units,
females over 9 units) on the heaviest drinking day in the past week, agestandardised percentage, persons, Wales health boards, 2011-2012
16
Figure 9
Percentage of adults reporting very heavy drinking (males, over 12 units,
females over 9 units) on the heaviest drinking day in the past week, Wales,
2008-12
17
Figure 10
Percentage of adults who reported drinking alcohol on the heaviest drinking day
in the past week, persons aged 16+, Great Britain, 2012
18
Figure 11
Comparison of sales and survey consumption data, weekly alcohol units per
person, 1996-2012
19
Figure 12
International alcohol consumption for selected countries, litres of pure alcohol
per person aged 15+, 2010
20
Table 3
Overview of definitions for person-based hospital admission indicators
21
Figure 13
Alcohol-specific hospital admissions by condition, 2012/13
22
Figure 14
Alcohol-attributable hospital admissions (narrow) by condition, 2012/13
23
Figure 15
Alcohol-attributable hospital admissions (broad) by condition, 2012/13
24
Figure 16
Hospital admissions (person based), European age-standardised rate (EASR)
per 100,000, males and females, all ages, Wales, financial years 2003/042012/13
25
Figure 17
Alcohol-specific hospital admissions, 3-year rolling crude rate per 100,000,
males and females aged under 18, Wales, financial years 2003/04-2012/13
26
Figure 18
Alcohol-specific hospital admissions (person-based), European age-standardised
rate per 100,000, males, all ages, Wales health boards, financial year 2012/13
27
2
Alcohol and health in Wales 2014: Wales profile
Figure 19
Alcohol-specific hospital admissions (person-based), European age-standardised
rate per 100,000, females, all ages, Wales health boards, financial year 2012/13
27
Figure 20
Alcohol-specific hospital admissions (person-based), persons, all ages, Wales,
2010/11-2012/13
28
Figure 21
Alcohol-specific hospital admissions, European age-standardised rate (EASR)
per 100,000, all ages, Wales, England & English regions, financial year 2012/13
29
Figure 22
Alcohol-specific hospital admissions, 3-year rolling crude rate per 100,000,
persons aged under 18, Wales & England, financial years 2003/04-2012/13
30
Figure 23
Alcohol-attributable hospital admissions (person-based narrow), European age
standardised rate (EASR) per 100,000, all ages, Wales, England & English
regions, financial year 2012/13
30
Figure 24
Alcohol-attributable hospital admissions (person-based, broad), European agestandardised rate (EASR) per 100,000, all age, Wales, England and English
regions, financial year 2012/13
31
Table 4
Overview of definitions for mortality indicators
32
Figure 25
Alcohol-specific and alcohol-attributable mortality by condition, 2010-12
33
Figure 26
Mortality, European age-standardised rate (EASR) per 100,000, males and
females, all ages, Wales, 2003-05 to 2010-12
34
Figure 27
Mortality from chronic liver disease and cirrhosis, European age-standardised
rate (EASR) per 100,000, males and females, all ages, Wales, 2003-05 to 201012
35
Figure 28
Months of life lost due to alcohol, males aged under 75, Wales health boards,
2010-2012
36
Figure 29
Months of life lost due to alcohol, females aged under 75, Wales health boards,
2010-2012
36
Figure 30
Alcohol-specific mortality, European age-standardised rate per 100,000, males,
all ages, Wales health boards, 2010-2012
37
Figure 31
Alcohol-specific mortality, European age-standardised
females, all ages, Wales health boards, 2010-2012
37
Figure 32
Alcohol-specific mortality, persons, all ages, Wales, 2003-12
38
Figure 33
Alcohol-specific mortality, European age-standardised rate (EASR) per 100,000,
all ages, Wales, England & English regions, 2010-2012
39
Figure 34
Alcohol-attributable mortality, European age-standardised rate (EASR) per
100,000, all ages, Wales, England & English regions, 2012
40
Figure 35
Self-reported alcohol consumption, age-standardised percentage by deprivation
fifth, males aged 16+, Wales, 2011-12
42
Figure 36
Self-reported alcohol consumption, age-standardised percentage by deprivation
fifth, females aged 16+, Wales, 2011-12
42
Figure 37
Mortality by deprivation fifth, European age-standardised rate (EASR) per
100,000*, all ages, Wales, 2003-05 to 2010-12
44
Figure 38
Hospital admissions by deprivation fifth (person-based), European agestandardised rate (EASR) per 100,000, all ages, Wales, financial years 2003/042012/13
46
3
rate
per
100,000,
Alcohol and health in Wales 2014: Wales profile
Foreword
Every week in Wales alcohol results in 29 deaths; around 1 in 20 of all deaths. These are not just
bar brawls and overdoses resulting from binge drinking. In fact, most result from long-term
drinking and its role in increasing risks of diseases such as cancer and cardiovascular disease.
Tragically, many of those affected by alcohol are unaware of how it is harming their health until
the disease process is well established.
The impact of alcohol on health also creates enormous pressures on our health systems. Every
week our hospitals handle as many as 1,000 admissions related to alcohol, increasing strains on
already stretched services. Such admissions are only the tip of an iceberg which includes many
more presentations at emergency departments, ambulance requests and GP appointments all
resulting from alcohol. The cost of alcohol to Wales extends even further through alcohol related
crime, domestic violence, abuse and neglect where individuals’ well-being, property and
sometimes families are harmed through the misuse of alcohol by others. Alcohol related long-term
disability and lost productivity at work mean that it is not just individuals but also the economy
that suffers from an unhealthy relationship with alcohol.
Our problems with alcohol in Wales must be tackled early in life. A greater proportion of children
drink alcohol in Wales than in England, Scotland or the Republic of Ireland. The normalisation of
drinking at a young age, particularly when associated with binge drinking, is especially harmful
and is linked with poor educational performance, sexual health problems in early life and sets the
pattern for harmful drinking behaviour in adulthood. Alcohol also hurts the poorest the most.
While this report shows that no parts of Wales are untouched by alcohol, its impact is greatest in
our most deprived communities which suffer substantially higher levels of both alcohol related
disease and deaths.
We can change our relationship with alcohol. The Welsh Government is taking forward a Minimum
Unit Price for alcohol; a targeted measure intended to reduce alcohol consumption in the heaviest
drinkers who favour very cheap alcohol but leave the moderate drinker with little or no changes to
what they pay for their drinks. Such measures have successfully improved health elsewhere and
can do the same in Wales. However, we also need to empower individuals in Wales to make the
right choices about their own drinking. Too many drinkers fail to recognise how even moderate
drinking can increase their risks of developing diseases such as cancer. Public health professionals
must rise to the challenge of informing the public about these risks
in an environment dominated by advertising intent on increasing
consumption of their products. Our experience with tobacco
suggests that sustained and population wide messages about
harms were only possible once legislation stipulated prominent
health information on all advertisements and products. The risks
related to alcohol use are now clear - what is needed is the policy
to allow them to be communicated at scale to the public.
The Welsh Government’s 10 year substance misuse strategy,
Working Together to Reduce Harm (2008-18), is clear that a multipronged approach involving individuals, the health service, the
alcohol industry and government is required if we are to make
significant progress in tackling this public health challenge. I
congratulate my colleagues at the Public Health Wales Observatory
for producing this comprehensive report which sets out the scale of
the challenges we face if we are to prevent the harms related to
alcohol consumption across Wales. It provides a clear indication
that while we are making progress much more is still to be done if
we want to reduce the avoidable harms that alcohol causes to
families, business and communities across Wales.
4
Professor Mark Bellis
Director of Policy,
Research and Development
Public Health Wales
Alcohol and health in Wales 2014: Wales profile
Key Messages
Alcohol remains a major cause of death and illness in Wales. Around 1,500 deaths in
Wales are attributable to alcohol each year (4.9% of all deaths).
Drinking in children and young people remains a concern with 17% of males and 14%
of females aged 11-16 in Wales drinking alcohol at least once a week. This is higher
than in Scotland, Ireland, England and the survey average.
Adult consumption has fallen slightly in Wales since 2008 for all three measures:
drinking above guidelines, heavy (binge) drinking and very heavy drinking. This mirrors
UK sales data which has fallen since its peak in 2004.
Adults under 45 now drink less. Drinking above guidelines, heavy and very heavy
drinking have decreased in adults aged under 45, particularly in males and the
youngest females, and increased or stayed the same for adults aged 45 and over.
Whilst a reduction in average drinking is good news, it is not uniform across the age
groups and masks persistent or increased drinking in over 45 year olds.
More young males aged 16-24 and more older females aged 75+ have abstained from
alcohol in the past year, whilst the percentage in the other age groups has largely
remained the same.
The comparison with UK alcohol sales data suggests that surveys including the Welsh
Health Survey underestimate alcohol consumption, and may still only represent 60% of
the alcohol sold.
The hospital admission rate in children and young people under 18 for alcohol-specific
conditions has been decreasing steadily for several years, but around 410 young people
were still admitted per year in the most recent period.
The rate of individuals admitted to hospital for alcohol-specific conditions has remained
static since 2006/07. The rate of alcohol-attributable admissions (narrow) has
decreased slowly since 2006/07, but in contrast, the rate for the broad measure has
increased steadily over time.
Both the alcohol-specific hospital admission rates and alcohol-attributable admission
rates are lower in Wales than in England for males, but more similar for females.
The months of life lost in Wales with 13.5 months for males and 6.5 months for
females was higher than in England with 11.5 months for males and 5.4 months for
females.
There were around 250 alcohol-specific deaths in males and 140 in females per year in
Wales, based on the period 2010-12, and alcoholic liver disease accounts for the
majority of these. Mortality rates are higher in Wales than in England.
