Women and alcohol

Summary

• Roughly four fifths of women in England report drinking alcohol, with average consumption at nine units a week
• Around one in seven women drink more than the Chief Medical Officers’ weekly low risk guideline
• Women make up 39% of the alcohol-related hospital admissions in England
• Evidence suggests that many of alcohol’s effects pose a greater risk to women’s physical health at lower consumption levels than men
• There is also evidence of alcohol marketing targeting women, perpetuating gendered stereotypes
• Policy options to reduce alcohol harm among women include population-level policies, marketing restrictions, and education and information.

Introduction

In recent decades, the historical gender gap between men and women’s alcohol consumption has closed.

79% of women in England reported drinking in the last year and average alcohol consumption among women is nine units a week. Around one in seven women drink more than the Chief Medical Officers’ weekly low risk guideline (14 units a week).

The number of women in alcohol treatment in England has been relatively stable over the past decade, with women comprising 40% of ‘alcohol only’ clients. Women make up 39% of the alcohol-related hospital admissions in England. The rate of alcohol-related admissions to NHS hospitals in England has continually risen over the last decade: for women in 2018/19 this was almost 1.5 times the rate compared with a decade previously. Alcohol-related and alcohol-specific mortality have been relatively stable among women over this time period.

Evidence suggests that many of alcohol’s effects pose a greater risk to women’s physical health at lower consumption levels than men, and some risks are accentuated with ageing.

Some alcohol-related harms impact exclusively or overwhelmingly on women:

• Breast cancer: There is a linear and dose-response relationship between alcohol and breast cancer, meaning even moderate alcohol consumption carries an increased risk
• Reproductive health: Alcohol has been linked to menstrual cycle dysfunction, a decreased chance of conceiving, increased odds of seeking fertility treatment, worse outcomes from assisted reproductive technology (ART) treatment, and earlier menopause
• Pregnancy:The Department of Health & Social Care advises pregnant women and women trying to conceive against drinking alcohol, but the UK has the fourth highest rate of drinking during pregnancy in the world. Drinking during pregnancy is associated with pregnancy loss, an increased likelihood of low birthweight, preterm birth and being small for gestational age, as well as Foetal Alcohol Spectrum Disorders. Heavy episodic drinking is associated with teenage pregnancy, which has long term socio-economic consequences
• Domestic violence: Seven in 10 victims of partner violence are women, and alcohol is a major contributor to partner violence
• Sexual assault: Almost four in ten victims of serious sexual assault reported the offender(s) were under the influence of alcohol. The same proportion of victims reported they were under the influence of alcohol themselves during this serious sexual assault

Research shows some alcohol marketing targets women, and representations of women in alcohol and nightlife venue marketing reproduce gendered stereotypes.

There are research gaps particularly interventions designed for women and alcohol marketing, and more collaboration between different stakeholder groups is needed. Policy options to reduce alcohol harm among women include population-level policies, marketing restrictions, and education and information. Alcohol treatment services could adapt to better support the needs of women.

Trends in women’s alcohol consumption and harm

How many women drink?

The 2018 Health Survey for England shows 79% of women in England had drunk alcohol in the past year (compared with 86% of men) [1].

While alcohol use and related harms were historically more prevalent among men than women, this gender gap has now closed. Explanations offered for the rise in women’s drinking include women’s changing role in the workplace and changing education levels, lifestyle shifts, alcohol advertising targeting women, ‘empty nest syndrome’ among middle-aged women when their children leave home, and ‘ladette’ culture among younger women.

A 2016 analysis of 68 international studies, with a combined sample size of over 4 million people, found the male-to-female ratios of alcohol use and related harm shrunk dramatically across birth cohorts from 1891 to 2001. For those born in the early 1900s, men were 2.2 times more likely than women to drink alcohol, three times more likely to drink in a way that suggested problematic use, and 3.6 times more likely to experience alcohol-related harms. However, for those born in the late 1990s, men were only 1.1 times more likely to drink alcohol than women, 1.2 times more likely to drink in a way that suggested problematic use, and 1.3 times more likely to experience alcohol-related harms [2].

Figure 1  Prevalence of any alcohol use (%) in females (x-axis) and males (y-axis) by five-year birth cohort ⁱ

How often and how much do women drink?

The 2018 Health Survey for England data show 50% of women drank in the past week (compared with 65% of men). In 2018, 9% of women drank on five or more days in the past week [3], similar to previous years [4]. Older women are more likely to drink regularly.

