Friday July 19 2013
Women shouldn't regularly drink more than 2-3 units a day
"More young women dying of alcohol abuse," The Times reports, with much of the media covering the story that the risk of death from an alcohol-related health condition has increased for women in their 30s and 40s.
The study behind this headline looked at three decades of reliable national statistics on alcohol-related deaths from Glasgow, Liverpool and Manchester. For all three cities, alcohol-related deaths have increased for both men and women over the past 30 years, with men typically having death rates higher than women. However, for the youngest group of people included in this study – those born in the 1970s – the number of deaths in women has increased compared with previous generations.
The researchers do not offer explanations for why the number of alcohol-related deaths is increasing. A number of newspapers mention that the increase could be due to the “ladette” culture that developed in the 1990s, in which it became more socially acceptable for young women to drink as heavily as men. This is plausible but cannot be proven by the evidence presented in this study.
Whatever the reason, the researchers say, “it is imperative that this early warning sign is acted upon”. Given that this increase has been seen in all three cities they say that “failure to have a policy response to this new trend may result in the effects of this increase being played out for decades to come”.
A better understanding of how past increases in alcohol-related deaths came about is needed to help to prevent such patterns repeating themselves in the future.
Where did the story come from?
The study was carried out by researchers from the Glasgow Centre for Population Health and no sources of external funding are reported.
The study was published in the peer-reviewed Journal of Epidemiology and Community Health and has been made free to download on an open access basis.
The media generally reflects the findings of this study accurately. The findings have prompted a great deal of speculation as well as moralising. It is important to remember that even expert opinion is not evidence.
Finally, the main purpose of the study – to see why there are such stark health inequalities between Scotland and other countries in western Europe – is barely mentioned by the media.
What kind of research was this?
This was a cross-sectional time trend analysis looking at health inequalities in general, and alcohol-related deaths in Scotland.
The researchers explain that Scotland has the highest working age mortality rate in western Europe. The “excess” poor health is believed to be partly due to greater deprivation in the country’s cities and the country’s industrial past.
However, the increasing health gap between Scotland and the rest of Britain cannot be due solely to deprivation. For example, premature mortality in Glasgow is 30% higher than in Liverpool and Manchester, both of which have similar levels of poverty and poor health, and an industrial past. The researchers use the phrase “the Scottish effect” to describe this unexplained health gap.
It is notable that trends in alcohol-related deaths for both Scotland and the UK have markedly deviated from other western European countries. Alcohol-related deaths are said to have a strong deprivation and gender gradient. In this cross-sectional time trend analysis, the researchers wanted to analyse the trend in alcohol-related deaths in Glasgow from 1980 to 2011, and compare this with Liverpool and Manchester.
What did the research involve?
The researchers obtained data on population and mortality for Glasgow from the National Records of Scotland and for Liverpool and Manchester from the Office for National Statistics.
Alcohol-related deaths were defined using codes from the International Classification of Diseases (ICD). The researchers included any deaths related to alcohol use, which were grouped broadly into categories of:
- the various forms of alcohol-related liver disease
- alcohol-related damage to an organ of the body other than the liver
- mental and behavioural disorders due to use of alcohol
- poisoning by exposure to alcohol (accidental or intentional)
Death rates were standardised with five-year rolling averages calculated. People were categorised into their decade of birth so that mortality rates for each birth cohort could be calculated. As this study was looking at 1980 onwards, the youngest birth cohort in this study was those born in the 1970s, and the oldest those born in the 1910s.
The researchers give the following as an example: to calculate the alcohol-related mortality rate for 1999 for people born in the 1960s, they would calculate this as the number of alcohol-related deaths in 1999 among those born in the 1960s divided by all people born in the 1960s. They looked at trends in alcohol-related deaths by age and by sex.
What were the basic results?
In the early 1980s alcohol-related deaths were reported to be three times higher in Glasgow than in Manchester or Liverpool. In all three cities, the number of alcohol-related deaths increased over the 30-year period, with the greatest increase in Glasgow.