Mortality and hospital admissions are strongly related to deprivation, where rates in the
most deprived areas are much higher than in least deprived areas. In contrast selfreported heavy and very heavy drinking is very similar between the deprivation fifths.
There is no sign of improvement in the gap in mortality between the most deprived and
the least deprived areas over time.
Inequalities in alcohol-specific and alcohol-attributable (narrow) hospital admissions
have slightly narrowed over time, but slightly widened for the alcohol-attributable
(broad) admissions.
5
Alcohol and health in Wales 2014: Wales profile
1 Introduction
Alcohol is deeply engrained in the culture of Wales, the UK and many other countries. Many
people enjoy alcoholic drinks in moderation but alcohol use can lead to significant harm to the
individual, their families and society. The harmful use of alcohol ranks among the top five risk
factors for disease, disability and death throughout the world. Alcohol is a causal factor in
more than 200 disease and injury conditions and accounts for an estimated 5.9% of all
deaths worldwide1. Alcohol use and its consequences remain a major public challenge in
Wales, the UK and elsewhere.
The Welsh Government’s substance misuse strategy Working together to reduce harm was
published in 2008 and in its 10 year plan sets out a national programme for tackling and
reducing the harms associated with alcohol, drugs and other substances2. The Welsh
Government publishes annual reports on progress made to implement the strategy2 and data
from our publication has been provided for the forthcoming 2014 annual update.
This new publication provides the latest update to our report A profile of alcohol and health in
Wales published five years ago3. Public Health England have recently published revised
methods for mortality and hospital admission indicators4 following a review of alcoholattributable fractions5 for each condition linked to alcohol. We have implemented these
methods using Wales data, providing a comprehensive and extended set of indicators for
Wales, allowing a comparison to England.
This Wales report provides further analysis with interpretation on alcohol consumption
patterns, hospital admissions, mortality and socio-economic inequalities. The publication
contains an extensive local component to support local action with seven health board
summary documents, each with data provided at the lowest geographical level possible. The
main indicator data is published in interactive data files and details of definitions, methods
and caveats as well as guidance on interpretation can be found in the technical guide.
6
Alcohol and health in Wales 2014: Wales profile
2
Definition of units and
drinking patterns
Alcohol units
In the UK alcoholic drinks are measured in units where one unit corresponds to approximately
8 grams (g) or 10 millilitres (ml) of pure alcohol. The number of units in a drink depends on
the size of the drink and its alcohol strength (Table 1).
The number of units in a drink is calculated by multiplying the amount in millilitres (ml) by its
strength (Alcohol by Volume, ABV which is measured as a percentage) and dividing the result
by 1,000. For example, to work out the number of units in a pint (568ml) of strong lager
(ABV 5.2%):
5.2 (%) x 568 (ml) ÷ 1,000 = 2.95 units
Further information and a unit calculator can be found on the NHS choices website on alcohol
units6.
Table 1 Number of units in common alcoholic drinks
Drink (and unit strength (ABV))
Quantity
Units
Small glass red/white/rosé wine (12%)
125 ml
1.5 units
Standard glass red/white/rosé wine (12%)
175 ml
2.1 units
Large glass red/white/rosé wine (12%)
250 ml
3 units
Pint of lower-strength lager/beer/cider (3.6%)
Pint (568 ml)
2 units
Pint of higher-strength lager/beer/cider (5.2%)
Pint (568 ml)
3 units
Bottle of lager/beer/cider (5%)
330 ml
1.7 units
Can of lager/beer/cider (4.5%)
440 ml
2 units
Alco pop (5.5%)
275 ml
1.5 units
Single small shot of spirits (Gin, rum, vodka,
whisky, tequila, sambuca) (40%)
25 ml
1 unit
Source: NHS Live Well website6
Drinking guidelines
Current NHS guidelines state that men should not regularly drink more than 3 to 4 units per
day and women not more than 2 to 3 units per day6. ‘Regularly’ is defined as drinking this
amount every day or most days of the week. It also recommends taking a break from alcohol
for 48 hours after a heavy drinking session6.
Data on drinking patterns comes from surveys which tend to collect information on the
amount drunk on the heaviest drinking day in the past week rather than the usual
consumption. This may be very different and so has implications for the interpretation of
survey data.
7
Alcohol and health in Wales 2014: Wales profile
Drinking above guidelines
Drinking above guidelines from surveys is usually measured as men drinking more than 4
units and women drinking more than 3 units on their heaviest drinking day in the previous
week.
Heavy (binge) drinking
The term ‘binge drinking’ usually refers to drinking lots of alcohol in a short period of time or
drinking to get drunk6. In surveys, it is usually measured as drinking more than twice the
daily guidelines, which for men is more than 8 units and women more than 6 units on their
heaviest drinking day in the previous week. In Welsh Health Survey (WHS) publications this
measure has been referred to as either ‘heavy drinking’ or ‘binge drinking’. To avoid any
confusion with the new ‘very heavy drinking’ category, both terms ‘heavy (binge) drinking’
are used throughout this report for Wales or Great Britain data (see Table 2).
Very heavy drinking
The new measure ‘very heavy drinking’ has no formal definition, but has recently been used
in Office for National Statistics (ONS) publications and referred to as men drinking more than
12 units and women drinking more than 9 units on their heaviest drinking day in the previous
week. This definition, listed in Table 2, is used throughout this report where survey results
are reported for Wales and Great Britain.
Abstainers
Abstainers in this report refer to WHS respondents who report not having drunk alcohol in the
past 12 months (Table 2). It includes those abstainers who have not consumed alcohol during
this fixed period of time regardless of whether they have drunk alcohol before this 12 month
period.
Table 2 Overview of definitions used
Category
Abstainer
Men
Not drinking at all in the previous
12 months
Women
Not drinking at all
previous 12 months
Drinking above guidelines
Drinking more than 3-4 units on
the heaviest drinking day in the
previous week
Drinking more than 2-3 units on
the heaviest drinking day in the
previous week
Heavy (binge) drinking
Drinking more than 8 units on the
heaviest drinking day in the
previous week
Drinking more than 6 units on
the heaviest drinking day in the
previous week
Very heavy drinking
Drinking more than 12 units on
the heaviest drinking day in the
previous week
Drinking more than 9 units on
the heaviest drinking day in the
previous week
8
in
the
Alcohol and health in Wales 2014: Wales profile
3 Patterns of drinking
Drinking patterns in this report come from surveys which ask a sample of the population
about their drinking habits. These results are then used as an estimate of the drinking
patterns of the population. The main source is the annual Welsh Health Survey, which asks
around 15,000 residents in Wales about their health and lifestyle choices.
Another survey is the Opinions and Lifestyle Survey (OPN), run by the ONS (formerly called
the General Household Survey), which covers adults in England, Scotland and Wales, and
allows comparisons across Great Britain.
Further information on these surveys can be found in the technical guide that accompanies
this publication (see inside cover). Additionally, socio-economic patterns of alcohol
consumption are discussed in Section 6.
3.1
Drinking in children and young people
Drinking alcohol can damage a child’s health. It is recommended that children do not drink
alcohol, however if children do drink alcohol, they should not do so until they are at least 15
years old7.
Drinking in children and young people remains of concern with 17%
of males and 14% of females aged 11-16 in Wales drinking alcohol
at least once a week in 2009/10, the most recent data available
(Figure 1). The Health Behaviour in School-aged Children survey
(HBSC) is conducted in 40 countries, including Wales, England and
Scotland. The percentages of young people drinking alcohol at least
once a week are higher in Wales than in Scotland, Ireland, England
and the survey average for both males and females. The
percentage in Wales is twice that of Ireland. At the health board
level, the higher percentages of 11-16 year olds are in Cwm Taf
UHB (males and females both 18%) and Aneurin Bevan UHB (males
21% and females 20%). The largest difference between males and
females is in Powys with twice the percentage of males (20%)
reporting drinking than females (10%), although this may be the
result of bias associated with small numbers of survey respondents.
9
“
…Drinking in children
and young people
remains a concern
with 17% of males
and 14% of females
aged 11-16 in Wales
drinking alcohol at
least once a week.
This is higher than in
Scotland, Ireland and
England and the
survey average ...”
Alcohol and health in Wales 2014: Wales profile
Figure 1
Percentage of persons who reported drinking alcohol at least once a week,
boys and girls aged 11-16*, 2009/10
Produced by Public Health Wales Observatory, using HBSC (WG)
Lowest (Iceland)
Highest (Czech Republic)
25
19
HBSC average
14
9
Scotland
15
13
Ireland
Boys
4
2
Girls
8
5
England
15
11
Wales
17
14
Betsi Cadwaladr UHB
15
18
Powys tHB
20
10
Hywel Dda UHB
16
12
ABM UHB
16
16
Cardiff & Vale UHB
13
10
Cwm Taf UHB
18
18
Aneurin Bevan UHB
21
20
*Country level data only includes ages 11, 13 and 15
3.2
Consumption in adults in Wales
Survey results are essential to identify characteristics of those drinking, and patterns are
reported by sex and age-group. WHS respondents were asked questions on their drinking
habits including how often they drank, if they are a non-drinker and, if they are a drinker,
how much of each type of alcohol they drank on the heaviest day in the last week. The
questions changed from 2008 onwards and the data is not comparable to earlier results.
Trends or comparisons over time are shown wherever possible.
Please note that surveys are known to underestimate drinking in the population, as shown in
comparisons to alcohol sales data, see section 3.4. They are, however, useful measures for
comparing areas or groups of people.
10
Alcohol and health in Wales 2014: Wales profile
3.2.1
Frequency of alcohol consumption in Wales
Males drink more frequently than females. Around 46% of males reported either drinking
once or twice a week or 3 to 4 days a week on average for the past 12 months compared to
35% for females. The percentage of adults who have not drunk alcohol in the past 12 months
is higher for females than males (Figure 2), also measured as abstainers (section 3.2.3).