Figure 2  Proportion of alcohol drunk in the last week

The average number of drinking days in the last week is fairly similar for women (2.7 days) and men (3.1 days) [5].

Average weekly alcohol consumption among women who drank in the past year was nine units a week in 2018 (compared with 15.5 units among men) and has been fairly stable in recent years. In England in 2018, around one in seven women (14%) exceed the recommended drinking guidelines of 14 units a week, compared with almost one in three men (30%) [6].

In terms of trends over time, the Health Survey for England reports estimated maximum consumption on any one day in the past week using a consistent methodology dating back to 2006. Between 2006 and 2013, the proportion of women consuming more than 3 units on any day in the last week dropped from 33% to 27%. Since 2013, this proportion has fluctuated between 25% and 27%; in 2018, it was 25%. The equivalent figure for men has been in gradual decline since 2009 [7]:

Figure 3  Maximum amount drunk on any day in the last week, 2006-2018

What do women drink?

In 2014 the Financial Times reported that seven of every ten bottles of wine purchased in British supermarkets were bought by women [8]. One influence on these choices around drink type is marketing (see further down document for more information).

Data on drink type are routinely collected in the Health Survey for England but not published in the annual reports. According to Office for National Statistics figures for 2014, the most popular drink for women is wine. 70% of women ‘binge drinkers’* who reported drinking in the last week consumed wine on their heaviest drinking day, compared to 33% who consumed spirits or liqueurs and 22% who drunk normal strength beer, stout, lager or cider. For women non-binge drinkers** who reported drinking in the last week, 61% reported consuming wine on their heaviest drinking day last week [9].

Inequalities, women and alcohol

The Health Survey for England data show women’s alcohol consumption by deprivation quintile. In 2018, there was a clear gradient in teetotalism which was more common in more deprived groups, and with low risk and increasing risk drinking which was more common in more advantaged groups. There was not a gradient in higher risk drinking (>35 units a week) [10]:

Figure 4  Women’s alcohol consumption categories by deprivation quintile

This is an example of a longstanding pattern that is also seen in other data sources and using other measures of socio-economic status. The Opinions and Lifestyles Survey 2014 found a higher proportion of economically inactive and unemployed women were teetotal than those in employment (34% and 37% respectively vs 16%). The same survey found a higher proportion of women in employment drank in the week prior to the survey than women who were unemployed or economically inactive (60% vs 39% and 44% respectively) [11].

While women in more advantaged groups drink more often and more heavily than their more disadvantaged counterparts – a pattern also seen among men – disadvantaged women experience more of the harms from alcohol, in a pattern known as the ‘alcohol harm paradox’ [12].

For more detail of the relationship between alcohol and socio-economic inequality, read our alcohol and health inequalities briefing.

Women and specialist alcohol treatment

Estimates of the number of alcohol dependent women are scarce, with the official estimate of close to 600,000 alcohol dependent adults potentially in need of specialist treatment in England not disaggregated by gender [13]. The Adult Psychiatric Morbidity Survey 2014 estimated 0.6% of women in England were alcohol dependent (compared with 1.9% of men) [14]. As with men, the majority of women potentially in need are not in receipt of specialist treatment.

The National Drug Treatment Monitoring System 2018-19 estimates women comprise 40% of all ‘alcohol only’ clients (30,345 out of 75,555 individuals) as well as 27% of ‘non-opiate and alcohol’ clients (7,770 out of 28,598 individuals)*** in England [15]. The number of women in alcohol treatment in England has been relatively stable over the past decade [16].

Out of all individuals in substance use treatment, a greater proportion of women reported either living with a child or being a parent when they started treatment compared with men (58% of women, 48% of men) [17]. This is one example of why gender equity is an important consideration in the design of treatment services. Treatment services have typically been designed around male service users, with few designed to meet women’s needs, for example through providing childcare or women-only spaces [18].

Alcohol-related hospital admissions and mortality among women

Women make up 39% of alcohol-related hospital admissions in England, despite being more likely to be non-drinkers and also drinking less on average compared with men [19].

The rate of alcohol-related admissions to NHS hospitals in England has risen a small amount over the last decade: for women in 2018/19 there were 494 admissions per 100,000, compared with 440 in 2008/09 (by narrow measure) [20]:

Figure 5  Admission episodes for alcohol-related conditions (Broad)

Figure 6  Admission episodes for alcohol-related conditions (Narrow)

Alcohol-related and alcohol-specific mortality among women have been relatively stable over this time period; with 28.7 alcohol-related deaths and 7.0 alcohol specific deaths per 100,000 women in 2018 [21].