There were 24 alcohol-related deaths per 100,000 of the Glasgow population in 1981. By 2008 this had more than doubled to a peak of 64 per 100,000 in 2008. Comparatively, the greatest rise seen for both Manchester and Liverpool was only half that seen in Glasgow – an increase of 19 per 100,000.
Alcohol-related deaths by age in Glasgow, Manchester and Liverpool
In all three cities, the highest burden of alcohol-related deaths was among those in their 40s and 50s. Across all “birth cohorts” (groups of people born in the same decade), the alcohol-related death rate began to rise in those aged 30–40 years of age, reaching a peak in those 50 and 60, then falling in those over 65 years of age.
Alcohol-related deaths by gender in Glasgow, Manchester and Liverpool
By sex, the number of alcohol-related deaths is roughly 2-3 times higher in men than women in all three cities, and this has remained fairly consistent over time. The increase in alcohol-related deaths over the past 30 years in all three of these cities has tended to be an increase for both men and women.
However, while the increases in Manchester and Liverpool rose quite steadily each year, in Glasgow there was a massive rise in alcohol-related deaths among men over the 10 years from around 1990 up to 2000, with a comparatively smaller rise among Glasgow women. The alcohol-related deaths for Glasgow men before this period had fluctuated between 30 and 40 deaths per 100,000 people for decades but by 2000 to 2004 they had reached around 85 per 100,000. When the researchers looked by birth cohort year, the steepest rise at this time in Glasgow was among those born in the 1940s and 50s, who would have been in their 40s and 50s in this decade.
From around 2003 onwards, the deaths among Glasgow men and women have both begun to decline, though deaths among men have fallen by a greater extent from their high peak. However, when the researchers looked again by birth cohort, the decline in death rate has been seen in Glasgow in all but the youngest birth cohort – those born in the 1970s. Notably, for this youngest age group, the gap between men and women in the number of alcohol-related deaths is narrowing – not just in Glasgow, but in all three cities.
Most alcohol-related deaths across all birth cohorts were liver-related. Around a quarter of deaths were due to mental and behavioural disorders related to alcohol, and far smaller numbers were due to other organ damage or poisoning.
How did the researchers interpret the results?
The researchers focus on their observation of the recent narrowing in the gap between men and women born in the 1970s in terms of alcohol-related mortality. They say “it is imperative that this early warning sign in young women in the UK is acted on if deaths from alcohol are to reduce in the long term”.
This time trend study has looked into the change in the number of alcohol-related deaths in Glasgow, Liverpool and Manchester over the last 30 years, by sex and by age group. One of the strengths of the study is that it used reliable national statistics and mortality data to examine the trends in these three cities.
The observations included a general rise in the number of alcohol-related deaths in all three cities over the 30 years, with the number of deaths in men tending to exceed those in women by about 3–4 times.
A startling observation, ignored by the media, was the great rise in alcohol-related deaths in men in Glasgow in the 1990s, which is something for which the researchers say “there is no obvious single cause”.
For the rise in general, it is thought that the relatively recent trend for heavy binge drinking may play a role, though it is not possible to examine the influence of drinking patterns using this study.
The study was also not able to examine the number of incidents or accidents involving alcohol – for example, people who suffered harm or injury themselves, or inflicted harm or injury upon others, while under the influence of alcohol. However, accurate data on this would be difficult to gather.
Another important finding across all three cities has been a narrowing in the gap between men and women in the number of alcohol-related deaths for the youngest people in this study – those born in the 1970s. Though this study can offer no explanation as to why the number of alcohol-related deaths among women seems to be increasing, the researchers say that “it is imperative that this early warning sign is acted upon”. Given that this increase has been seen in all three cities they say that “failure to have a policy response to this new trend may result in the effects of this increase being played out for decades to come”.
This study provided very few explanations for the sometimes dramatic trends in alcohol-related deaths observed over the last 30 years – in particular the massive peak of deaths among men in Glasgow during the 1990s. This large knowledge gap needs to be filled by further research if there can be any hope for a sustained effort to prevent similar peaks in deaths in the future.
Analysis by Bazian. Edited by NHS Choices. Follow Behind the Headlines on Twitter.