Figure 2
Reported average frequency of drinking alcohol, percentages, males
and females aged 16+, Wales, 2012
Produced by Public Health Wales Observatory, using WHS (WG)
Males
10
15
Never
3.2.2
7
13
9
13
15
17
Once or Once every Once or
twice a
couple of
twice a
year
months
month
32
25
14
10
Fem ales
4
3
9
5
Once or
Three or Five or six Almost
twice a four days a days a
every day
week
week
week
Alcohol consumption trend in Wales
More males than females report drinking above guidelines, heavy
(binge) drinking or very heavy drinking (see Table 2 for definitions).
The percentages of adults have slightly decreased over time for all
three measures although the differences between the earliest and
most recent period are unlikely to be statistically significant (Figure
3).
11
“
…Adult consumption
has fallen slightly
across Wales since
2008 for all three
measures. This
mirrors UK sales data
which has fallen since
its peak in 2004 ...”
Alcohol and health in Wales 2014: Wales profile
Figure 3
Percentage of adults who reported drinking on the heaviest drinking day in
the past week, age-standardised*, males and females aged 16+, Wales,
2008-2013
Produced by Public Health Wales Observatory, using WHS (WG)
Males
95% confidence interval
Fem ales
Drinking above guidelines
51
51
51
49
38
38
37
48
48
38
37
36
Heavy (binge) drinking
34
33
33
33
31
32
22
21
21
22
21
20
Very heavy drinking
20
19
19
19
17
18
13
12
11
12
12
11
2008
2009
2010
2011
2012
2013
* Using aggregated weightings from the 2013 European Standard Population
3.2.3
Abstainers
The percentage of adults who report not having drunk alcohol in
the past 12 months differs between age groups (Figure 4). The
percentage of abstainers is generally higher in females compared
to males and is lowest in the 25-64 age range. The percentages
have largely remained the same or increased slightly across the
age groups between 2008-09 and 2011-12. The greatest
increases are in females aged 75+ years, and males aged 16-24
years where the percentage has increased from 11% in 2008-09
to 14% in 2011-12 (Figure 4).
12
“
… More young males
aged 16-24 and more
older females aged 75+
have abstained from
alcohol in the past year,
whilst percentage in the
other groups has largely
remained the same ... ”
Alcohol and health in Wales 2014: Wales profile
Figure 4
Percentage of adults who reported abstaining from alcohol, males and females aged
16+, Wales, 2008-9 and 2011-12
Produced by Public Health Wales Observatory, using Welsh Health Survey (WG)
95% confidence
interval
75+
Males 2008-9
Females 2008-9
Males 2011-12
Fem ales 2011-12
18
65-74
11
18
38
11
22
55-64
8
8
15
45-54
8
8
11
35-44
7
7
10
25-34
7
7
9
16-24
3.2.4
11
14
35
23
15
11
8
9
12
11
Drinking patterns by age group and over time
Drinking patterns and the change over time vary by age group and
measure (Figure 5, 6, 7).
Drinking above guidelines
Drinking above guidelines has decreased substantially in the
younger age groups (16-24, 25-34 and 35-44), most notably in
males (Figure 5), however, there was no change for females aged
25-34. For example, the percentage in young males aged 16-24
has fallen from 51% to 42% and in young females from 46% to
39%. In those aged 45 and above the percentages have either
remained the same or have slightly increased between the time
periods, with the 45-54 year old age group having the highest
percentages (Figure 5).
13
“
… Adults under 45 now
drink less. Drinking above
guidelines, heavy and very
heavy drinking have
decreased in adults aged
under 45, particularly in
males and the youngest
females, and increased or
stayed the same for adults
aged 45 or over ... ”
Alcohol and health in Wales 2014: Wales profile
Figure 5
Percentage of adults who reported drinking above guidelines, males and females aged
16+, Wales, 2008-09 and 2011-12
Produced by Public Health Wales Observatory, using Welsh Health Survey (WG)
95% confidence
interval
75+
22
65-74
42
55-64
45-54
35-44
25-34
16-24
54
57
61
59
51
Males 2008-09
Females 2008-09
Males 2011-12
Fem ales 2011-12
23
10
9
44
24
55
38
58
48
47
55
47
49
52
43
42
39
22
36
43
46
Heavy (binge) drinking
The pattern of change in heavy (binge) drinking is similar to that of drinking above guidelines.
Percentages have significantly decreased in males aged under 45 and females aged 16-24,
and increased or stayed the same for those aged 45 and over (Figure 6). Most notably, the
percentage in females aged 55-64 has increased from 14% to 18% and in males aged 65-74
from 18% to 21%.
Very heavy drinking
The decrease in very heavy drinking for males aged under 45 and females aged 16-24 is very
similar to heavy (binge) drinking. Very heavy drinking has slightly increased or stayed the
same for males and females aged 45 and over (Figure 7). Males aged 25-34 and females
aged 16-24 still report the highest percentages but the differences between the age groups
are becoming smaller. Whilst the reduction in average drinking in Wales is good news, it is
not uniform across the age groups and masks persistent or increased drinking in males and
females over 45.
14
Alcohol and health in Wales 2014: Wales profile
Figure 6
Percentage of adults who reported heavy (binge) drinking, males and females aged
16+, Wales, 2008-09 and 2011-12
Produced by Public Health Wales Observatory, using Welsh Health Survey (WG)
95% confidence
interval
Males 2008-09
Females 2008-09
Males 2011-12
Fem ales 2011-12
75+
8
65-74
6
18
55-64
33
45-54
39
2
21
8
33
18
40
27
6
14
26
35-44
45
39
29
30
25-34
45
39
30
30
32
29
16-24
39
35
Figure 7
Percentage of adults who reported very heavy drinking, males and females aged 16+,
Wales, 2008-09 and 2011-12
Produced by Public Health Wales Observatory, using Welsh Health Survey (WG)
95% confidence
interval
Males 2008-09
Females 2008-09
Males 2011-12
Fem ales 2011-12
75+
3
65-74
7
55-64
14
45-54
21
35-44
25-34
16-24
27
31
29
1
8
2
13
8
22
14
24
16
26
19
24
21
6
13
17
20
25
15
Alcohol and health in Wales 2014: Wales profile
3.2.5
Geographical variation in very heavy drinking
The percentage of adults aged 16 and over who report very heavy drinking varies between
health boards. The percentage for persons in Powys tHB (12.6%) and Hywel Dda UHB
(13.3%) is statistically significantly lower than Wales, whilst Cwm Taf UHB (17.6%) is
statistically significantly higher (Figure 8).
Figure 8
Percentage of adults reporting very heavy drinking (males over 12 units,
females over 9 units) on the heaviest drinking day in the past week, agestandardised percentage*, persons, Wales health boards, 2011 -2012
Produced by Public Health Wales Observatory, using Welsh Health Survey (WG)
95% confidence interval
Betsi Cadwaladr UHB
14.6
Powys tHB
12.6
Hywel Dda UHB
13.3
ABM UHB
15.3
Cardiff & Vale UHB
14.4
Cwm Taf UHB
17.6
Aneurin Bevan UHB
15.9
Wales = 14.9
* Using aggregated weightings from the 2013 European Standard Population
The percentages vary considerably at sub-local authority level (upper super output areas)
ranging from 8.4 to 21.8 (see map) (Figure 9). The map for each health board separately is
included in the Health Board summary document (see inside cover). The highest percentages
of adults reporting very heavy drinking are in USOAs within Rhondda Cynon Taf, Caerphilly,
Bridgend and Blaenau Gwent.
16
Alcohol and health in Wales 2014: Wales profile
Figure 9
17
Alcohol and health in Wales 2014: Wales profile
3.3
Drinking patterns and trends in Great Britain
The percentages of drinking for the three definitions in Wales are higher than in England, but
lower than in Scotland (Figure 10). According to the Opinions and Lifestyle Survey (OPN),
27% of adults in Wales reported heavy (binge) drinking on the heaviest drinking day in the
past week, higher than in England with 25% and lower than in Scotland with 33%. In Wales,
16% reported very heavy drinking compared to 13% in England and 19% in Scotland.
The data can differ between surveys. The OPN reports that 55% of adults drink above
guidelines in Wales whereas the WHS reports 42% for 2012. This is likely to be due to a
relatively small sample size for Wales in the OPN (unweighted sample of 370 adults) and a
different survey design. The WHS percentages of those reporting heavy (binge) drinking
(26%) or very heavy drinking (14%) are much closer.
Figure 10
Percentage of adults who reported drinking alcohol on the heaviest
drinking day in the past week, persons aged 16+, Great Britain, 2012
Produced by Public Health Wales Observatory, using Opinions and Lifestyle Survey (ONS)
Drinking above guidelines
51
25
England
3.4
13
Heavy (binge) drinking
Very heavy drinking
55
63
27
16
Wales
33
19
Scotland
Comparison of survey results and sales data
The average sale of alcohol per week per person in the UK has fallen
from 22.3 units of alcohol at its peak in 2004 to 18.6 units in 2012
(Figure 11). Sales data are derived from HM Revenue and Customs
duty charged on alcohol produced or processed in the UK, or brought
into the UK for consumption but not necessarily consumed. Sales
figures are not available for Wales separately as it is not possible to
identify where in the UK the product was manufactured or released
for consumption.
Survey data is known to underestimate consumption in the
population and only capture around 60% of the true figures8,9. In
2010, the average weekly alcohol consumption per adult in Great
Britain (General Lifestyle Survey (GLF) survey data) was 11.5 units,
much lower than UK sales data of 19.6 units in 2010.