In contrast to the higher alcohol consumption among more affluent groups described earlier, there is a clear and persistent gradient in alcohol-related admissions and mortality with more deprived groups experiencing higher rates. In England in 2017/18 the most deprived decile had a 50% higher rate of alcohol-related hospital admissions (broad measure) compared with the least deprived decile (2,839 per 100,000 compared with 1,814), as well as a 50% higher rate of alcohol-related mortality (58.7 per 100,000 compared with 38.3) [22]. These figures are not gender specific. However, studies have identified concerning trends in alcohol-related mortality among young women in deprived areas in certain cities in the UK, particularly among a cohort of women born in the 1970s [23].

Effects of alcohol on women

Sex differences in risk

Many adverse effects of alcohol are common to both women and men. However, evidence suggests many of these effects are a greater risk to women’s health at lower consumption levels than men.

In the past this was reflected in UK drinking guidelines with lower levels for women, however since 2016 there has been a universal drinking guideline of 14 units a week, which took into account the greater risks of acute harms among men.

The Royal Medical Colleges reported that studies into alcohol-related harm in women consistently found the consumption level at which relative risk of mortality starts to rise is around 16 grams of pure alcohol per day (two UK units). It also known that women reach higher blood ethanol concentrations than men following the same dose of ethanol, regardless of body weight. This is because ethanol is soluble in water and women have a higher ratio of body fat to body water than men [24].

Higher blood ethanol concentrations for a given level of consumption have a role in development of alcohol dependence. Women experience an increased risk of dependence at an earlier stage or lower consumption levels than men [25]. Physiological changes as a result of heavy drinking – including liver disease such as cirrhosis and hepatitis – also have a shorter onset time and occur at lower consumption levels among women compared with men [26]. Women are also more sensitive to brain damage caused by alcohol dependence, with alcohol dependent women’s brain volumes (measures of grey and white matter) affected more than their male counterparts [27].

It has also been found that women who drink excessively develop more medical problems than men [28]. This is consistent with women in developed countries having a generally higher burden of chronic disease morbidity – but not mortality – than men [29].

Alcohol and ageing

As women age alcohol poses some unique issues. Older women have less lean muscle mass and continue to lose this with age, which makes them more susceptible to effects of alcohol. Liver enzymes which process alcohol and medication lose effectiveness with age, and the central nervous system becomes more sensitive. This poses a greater problem for older women who drink, as older women tend to take more medication [30]. Older women are also more likely to experience falls than older men, the risk of which is also increased by certain medications [31,32].

Possible benefits of moderate drinking for cardiovascular health for women over the age of 55 have received attention. The relationships between alcohol consumption and cardiovascular disease are complex [33]. This is reflected in the guidelines from the UK Chief Medical Officers’, which highlight that recent evidence suggests “net benefits from small amounts of alcohol are less than previously thought and are significant in only a limited part of the population”. They note this benefit only applies to women over 55 years, “for whom the maximum benefit is gained when drinking around 5 units a week, with some beneficial effect up to around 14 units a week” [34].

For information on the health impacts of alcohol more generally, read our briefings on alcohol and health.

Some alcohol-related harms impact exclusively or overwhelmingly on women:

Breast cancer

Studies first indicated a correlation between alcohol use and breast cancer in the mid 1990s. An article in the New England Journal of Medicine suggested that the balance of risks and benefits for women was complicated by a light-to-moderate alcohol intake [35].

A comprehensive review published by the World Health Organization in 2004 established there is a clear linear dose-response relationship between alcohol consumption and breast cancer [36]. Therefore, standard medical advice commonly dictates that in order to limit the likelihood of getting breast cancer, women ought to avoid or limit the consumption of alcohol [37].

Since 2012, alcohol has been classified as a group one carcinogen by the World Health Organization’s International Agency for Research on Cancer [38]. Alcohol is known to cause seven types of cancer: mouth and upper throat, laryngeal, oesophageal, breast, liver, and bowel cancers.

Figure 7  Which cancers are affected by alcohol?