18
“
… The comparison with
UK alcohol sales data
suggests that surveys
including the Welsh
Health Survey
underestimate alcohol
consumption, and may
still only represent 60%
of the alcohol sold ... ”
Alcohol and health in Wales 2014: Wales profile
The estimated weekly consumption for adults in Wales from the WHS was 10.7 units in 2007
showing a similar gap to sales data (21.5 units). More recent figures for Wales cannot be
produced robustly because of question changes in 2008. Additionally, the estimates since
2008 are higher as they take into account the increasing strength of drinks. The WHS sales
data gap is therefore now expected to be more similar to the GLF.
The GLF survey data does not include Northern Ireland and includes persons aged 16+
whereas the sales figures are for persons aged 15+ and it is unclear how accurate sales
figures are as an estimate of consumption. People tend to understate the amount they drink,
partly unintentionally, as they genuinely forget some of the drinking or because it is difficult
to estimate units due to different strengths and serving sizes8. Some may deliberately
understate the amount drunk, because they feel that it is not socially acceptable, and very
heavy drinkers are underrepresented in surveys8. Drinking reported for the previous week
may also not represent usual drinking patterns or include special events. While there may be
underestimation in the survey results, survey data is still essential to identify characteristics
of those drinking, for example comparisons between age groups, males and females, or
frequency of drinking which sales data cannot provide.
Figure 11
Comparison of sales and survey consumption data, weekly alcohol units
per person, 1996-2012
Produced by Public Health Wales Observatory, using HMRC, BBPA, GLF (ONS) & WHS (WG)
UK sales data, aged 15+, United Kingdom
GLF survey data, aged 16+, Great Britain*
WHS data, aged 16+, estim ate before question change, Wales**
25
20
15
10
5
0
1996
1998
2000
2002
2004
2006
2008
2010
*GLF weekly consumption data unavailable for 2007
** 03/04, 04/05, 05/06 (charted as 2004, 2005, 2006), 2007 WHS figure from WAG
19
2012
Alcohol and health in Wales 2014: Wales profile
3.5
International comparison of consumption
Drinking patterns can vary considerably from country to country. The recent report by the
World Health Organisation1 shows that globally people aged 15 and over drink on average 6.2
litres of pure alcohol per year. In the UK the consumption per person aged 15 and over was
an estimated 11.6 litres of pure alcohol in 20101 (Figure 12). This was compared to other
selected Western countries and is lower in the UK than in Australia or France with 12.2 litres,
but higher than the Netherlands with 9.9 litres or Italy with 6.7 litres of alcohol (Figure 12).
The sources for the data shown differ between countries but mostly use official data sources
such as tax or sales rather than survey data1. These have been adjusted to exclude tourists
and also include an estimate of unrecorded consumption.
Figure 12
International alcohol consumption* for selected countries,
litres of pure alcohol per person aged 15+, 2010
Produced by Public Health Wales Observatory, using WHO (2014)
95% confidence interval
Portugal
12.9
Finland
12.3
Australia
12.2
France
12.2
Luxem bourg
11.9
Ireland
11.9
Germ any
11.8
UK
11.6
Denm ark
11.4
Spain
11.2
Belgium
Austria
11.0
10.3
Netherlands
9.9
Sweden
9.2
USA
9.2
Italy
6.7
* Excludes tourist consumption and includes estimated unrecorded consumption
20
Alcohol and health in Wales 2014: Wales profile
4 Hospital admissions
Hospital admission relating to alcohol is an important indicator of morbidity and the effect of
alcohol misuse on the health of the population. In this report we focus on indicators
measuring individuals admitted to hospital at least once a year (person-based) for three
different definitions, following revised methods used in England5. These are very briefly
explained in table 3, but for further details please see the technical guide.
Table 3 Overview of definitions for person-based hospital admission indicators
Admissions indicator
Brief definition
Alcohol-specific hospital admissions
Measures individuals admitted with alcohol-specific
(wholly attributable) conditions, either in the primary
diagnosis (main reason) or in secondary diagnoses. This
list of conditions has been revised.
Alcohol-attributable hospital admissions
(person-based, narrow)
Measures individuals admitted with alcohol-attributable
conditions (either wholly or partly attributable to alcohol)
at least once a year, either as the primary diagnosis
(main reason) or an external cause (e.g. injuries) as a
secondary diagnosis, whichever is most linked to alcohol
(highest fraction). This is a new measure.
Alcohol-attributable hospital admissions
(person-based, broad)
Measures individuals admitted with alcohol-attributable
conditions (either wholly or in part attributable to alcohol)
at least once a year, either as the primary diagnosis
(main reason) or a secondary diagnosis, whichever is
most linked to alcohol (highest fraction). This measure
has been revised.
The alcohol-attributable admissions use attributable fractions for each condition, which is
based on evidence in academic literature for a relationship between alcohol and the condition
well as the prevalence of drinking. We use the same fractions produced for England in our
calculation of the indicator data and assumptions are made that these equally apply to Wales
and, for example, each local authority in Wales. They are therefore considered estimates and
need to be interpreted with care.
Episode-based alcohol-attributable indicators have also been produced and are published in
interactive data files (see inside cover), but not included in this report. They measure
admitting episodes where each individual could be counted more than once a year. These
relate more to the burden on the health service whilst this report focusses on the effect of
alcohol on the population and therefore on person-based analyses.
Socio-economic patterns in hospital admissions are examined in section 6. Further details on
the indicators and methods are contained in the technical guide (see inside cover).
21
Alcohol and health in Wales 2014: Wales profile
4.1
The scale and breakdown by condition
Alcohol-specific hospital admissions
Around 6,700 males and 3,500 females have been admitted to hospital with an alcoholspecific condition (Figure 13). Each individual is only counted once in this definition for one
alcohol-specific condition (Figure 13) but this may not be the only alcohol-specific condition
noted or the main reason for admission. Nevertheless, it gives an indication of the
contribution of each condition to the measure. Over half of females and over two thirds of
males had a diagnosis of mental and behavioural disorder, including for example acute
intoxication. Just over 1 in 5 females had a diagnosis of ethanol poisoning, twice the
proportion in males. This measure underestimates admissions due to alcohol as only those
entirely down to alcohol are included.
Figure 13
Ethanol
poisoning
Alcoholic liver
disease
21.2
15.1
6,700
Males
Alcohol-specific
hospital admissions
2012/13
8.8
2.8
2.5
Ethanol
poisoning
3,500
15.2
Females
Other
Chronic
& acute
pancreatitis
Alcohol-specific
hospital admissions
2012/13
Alcoholic liver
disease
3.3
Other
1.8
Chronic
& acute
pancreatitis
70.8
Mental &
behavioural
disorders
58.5
Mental &
behavioural
disorders
Alcohol-attributable hospital admissions (narrow)
An estimated 8,100 males and 4,600 females were admitted to hospital with an alcoholattributable condition using the narrow definition (Figure 14). This includes individuals with an
alcohol-attributable condition as the main reason for admission (primary diagnosis) or an
external cause (e.g. injury) as a secondary diagnosis. This measure may also understimate
admissions due to alcohol as it is not considering all secondary diagnoses, but it is likely to be
more comparable over time.
Around a third of these hospital admissions were for conditions that are wholly attributable to
alcohol (32.1% in males and 38.7% in females) and the main reason (primary diagnosis) for
admission. These therefore include some of the individuals in the other definition of alcoholspecific admissions but not all, as those with only an alcohol-specific secondary diagnosis are
not included in this narrow definition. Almost a third of males and 1 in 5 females were
admitted for unintentional injuries. Please note that cardiovascular disease accounted for
9.6% of narrow hospital admissions in males (8.5% of females).
22
Alcohol and health in Wales 2014: Wales profile
Figure 14
Cancer
9.5
Respiratory
7.5
9.6
8,100
8.6
Other
6.2
Cardiovascular
Unintentional
injuries
Other
Cardiovascular
8.5
Respiratory
5.5
Intentional
injuries
Cancer
4.1
13.4
4,600
3.8
Males
Females
Alcohol-attributable
admissions (narrow)
2012/13
32.1
31.0
Alcohol-attributable
admissions (narrow)
2012/13
Unintentional
injuries
Wholly
attributable
21.5
Pregnancy &
childbirth
38.7
Wholly
attributable
Alcohol-attributable hospital admissions (broad)
An estimated 21,700 males and 12,300 females were admitted to hospital with an alcoholattributable condition using the broad definition (Figure 15). This includes individuals with at
least one diagnosis that relates to an alcohol-attributable condition, but not necessarily the
main reason for attendance. It considers the admission with the condition with the strongest
link to alcohol i.e. the largest fraction. Coding of secondary diagnoses may vary locally and so
this measure may vary in its ability of picking up all individuals with an admission related to
alcohol.
Almost half of the admissions had a diagnosis of cardiovascular disease, where a relatively
high proportion are linked to alcohol and which is a common diagnosis. Any change to
hospital admissions for cardiovascular disease with a link to alcohol is therefore strongly
reflected in the pattern of alcohol-attributable hospital admissions using the broad definition.
Please note that these groups only contain certain conditions that are linked to alcohol, for
example, only some cardiovascular diseases. For a list of these please see the technical
guide. Risk factors for cardiovascular disease are wide ranging, and any pattern in
cardiovascular disease could be the result of risk factors other than alcohol consumption. For
these reasons this measure needs to be interpreted with caution, particularly when
considering a trend (see Section 4.2). Please also see the technical guide for further details.
Just under a third of individuals were admitted with conditions that are wholly attributable to
alcohol consumption. This measure contains all the individuals in the definition of alcoholspecific hospital admission.
23
Alcohol and health in Wales 2014: Wales profile
Figure 15
Nervous
system
Nervous
Cancer Other
system
Digestive
Unintentional
injuries
4.3
Cancer
3.2 3.0
1.7
5.2
6.1
21,700
Wholly
attributable
4.2
30.9
4.0 2.1
5.5
Unintentional
injuries
8.0
Other Digestive
12,300
Males
Females
Alcohol-attributable
admissions (broad)
2012/13
Alcohol-attributable
admissions (broad)
2012/13
48.9
Cardiovascular
Wholly
attributable
28.4
48.7
Cardiovascular
Hospital admission trends over time
The rate of individuals admitted to hospital for alcohol-specific
(wholly attributable) conditions increased during the early years of
the period shown but has remained stable since 2006/07. In our
previous publication A profile of alcohol and health in Wales3 we
recommended that the alcohol-attributable trend (now called broad)
should be considered alongside the alcohol-specific measure to
ensure that a trend over time is not merely reflecting a change in
trend in one of the attributable conditions, rather than the impact of
alcohol in general over time. This is more complex with two alcoholattributable definitions showing different trends. The rate of alcoholattributable admissions (narrow) has slowly decreased since
2006/07. This is in contrast to the broad measure which has steadily
increased over time (Figure 16).
“
… The rate of individuals
admitted to hospital for
alcohol specific conditions has remained
static since 2006/07,
but in contrast the rate
for the broad measure
has increased steadily
over time ... ”
The broad admission indicator is heavily influenced by patterns in cardiovascular disease
linked to alcohol (Section 4.3) with half of the individuals counted with cardiovascular disease
in either the primary or secondary diagnosis. Although not shown here, the trend in
admissions with cardiovascular disease linked to alcohol has been rising over time and so this
is reflected in the trend in the broad alcohol-attributable admissions. It may not, however,
reflect the impact of alcohol on cardiovascular disease as there are many risk factors other
than alcohol and treatment may change over time. Also, coding of secondary diagnosis may
vary over time as well as between areas which needs to be taken into account when
interpreting the broad alcohol-attributable admission measure. The narrow admissions
measure is not affected in the same way and is therefore a more suitable measure for
tracking trends over time instead of the broad admission indicator. It is still advisable,
however, to consider the alcohol-specific trend alongside the narrow measure.
The rate in males is consistently higher than the female rate, for all measures.
24
Alcohol and health in Wales 2014: Wales profile
Figure 16
Hospital admissions (person based), European age-standardised rate
(EASR) per 100,000*, males and females, all ages, Wales, financial
years 2003/04-2012/13
Produced by Public Health Wales Observatory, using fractions (PHE), PEDW (NWIS) &
MYE (ONS)
95% confidence interval
Fem ales
Males
456
450
454
461
457
457
222
234
234
229
229
238
237
228
622
619
611
596
589
322
321
316
317
316
297
1,450
1,452
1,496
1,484
1,532
1,530
1,534
707
739
740
759
750
783
781
783
2009/10
2010/11
2011/12
2012/13
197
458
2008/09
199
431
2007/08
404
2006/07
393
2005/06
Alcohol-specific
Alcohol-attributable (narrow)
584
583
607
630
298
293
316
328
562
Alcohol-attributable (broad)
601
638
2004/05
1,275
2003/04
1,199
1,392
*Using 2013 European Standard Population
Hospital admissions aged under 18
The rate for children and young people aged under 18 admitted for
alcohol-specific (wholly attributable) conditions has statistically
significantly decreased since the period 2005-07 in both males and
females (Figure 17). This downward trend is good news, but around
410 young people were still admitted per year in the most recent
period. People aged under 18 are the only group with a higher rate
in females than in males. This rate in all young people under 18 in
Wales is consistently higher than in England although the difference
is slightly narrowing over time (Figure 22 in Section 4.4).
25
“
…The hospital admission
rate in children and young
people under 18 for
alcohol- specific conditions
has been decreasing
steadily for several years,
but around 410 young
people are still admitted
per year in the most
recent period... ”
Alcohol and health in Wales 2014: Wales profile
Figure 17
Alcohol-specific hospital admissions, 3-year rolling crude rate per 100,000,
males and females aged under 18, Wales, financial years 2003/04 2012/13
Produced by Public Health Wales Observatory, using PEDW (NWIS) & MYE (ONS)
Males
118
Fem ales
126
95% confidence interval
117
108
103
91
89
95
96
4.3
2004/052006/07
2005/062007/08
78
88
74
2003/042005/06
84
2006/072008/09
2007/082009/10
68
2008/092010/11
62
2009/102011/12
52
2010/112012/13
Geographical variation in hospital admissions
Alcohol-specific hospital admission rates vary locally, amongst health board areas and upper
super output areas. The rate for males in Powys tHB and Hywel Dda UHB is statistically
significantly lower than for Wales, and statistically significantly higher in Aneurin Bevan UHB
and Cwm Taf UHB (Figure 18). The rates for females are statistically significantly lower in
Powys tHB and in Cardiff and Vale UHB than for Wales, and statistically significantly higher in
Betsi Cadwaladr UHB and Aneurin Bevan UHB (Figure 19). Other indicators, although not
shown here, are included in the health board summary documents (see inside cover).
26
Alcohol and health in Wales 2014: Wales profile
Figure 18
Alcohol-specific hospital admissions (person-based), European agestandardised rate per 100,000*, males, all ages, Wales health boards,
financial year 2012/13
Produced by Public Health Wales Observatory, using PEDW (NWIS) & MYE (ONS)
95% confidence interval
Betsi Cadwaladr UHB
461
Powys tHB
297
Hywel Dda UHB
401
ABM UHB
446
Cardiff & Vale UHB
473
Cwm Taf UHB
526
Aneurin Bevan UHB
510
Wales = 457
* Using the 2013 European Standard Population
Figure 19
Alcohol-specific hospital admissions (person-based), European agestandardised rate per 100,000*, females, all ages, Wales health boards,
financial year 2012/13
Produced by Public Health Wales Observatory, using PEDW (NWIS) & MYE (ONS)
95% confidence interval
Betsi Cadwaladr UHB
258
Powys tHB
188
Hywel Dda UHB
219
ABM UHB
209
Cardiff & Vale UHB
195
Cwm Taf UHB
226
Aneurin Bevan UHB
250
Wales = 228
* Using the 2013 European Standard Population
27
Alcohol and health in Wales 2014: Wales profile
Rates for alcohol-specific hospital admissions vary considerably in smaller areas of Wales
(upper super output areas) and range from 165 to 613, see map (Figure 20). A separate map
for each health board is included in the health board summary document (see inside cover).
Figure 20
4.4
Comparison to England and English regions
Both the alcohol-specific hospital admission rates and alcohol-attributable admission rates for
all ages in males are lower in Wales than in England but more similar for females (Figure 21,
23 and 24). However, rates amongst the English regions vary considerably.
Alcohol-specific admission rates range from 344 per 100,000 in the East of England to 740 in
the North West for males (from 172 to 364 respectively for females). The range is narrower
for Wales health board areas, where rates vary from 297 to 526 for males and 188 to 258 for
females (Figure 18 and Figure 19 in Section 4.3).
The patterns are similar for alcohol-attributable admissions although the Wales rate for
alcohol-attributable admissions (narrow) is only slightly lower than England. There are small
differences in the method or data between the England and Wales rates. For example, English
rates only consider patients in English hospitals, whereas the Wales rates include patients
resident in Wales but treated in English hospitals. Also, there are some coding issues where
28
Alcohol and health in Wales 2014: Wales profile
around 2% of hospital spells in Wales in the most recent year do not have a diagnosis coded.
It is not clear how this affects the person-based analysis but it is unlikely to affect the
interpretation of the differences greatly. Please see the technical guide for further details of
the methods.
In contrast to rates for all ages, rates in young people aged under 18 in Wales are
consistently higher than in England (Figure 22), although all are decreasing. The differences
between them have been narrowing but are still substantial in the most recent period of
2010-12 with a crude rate of 65 per 100,000 in Wales compared to 45 per 100,000 in
England.
Figure 21
Alcohol-specific hospital admissions, European agestandardised rate (EASR) per 100,000, all ages, Wales, England
& English regions, financial year 2012/13
Produced by Public Health Wales Observatory, using PEDW (NWIS), MYE (ONS)
& LAPE (PHE)
95% confidence interval
Wales
457
228
England
507
232
North East
702
342
North West
740
364
Yorkshire and the Hum ber
522
244
East Midlands
406
198
West Midlands
472
213
East of England
344
172
London
529
188
South East
376
188
South West
468
238
Males
Fem ales
29
Alcohol and health in Wales 2014: Wales profile
Figure 22
Alcohol-specific hospital admissions, 3-year rolling crude rate per 100,000,
persons aged under 18, Wales & England, financial years 2003/04 2012/13
Produced by Public Health Wales Observatory, using PEDW (NWIS) & MYE (ONS)
Wales
110
106
98
England
95% confidence interval
102
88
79
68
2003/042005/06
2004/052006/07
2005/062007/08
2006/072008/09
63
2007/082009/10
57
2008/092010/11
73
52
2009/102011/12
Figure 23
Alcohol-attributable hospital admissions (person-based,
narrow), European age-standardised rate (EASR) per 100,000,
all ages, Wales, England & English regions, financial year
2012/13
Produced by Public Health Wales Observatory, using PEDW (NWIS), MYE (ONS)
& LAPE (PHE)
95% confidence interval
Wales
562
297
England
589
306
North East
735
388
North West
721
383
Yorkshire and the Hum ber
624
317
East Midlands
568
295
West Midlands
578
304
East of England
492
264
London
557
260
South East
496
267
South West
547
303
Males
Fem ales
30
65
45
2010/112012/13
Alcohol and health in Wales 2014: Wales profile
Figure 24
Alcohol attributable hospital admissions (person-based, broad),
European age-standardised rate (EASR) per 100,000, all ages,
Wales, England & English regions, financial year 2012/13
Produced by Public Health Wales Observatory, using PEDW (NWIS), MYE (ONS)
& LAPE (PHE)
95% confidence interval
Wales
England
North East
North West
Yorkshire and the Hum ber
East Midlands
West Midlands
East of England
London
South East
South West
Males
Fem ales
1,534
783
1,676
832
2,117
1,065
2,025
1,033
1,753
866
1,493
752
1,638
829
1,454
722
1,784
842
1,410
705
1,544
773
31
Alcohol and health in Wales 2014: Wales profile
5 Mortality
5.1
Alcohol-specific and alcohol-attributable mortality
Mortality linked to alcohol can be measured using different definitions. In this report we show
alcohol-specific mortality and alcohol-attributable mortality. These are very briefly explained
in table 4, but for further details please see the technical guide.