More recently, the relationship between moderate alcohol intake and cancer incidence in women has been established. A 2009 publication using data from the Million Women Study**** identified that for women up to age 75, the risk of breast cancer rises by 11 cases per 1,000 for every 10g of alcohol (ie every additional drink) consumed daily [39]. This finding was consistent even at low consumption levels; for instance, a group of 1,000 women who have just 1 drink a day will have 11 more cases of breast cancer compared to non-drinkers. It was also estimated that 11% of all breast cancer in women in the United Kingdom is attributable to alcohol, equivalent to 5,000 cases annually [40]. A subsequent meta-analysis in 2012 found a 4% increase in the risk of breast cancer among people drinking one alcoholic drink per day, with three or more drinks per day associated with an increased risk of 40-50% [41].

The updated low risk drinking guidelines in 2016 were supported by a review from an expert group of the available evidence on alcohol’s health effects. In this report, it is stated that the evidence base on women’s breast cancer risk had grown since the previous drinking guidelines in 1995, specifically, that this risk starts from any level of alcohol consumption, and rises in a dose-response manner. The official guidelines now state: “The risk of developing a range of health problems (including cancers of the mouth, throat and breast) increases the more you drink on a regular basis” [42].

Reproductive health

Alcohol use has several impacts on women’s reproductive health. In a 2017 review, heavier drinking and heavy episodic drinking were linked to menstrual cycle dysfunction, poorer measures of ‘ovarian reserve’ (a clinical measure of a woman’s reproductive potential), a decreased chance of conceiving, increased odds of seeking fertility treatment, worse outcomes of assisted reproductive technology (ART) treatment, and earlier menopause [43]. Mirroring an issue common across medical research relating to women’s health [44], the authors identified a paucity of high quality studies in this area.

One Danish study found that even small amounts of alcohol (one to six drinks a week) can affect a woman’s fertility, with those drinking one to six drinks a week almost twice as likely to experience infertility compared with those drinking less than one drink a week [45].

There is evidence that some of the negative impact of alcohol on fertility can be reversed. For example, a small study from New Zealand found women who reduce their drinking or do not drink at all during fertility treatment were twice as likely to conceive as those who did not alter their drinking patterns before treatment [46].

Pregnancy and postnatal care

A recent international study found the UK has the fourth highest rate of drinking during pregnancy in the world [47]: in 2017, 11% of pregnant women drank in the last week [48].

The UK Department of Health advises against pregnant women or women trying to conceive drinking alcohol, and warn that “drinking during pregnancy can lead to long-term harm to the baby, with the more you drink, the greater the risk” [49]. The Scottish Chief Medical Officer also advises against pregnant women drinking alcohol during pregnancy, as “there is no ‘safe’ time for drinking alcohol during pregnancy and there is no ‘safe’ amount”. This is based on the precautionary principle: [50,51] there is an absence of evidence that small amounts of alcohol are harmful, but this does not equate to small amounts being safe.

Drinking during pregnancy has multiple effects, including Foetal Alcohol Spectrum Disorders (FASD) (see next section). Several studies have identified associations between alcohol consumption and pregnancy loss [52]. A 2011 systematic review identified heavy drinking is associated with increased likelihood of low birthweight, preterm birth and being small for gestational age [53].

Postnatally, while evidence on the effect of alcohol consumption while breastfeeding on infants is mixed and limited [54,55,56,57], NHS guidance on drinking during breastfeeding recommends that breastfeeding mothers follow the drinking guidelines and also avoid breastfeeding for two to three hours for every drink to avoid exposing the baby to any alcohol in milk. The guidance also stated parents should not share a bed or sofa with their babies if they have drunk any alcohol, as this has a strong association with sudden infant death syndrome (SIDS) [58].

Research conducted with UK midwives for the Institute of Alcohol Studies found most midwives advised against drinking during pregnancy. However, knowledge of the Chief Medical Officers’ guidelines was lacking as these guidelines had not been communicated to midwives and there was limited training on alcohol available. The research also identified there is no standardised approach to addressing alcohol consumption during antenatal appointments, meaning assessment and recording of alcohol consumption is inconsistent across the UK and within each country [59].

Among teenagers, heavy episodic drinking is associated with an increased risk of becoming pregnant, with rates of teenage pregnancy in the UK the highest in Western Europe [60]. Teenage pregnancy and early motherhood has been associated with poorer long-term socioeconomic outcomes, though these are difficult to separate from initial disadvantage that also contribute to teenage pregnancy [61].

Foetal Alcohol Spectrum Disorders (FASD)

Alcohol reaches the foetus by crossing the placental barrier. FASD is an umbrella term for lifelong brain-based neurodevelopmental disorders that covers: Foetal Alcohol Syndrome (FAS), Alcohol-Related Neurodevelopmental Disorder, Alcohol-Related Birth Defects, Foetal Alcohol Effects and partial Foetal Alcohol Syndrome [62].