Table 4 Overview of definitions for mortality indicators
Mortality indicator
Brief definition
Alcohol-specific mortality
Includes deaths with an underlying cause of death that is
wholly attributable to alcohol e.g. alcoholic liver disease.
Additional causes of death were added in the revision of
methods.
Alcohol-attributable mortality
Includes deaths with an underlying cause of death that is
either wholly or partly attributable to alcohol. Revised
attributable fractions for each cause of death are applied in
the calculation. This wider definition is an estimate based on
fractions assuming they apply equally to Wales.
Public Health England have recently published revised methods for these4. Data using the
revised method is not comparable to previously published data. Alcohol-specific mortality has
been produced instead of alcohol-related mortality (ONS definition) in this updated report as
it is easily comparable to England and English regions. Socio-economic inequalities in
mortality are examined in Section 6.
Around 250 males and 140 females die from an alcohol-specific condition per year in Wales,
based on the period 2010-12 (Figure 25). Alcoholic liver disease accounts for the majority of
these deaths (82% for males, 86% for females) (Figure 25). Alcohol-specific mortality only
includes deaths entirely caused by alcohol, which underestimates the impact of alcohol on
mortality in the population.
32
Alcohol and health in Wales 2014: Wales profile
Figure 25
Mental &
behavioural
disorders
Mental &
behavioural
disorders
9.0
Accidental
poisoning by &
exposure to
5.5
alcohol
250
Males
82.0
3.5
8.2
140
Females
Other
Alcohol-specific
deaths per year
2010-12
86.0
Alcoholic
liver disease
5.3
0.5
Accidental
poisoning by &
exposure to
alcohol
Other
Alcohol-specific
deaths per year
2010-12
Alcoholic
liver disease
Unintentional
injuries
12.7
Intentional
injuries
Respiratory
Cardiovascular
8.9
Respiratory
Digestive
7.3
13.2
6.8
Unintentional
injuries
8.3
990
7.5
Males
25.5
Wholly
attributable
Alcohol-attributable
deaths per year
2010-12
5.1
1.3
5.1
530
Cardiovascular
2.0
0.9
Females
Digestive
Other
Wholly
attributable
26.3
Intentional
injuries
Other
Alcohol-attributable
deaths per year
2010-12
38.4
Cancer
30.7
Cancer
Using the wider definition of alcohol-attributable mortality there were around four times as
many deaths as for alcohol-specific mortality, with 990 male and 530 female deaths per year
in Wales, based on the period 2010-12 (Figure 25). These avoidable deaths are a sizeable
proportion of overall deaths in Wales accounting for 6.6% of all male deaths, and 3.3% of all
female deaths.
Wholly attributable causes account for around one in four of these deaths (26%) in both
males and females. Cancer accounts for 31% of the alcohol-attributable deaths in males and
38% for females shown in the purple colour on the chart. Cardiovascular disease (shown in
darker blue) accounts for over 7% of male and 13% of female deaths.
33
Alcohol and health in Wales 2014: Wales profile
Trends over time
Rates vary slightly over time (Figure 26). Alcohol-specific mortality rates for males were
rising slowly with the peak in 2007-09. The rates have since declined but overall are still
higher in the most recent period of 2010-12 compared to the earliest in 2003-05. The trend
for females shows rates having risen slightly over time. Alcohol-attributable rates also vary a
little over time with a slight decrease in recent years. The alcohol-attributable rates in 201012 are slightly lower than in 2003-05 but this small difference is not statistically significant.
Mortality rates for males are around twice those for females (Figure 26).
Figure 26
Mortality, European age-standardised rate (EASR) per 100,000*, males
and females, all ages, Wales, 2003-05 to 2010-12
Produced by Public Health Wales Observatory, using fractions (PHE), ADDE & MYE (ONS)
Males
Fem ales
Alcohol-specific
15.0
15.3
7.6
8.2
95% confidence interval
18.0
18.9
18.7
8.2
8.9
8.8
16.4
17.6
17.3
9.2
8.6
9.0
74.2
76.7
78.4
79.4
77.1
75.0
73.2
33.6
33.3
33.4
34.0
33.5
33.5
32.2
32.5
2005-07
2006-08
2007-09
2008-10
2009-11
2010-12
2003-05
75.9
2004-06
Alcohol-attributable
*Using 2013 European Standard Population
5.2
Chronic liver disease and cirrhosis
Around 260 males and 170 females die from chronic liver disease in Wales per year (based on
2010-12). The majority of these, just over 200 males and 120 females, died from alcoholic
liver disease. The remainder have a cause of death of chronic hepatitis (not elsewhere
specified) or fibrosis and cirrhosis of the liver which in the large part of cases is due to alcohol
but can also have other causes. There is only slight variation in rates over time and
differences are not statistically significant (Figure 27).
34
Alcohol and health in Wales 2014: Wales profile
Figure 27
Mortality from chronic liver disease and cirrhosis, European agestandardised rate (EASR) per 100,000*, males and females, all ages,
Wales, 2003-05 to 2010-12
Produced by Public Health Wales Observatory, using ADDE & MYE (ONS)
Fem ales
10.7
10.4
18.1
10.8
10.4
11.0
2010-12
9.7
18.6
2009-11
9.5
19.8
2008-10
9.2
2003-05
16.7
2007-09
20.6
2006-08
19.5
17.3
2005-07
17.9
95% confidence interval
2004-06
Males
*Using 2013 European Standard Population
5.3
Months of life lost
The months of life lost due to alcohol is an estimate of the
increase in life expectancy at birth that would be expected if all
alcohol-attributable deaths among persons aged under 75 years
were prevented. It was produced using the English method and
unfortunately does not include confidence intervals. This measure
is related to alcohol-attributable mortality including the use of the
same fractions, but includes only the alcohol-attributable deaths
in those aged under 75. As with the alcohol-attributable mortality
this indicator needs to be interpreted with care. For further
details please see the technical guide.
“
…The months of life lost in
Wales with 13.5 months
for males and 6.5 months
for females was higher
than in England with 11.5
months for males and 5.4
months for females... ”
The number of months of life lost due to alcohol in Wales was 13.5 months for males and 6.5
months for females in 2010-12 (Figures 28 and 29). The months of life lost in Wales was
higher than in England with 11.5 months for males and 5.4 months for females.
The months of life lost due to alcohol in males was lowest in Powys tHB and Hywel Dda UHB
with nearly 12 months and highest in Cwm Taf UHB with just over 18 months (Figure 28). For
females this ranged from around 6 months in Hywel Dda UHB and Cardiff and Vale UHB to
just over 8 months in Cwm Taf UHB.
35
Alcohol and health in Wales 2014: Wales profile
Figure 28
Months of life lost due to alcohol, males aged under 75, Wales health boards,
2010-2012
Produced by Public Health Wales Observatory, using ADDE, Life Tables for Wales & MYE (ONS)
Wales = 13.5
Betsi Cadwaladr UHB
13.1
Powys tHB
11.7
Hywel Dda UHB
11.7
ABM UHB
15.3
Cardiff & Vale UHB
13.2
Cwm Taf UHB
18.1
Aneurin Bevan UHB
12.4
Figure 29
Months of life lost due to alcohol, females aged under 75, Wales health
boards, 2010-2012
Produced by Public Health Wales Observatory, using ADDE, Life Tables for Wales & MYE (ONS)
Wales = 6.5
Betsi Cadwaladr UHB
6.9
Powys tHB
6.1
Hywel Dda UHB
5.9
ABM UHB
6.7
Cardiff & Vale UHB
6.0
Cwm Taf UHB
8.1
Aneurin Bevan UHB
5.8
36
Alcohol and health in Wales 2014: Wales profile
5.4
Geographical variation in mortality
Alcohol-specific mortality rates vary locally, amongst health boards and small areas. The rate
for males in Powys tHB is statistically significantly lower than Wales with 11 per 100,000, and
statistically significantly higher in Cardiff and Vale UHB and Cwm Taf UHB with around 22 per
100,000. The rate for females is statistically significantly higher than Wales in Cwm Taf UHB.
Figure 30
Alcohol-specific mortality, European age-standardised rate per 100,000*,
males, all ages, Wales health boards, 2010-2012
Produced by Public Health Wales Observatory, using ADDE & MYE (ONS)
95% confidence interval
Betsi Cadwaladr UHB
16
Powys tHB
11
Hywel Dda UHB
14
ABM UHB
19
Cardiff & Vale UHB
22
Cwm Taf UHB
22
Aneurin Bevan UHB
16
Wales = 17
*Adjusted for ICD-10 coding changes and using the 2013 European Standard Population
Figure 31
Alcohol-specific mortality, European age-standardised rate per 100,000*,
females, all ages, Wales health boards, 2010-2012
Produced by Public Health Wales Observatory, using ADDE & MYE (ONS)
95% confidence interval
Betsi Cadwaladr UHB
10
Powys tHB
7
Hywel Dda UHB
7
ABM UHB
9
Cardiff & Vale UHB
9
Cwm Taf UHB
Aneurin Bevan UHB
Wales = 9
13
8
*Adjusted for ICD-10 coding changes and using the 2013 European Standard Population
37
Alcohol and health in Wales 2014: Wales profile
Rates for alcohol-specific mortality vary considerably at upper super output area level ranging
from 5 to nearly 30 per 100,000, see map (Figure 32). A separate map for each health board
is included in the health board summary document (see inside cover).
Figure 32
5.5
Comparison to England and English regions
Mortality rates for both definitions are higher in Wales than in England (Figures 33 and 34)
but mortality rates amongst the English regions vary considerably. Alcohol-specific mortality
rates range from 10.3 per 100,000 in the East of England to 20.2 in the North West for males
(from 4.4 in London to 10.6 in the East of England for females). The range amongst English
regions is similar for Wales health board areas, where rates vary from 11 to 22 for males and
7 to 13 for females (Figures 30 and 31). The Wales rate at 17.3 is most similar to the rate for
the West Midlands at 17.7.
The patterns are similar for alcohol-attributable mortality and the Wales rate appears most
similar to the North East of England. There are minor method differences between the
England and Wales rates but these are unlikely to affect the interpretation of the differences
much. Please see the technical guide for more details.
38
Alcohol and health in Wales 2014: Wales profile
Figure 33
3
Alcohol-specific mortality, European age-standardised rate
(EASR) per 100,000, all ages, Wales, England & English regions,
2010-2012
Produced by Public Health Wales Observatory, using ADDE, MYE (ONS) & LAPE
(PHE)
Males Fem ales
95% confidence interval
Wales
England
North East
North West
Yorkshire and The Hum ber
East Midlands
West Midlands
East of England
London
South East
South West
17
9
15
7
19
9
20
11
16
7
14
6
18
9
10
5
12
4
12
5
14
6
39
Alcohol and health in Wales 2014: Wales profile
Figure 34
Alcohol-attributable mortality, European age-standardised rate
(EASR) per 100,000, all ages, Wales, England & English
regions, 2012
Produced by Public Health Wales Observatory, using ADDE, MYE (ONS) & LAPE
(PHE)
95% confidence interval
Males Fem ales
Wales
England
North East
North West
Yorkshire and The Hum ber
East Midlands
West Midlands
East of England
London
South East
South West
72
33
63
28
70
31
77
34
67
28
63
29
68
31
54
26
59
25
58
26
60
26
40
Alcohol and health in Wales 2014: Wales profile
6
Socio-economic
patterns
Alcohol is a source of socio-economic inequalities in health as shown in our previous
publication on inequalities10. This section looks at some key alcohol indicators and their
relationship to area deprivation in the absence of suitable individual level socio-economic
data. We use the Welsh Index of Multiple Deprivation 2011 (WIMD) to identify ‘fifths’ of
deprivation. The most deprived fifth contains the most deprived 20% of areas and the least
deprived fifth the least deprived 20% of areas. As with all area-based deprivation measures,
not everyone living in a deprived area is necessarily living in deprived circumstances and
equally, some people living in an area classed as least deprived may experience deprivation.
Generally, we found little difference in alcohol consumption between the deprivation fifths for
heavy (more than 8 units for males, 6 units for females) and very heavy drinking (more than
12 units for males, 9 units for females) (Figures 35 and 36). The exception is heavy drinking
in females which is slightly higher in the least deprived areas (25%) compared to all other
deprivation fifths (20%-21%), although this is unlikely to be statistically significant. For heavy
drinking and very heavy drinking we therefore found no apparent relationship to deprivation.
There is an inverse relationship for drinking above guidelines (more than 4 units for males, 3
units for females) where the proportion is higher in the least deprived areas (52% in males)
than in the most deprived areas (45% in males). This is slightly stronger in females with 44%
in the least deprived areas compared to 32% in the most deprived.
A recent academic study using WHS data (2003/04-2007) found that deprivation acted
differently on the risk of binge drinking between males and females at different age groups 11.
The proportion of respondents drinking above guidelines showed the same inverse
relationship as in our analysis shown in Figures 35 and 36. In contrast to our results, the
study found that binge drinking increased with increasing deprivation11. Our broad analysis
looks at all WHS respondents aged 16 and over combined for 2011-2012 and so may mask
different relationships for some age groups, but also covers newer WHS data after the
question change and adjustment for strength of drinks from 2008 onwards. The deprivation
measure may also have changed between the WIMD 2005 and the WIMD 2011. These
patterns should therefore be interpreted with caution and further analysis would be needed to
investigate the differing results.
41
Alcohol and health in Wales 2014: Wales profile
Figure 35
Self-reported alcohol consumption, age-standardised percentage by deprivation fifth,
males aged 16+, Wales, 2011-12
Produced by Public Health Wales Observatory, using WHS & WIMD 2011 (WG)
Drinking above guidelines
Heavy (binge) drinking
Very heavy drinking
95% confidence interval
52
32
17
Least deprived
49
32
17
48
Next least deprived
31
18
Middle
48
32
19
Next m ost deprived
45
31
17
Most deprived
Figure 36
Self-reported alcohol consumption, age-standardised percentage by deprivation fifth,
females aged 16+, Wales, 2011-12
Produced by Public Health Wales Observatory, using WHS & WIMD 2011 (WG)
Drinking above guidelines
Heavy (binge) drinking
Very heavy drinking
95% confidence interval
44
25
13
Least deprived
39
21
11
Next least deprived
36
21
11
Middle
34
21
13
Next m ost deprived
42
32
20
12
Most deprived
Alcohol and health in Wales 2014: Wales profile
The pattern in alcohol consumption is in stark contrast to outcomes in
both mortality and hospital admissions which show a strong
relationship to deprivation where rates in the most deprived areas are
much higher than in least deprived areas.
This report focuses on the rate ratio, a measure of relative inequality,
as it can be compared between causes and between males and females
independent of scale. It is produced by dividing the rate for the most
deprived areas by the rate in the least deprived areas. For comparisons
over time the absolute difference between the rates was also
considered to ensure consistent interpretation of patterns.
“
…Mortality and
hospital admissions
are strongly related to
deprivation. In
contrast self-reported
heavy and very heavy
drinking is similar
between the
deprivation fifths ...”
Alcohol-specific indicators are more strongly related to deprivation than
alcohol-attributable indicators. Alcohol-specific mortality and hospital admissions have the
largest rate ratios, with the rates in the most deprived areas for males between 3.3 and 4.6
times higher than the rates in the least deprived areas (between 2.5 and 3.1 times higher for
females) (Figures 37 and 38).
The rate ratios were lower for alcohol-attributable mortality where the rates for males in the
most deprived areas were around twice those in least deprived areas (around 1.7 times
higher for females) (Figure 37). The rate ratio is also higher for males than for females, and
mortality for both definitions is more strongly related to deprivation for males than females.
Inequalities in mortality have not changed significantly since 2003/05. Although there is
variation in rates over time, mortality rates have increased overall with only a slight drop in
rates for males in the most recent years (Figure 37). They have increased more in the most
deprived compared to the least deprived. The gap has therefore slightly widened, although
due to relatively small numbers of deaths and large confidence intervals this difference is not
statistically significant. Changes in the gap for alcohol-attributable mortality are much smaller
and similarly not statistically significant.
43
Alcohol and health in Wales 2014: Wales profile
Figure 37
Alcohol specific mortality by deprivation fifth, Eurpoean age-standardised rate (EASR) per
Mortality
byall
deprivation
fifth,
European
age-standardised rate (EASR) per 100,000*, all ages, Wales, 2003-05 to 2010-12
100,000*,
males
ages, Wales,
2003-05
to 2010-12
Produced
byHealth
Public Wales
HealthObservatory,
Wales Observatory,
using MYE
ADDE,
MYE &
(ONS),
(PHE) & WIMD 2011 (WG)
Produced
by Public
using ADDE,
(ONS)
WIMDfractions
2011 (WG)
Most deprived
45
Wales
Least deprived
Males
40
95% confidence interval
95% confidence interval
Alcohol-specific mortality
45
Fem ales
20
40
35
30
25
35
15
30
25
10
20
15
3.3
Males
2003-05
140
4.6
3.9
3.5
2004-06
2005-07
2006-08
rate ratio
3.1
2.7
2.8
2.8
2005-07
2006-08
2008-10
2009-11
3.4
3.9
4.0
Alcohol-attributable mortality
2007-09
*Using 2013 European Standard Population
120
2008-10 502009-11
3.6
3.0
2010-12
3.1
2007-09
3.4
2004-06
0
2.5
2003-05
3.6
2010-12
4.0
2009-11
3.9
2008-10
3.9
Fem ales
2010-12
40
100
80
30
60
20
40
2.1
2004-06
2005-07
2006-08
2007-09
2008-10
2009-11
2010-12
0
1.6
1.8
1.7
rate ratio
1.7
1.6
1.7
1.7
1.8
2010-12
2.1
2009-11
2.1
2008-10
2.0
2007-09
2.1
2006-08
2.1
2005-07
2.2
2004-06
2.0
2003-05
0
10
rate ratio
20
2003-05
0
Rate ratio
4.6
2007-09
10
5
3.3
rate ratio
3.5
3.4
2006-08
0
2005-07
5
2004-06
15
5
10
2003-05
20
*Using the 2013 European Standard Population
Interpreting inequalities trends in hospital admissions is more
complex due to the nature of the three different definitions used.
Please see Section 4.1 for a discussion on the definitions or the
technical guide for further details. Our analysis covers the
financial years 2003/04 to 2012/13.
“
… There is no sign of
improvement in the gap in
mortality between the
most deprived and the
least deprived areas over
time... ”
The rates of individuals being admitted to hospital with alcoholspecific conditions varies over time, but rates have overall been
rising in both the most and the least deprived areas (Figure 38).
The rates had increased relatively more in the most deprived in the first few years and so the
absolute rate difference had been rising. This difference narrowed again in the most recent
years to a similar level to the first year. The rate ratio, the relative measure of inequalities,
also fluctuated over time, but has dropped in the more recent years. The rate in the most
deprived areas for males has dropped from 4.4 times in the first period to 3.4 times the rate
in the least deprived in the most recent period (3.4 to 2.8 in females). Relative inequalities
have therefore narrowed over time, although due to large confidence intervals this difference
is not statistically significant in females. This suggests that a decrease in the most deprived
areas in recent years may have contributed to a slight narrowing of inequalities in alcoholspecific hospital admissions.
44
Alcohol and health in Wales 2014: Wales profile
The pattern for individuals admitted with alcohol-attributable
conditions using the narrow definition appears similar but far less
pronounced with only a very slight narrowing in inequalities that
is not statistically significant (Figure 38). The absolute rate
difference and the rate ratio have slightly decreased over time,
again due to a larger decrease in rates in the most deprived
areas than in the least deprived most recently.
“
…Inequalities in alcoholspecific and alcoholattributable (narrow)
hospital admissions have
slightly narrowed over
time, but slightly widened
for the alcohol-attributable
(broad) admissions... ”
Some conditions are more linked to deprivation than others and
are more prevalent and therefore contribute differently to
inequalities. Although the data is not shown here, wholly
attributable conditions made the biggest contribution to inequalities in the narrow admissions,
with mental and behavioural disorders making up the majority of these. Cancer,
cardiovascular disease and unintentional injuries had the opposite effect, where the
proportions of individuals admitted with these conditions were higher in least deprived areas
and so were causing the inequalities to narrow. Please note that these condition groups only
contain certain conditions that are linked to alcohol, for example, only some cardiovascular
diseases. For a list of these please see the technical guide.
The pattern of inequalities is slightly different for individuals admitted with alcoholattributable conditions using the broad definition. The rates in most deprived and least
deprived areas have consistently risen over time with only slight variation. The rate ratios
have fallen slightly from 2.2 to 1.9 for males and females, but the difference in females is not
statistically significant. The absolute rate difference has widened slightly over time, and the
reduction in rate ratio is largely down to the rise in rates and the relatively larger rise in the
least deprived areas.
As with the narrow definition discussed earlier, the conditions have different effects on
inequalities. Although the data is not shown here, wholly attributable conditions also make
the biggest contribution to inequalities in the broad admissions, with mental and behavioural
disorders making up the majority of these. As discussed in Section 4 admissions with
cardiovascular disease as a diagnosis (primary or secondary) make up a large proportion of
these admissions, but have a narrowing effect on inequalities. The proportion of individuals
admitted with a cardiovascular disease linked to alcohol is actually higher in the least
deprived areas than the most deprived. Please note that the diagnosis may not be the main
reason for admission in this definition.
The reasons why the outcomes measures are strongly related to deprivation but not to
consumption indicators are unclear. There may be differences in individuals’ drinking patterns
in the least and most deprived areas, for example there may be differences in the regularity
of heavy drinking episodes or the actual units consumed above the limit. It may also be that
underestimation of consumption in surveys is proportionally greater in deprived areas, or
heavier drinkers are underrepresented in the survey. As discussed earlier our broad analysis
may also mask relationships to deprivation by age groups. In addition the pattern in
outcomes due to alcohol may be influenced by other factors such as poorer general health
and differences in individuals seeking treatment or help.
45
Alcohol and health in Wales 2014: Wales profile
Figure 38
Hospital
bydeprivation
deprivation fifth,
fifth (person-based),
European age-standardised
Alcohol
specificadmissions
mortality by
Eurpoean age-standardised
rate (EASR) per rate (EASR) per 100,000*, all ages,
Wales,
financial
years
2003/04-2012/13
100,000*,
males
all ages,
Wales,
2003-05 to 2010-12
Produced
byHealth
Public Wales
HealthObservatory,
Wales Observatory,
using MYE
PEDW
(NWIS),
MYE (ONS)
, fractions (PHE) & WIMD 2011 (WG)
Produced
by Public
using ADDE,
(ONS)
& WIMD
2011 (WG)
1,000
500
900
450
800
400
700
350
600
300
25
500
250
400
200
20
300
150
3.3
4.6
3.9
Males
2003-05
2004-06
2005-07
3.0
3.2
2005/06
2006/07
2007/08
2009/10
2010/11
2.9
2.8
2012/13
rate ratio
3.1
3.1
2008/09
3.7
Fem ales
500
450
300
500
250
400
200
300
150
1.9
2.1
2.1
rate ratio
2.1
2.1
1.9
2.0
2010/11
0
2.0
2009/10
2007/08
50
2007/08
2006/07
2.0
2006/07
2005/06
2.1
2005/06
2004/05
2.1
2004/05
2003/04
2.2
2003/04
2.3
2012/13
2.3
2011/12
2.3
100
2010/11
2.3
rate ratio
2.3
2.2
Alcohol-attributable admissions (broad)
Males
2,500
1.9
1.8
2012/13
350
600
2008/09
400
700
0
3.4
3.4
3.9
4.0
3.6
Alcohol-attributable admissions (narrow)
2006-08
2007-09
2008-10
2009-11
2010-12
800
100
3.2
2004/05
0
3.4
2003/04
50
3.4
2012/13
3.9
2011/12
3.8
2010/11
4.1
3.5
*Using1,000
2013 European Standard Population
900
200
Fem ales
100
2009/10
4.2
2006/07
4.2
rate ratio
3.8
3.9
2009/10
Fem ales
1,400
1,200
2,000
1,000
1,500
800
1,000
600
400
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12
2012/13
0
*Using the 2013 European Standard Population
46
2.2
2.1
2.1
2.1
rate ratio
2.0
2.0
1.9
2.0
2.0
1.9
2012/13
1.9
2011/12
1.9
2010/11
1.9
2009/10
2.0
2007/08
rate ratio
2.0
2.0
2006/07
2.1
2005/06
2.1
2004/05
2.2
2003/04
0
200
2.2
2008/09
500
2003/04
0
Rate ratio
4.3
2005/06
5
0
4.4
2004/05
10
100
2003/04
15
2008/09
200
2008/09
30
2007/08
35
95% confidence interval
Alcohol-specific admissions
Males
40
95% confidence interval
Least deprived
2011/12
Wales
2011/12
Most deprived
45
Alcohol and health in Wales 2014: Wales profile
7 References
1. World Health Organisation. Global status report on alcohol and health 2014. Geneva:
WHO; 2014. Available at:
http://www.who.int/substance_abuse/publications/global_alcohol_report/en/
[Accessed: 23rd Sep 2014]
2. Welsh Assembly Government. Working together to reduce harm. The substance
misuse strategy for Wales 2008-2018. Cardiff: WAG; 2008. Available at:
http://wales.gov.uk/topics/people-andcommunities/safety/substancemisuse/publications/strategy0818/?skip=1&lang=en
[Accessed: 23rd Sep 2014)
3. Gartner A et al. A profile of alcohol and health in Wales. Cardiff: Wales Centre for
Health; 2009. Available at:
http://www.wales.nhs.uk/sites3/Documents/749/Alcohol%20Profile%20for%20Wales
%20(E).pdf [Accessed: 23rd Sep 2014]
4. Public Health England. User guide: Local alcohol profiles for England. [Online]. 2014.
Available at: http://www.lape.org.uk/downloads/LAPE%20User%20Guide_Final.pdf
[Accessed: 23rd Sep 2014)
5. Jones L, Bellis MA, Updating England-specific alcohol-attributable fractions. Liverpool:
John Moores University; 2014. Available at: http://www.cph.org.uk/wpcontent/uploads/2014/03/24892-ALCOHOL-FRACTIONS-REPORT-A4-singles24.3.14.pdf [Accessed: 23rd Sep 2014)
6. NHS Choices. Alcohol units. [Online]. 2013. Available at:
http://www.nhs.uk/Livewell/alcohol/Pages/alcohol-units.aspx [Accessed: 23 Sep
2014]
7. NHS Choices. Should my child drink alcohol? [Online]. 2013. Available at:
http://www.nhs.uk/chq/pages/2595.aspx?categoryid=62 [Accessed: 23rd Sept 2014)
8. Goddard E. Obtaining information about drinking through surveys of the general
population. National Statistics Methodology Series No. 24. [Online]. 2001.
http://www.ons.gov.uk/ons/guide-method/method-quality/specific/gss-methodologyseries/gss-methodology-series--24--obtaining-information-about-drinking-throughsurveys-of-the-general-population.pdf. [Accessed: 23rd Sep 2014]
9. Catto S, Gibbs D. How much are people in Scotland really drinking? Edinburgh: NHS
Health Scotland, ScotPHO; 2008. Available at:
http://www.scotpho.org.uk/publications/reports-and-papers/491-how-much-arepeople-in-scotland-really-drinking-a-review-of-data-from-scotlands-routine-nationalsurveys- [Accessed: 23rd Sep 2014]
10.Public Health Wales Observatory, 2011, Measuring inequalities 2011. Trends in
mortality and life expectancy in Wales. Public Health Wales NHS Trust; 2011. Available
at: http://www.wales.nhs.uk/sitesplus/922/page/58379 [Accessed: 1st Oct 2014]
11.Fone D et al. Socioeconomic patterning of excess alcohol consumption and binge
drinking: a cross-sectional study of multilevel associations with neighbourhood
deprivation. BMJ Open 2013:3:e002337. doi:10.1136/bmjopen-2012-002337.
Available at: http://bmjopen.bmj.com/content/3/4/e002337.full.pdf+html [Accessed:
23rd Sep 2014]
47