A major 2016 review identified there are over 400 medical conditions that co-occur with FASD, affecting nearly every system of the body, including the central nervous system, vision, hearing, cardiac, circulation, digestion, and musculoskeletal and respiratory systems, among others [63]. Those with FASD have also been found to experience higher rates of mental health problems, schooling disruption and trouble with the law [64] and are at increased risk of sleep problems [65]. Birth mothers of children with FASD often also face multiple disadvantage, including poverty, poor nutrition, illicit drug use, smoking, violence and a history of obstetric problems [66].

FAS is the most severe form of FASD [67]. The NHS lists symptoms of FAS as [68]:

• a head that’s smaller than average
• poor growth – they may be smaller than average at birth, grow slowly as they get older, and be shorter than average as an adult
• distinctive facial features – such as small eyes, a thin upper lip, and a smooth area between the nose and upper lip, though these may become less noticeable with age (though the National Organisation for FASD highlights it is a misconception that these features are necessary for a diagnosis) [69]
• movement and balance problems
• learning difficulties – such as problems with thinking, speech, social skills, timekeeping, maths or memory
• issues with attention, concentration or hyperactivity
• problems with the liver, kidneys, heart or other organs
• hearing and vision problems

A recent study in England indicated as many as 17% of children screen positive for FASD [70], requiring assessment to determine whether a diagnosis should be made. While Scotland introduced guidelines to identify FASD in 2019 [71], these are lacking elsewhere in the UK.

Domestic violence

In the year ending March 2019, It is estimated that 1.6 million women had experienced domestic abuse (partner or family non-physical abuse, threats, force, sexual assault or stalking) [72]. The 2018 Crime Survey for England and Wales found seven in ten victims of partner abuse in the past year were women [73]. Women are even more likely to experience repeated or serious victimisation. For example, 89% of victims reporting four or more incidents are women, and women are more likely to have injuries that require medical attention than men [74].

The World Health Organization states alcohol consumption – especially at hazardous and harmful levels – is a major contributor to the occurrence of intimate partner violence and links between the two are manifold [75]. A report for Alcohol Research UK (now Alcohol Change) found two-thirds of ‘domestic’ incidents known to the police were found to involve at least one of the couple concerned being ‘under the influence’ of alcohol [76]. A large Swedish study found men with alcohol use disorder were 4.4 times more likely to perpetrate intimate partner violence against women compared with matched controls [77]. Some studies have also found gendered associations with socio-economic inequality and victimisation. For example, a US study found a significantly higher likelihood of crime victimisation in disadvantaged neighbourhoods for women, but not men [78].

Victims of domestic abuse may use alcohol as a coping mechanism and, in some cases, this may be used by violent partners as an excuse for continued abuse [79]. Around 10% of clients accessing domestic violence support services (95% of whom were women) had an ‘alcohol misuse need’ [80], but not all refuges and other services accept survivors with alcohol support needs.

Sexual assault

The 2017 Crime Survey for England and Wales found 20% of women have experienced some type of sexual assault since the age of 16, equivalent to an estimated 3.4 million female victims (compared with 4% of men, equivalent to 631,000 male victims) [81].

Use of alcohol by both victim and perpetrator is commonly implicated in sexual assault [82]. Of those in the 2017 Crime Survey for England and Wales who reported they had been victims of rape or assault by penetration since they were 16 years of age – 96% of whom were women – 38% of victims reported that the offender(s) were under the influence of alcohol. The same proportion of victims (38%) reported they were under the influence of alcohol themselves during this serious sexual assault [83].

In a 2011 report on alcohol and sexual health by the Royal College of Physicians identified evidence that perpetrators may actually seek out intoxicated women, and that blood alcohol levels in victims in 60% of rape cases raised questions as to whether the victim would be in a position to give consent [84].

Distributional effects of alcohol taxes

A 2020 IAS report on the distributional effects of UK alcohol taxes found that women pay 39% of alcohol duty, on average £160 per head per year (the equivalent figure for men is £250) [85]. When alcohol duty is looked at by beverage type, women pay the majority of wine duty – 55% – but relatively little beer or cider duty. This suggests that the government’s decision to cut taxes on particular alcohol drinks and not others has had distributional implications. In 2020/21, in real terms, compared with 2012/13, duty on beer, which overwhelmingly falls on men, has fallen by 19% in real terms. By contrast, wine duty has been cut by 3% over that period [86].

What can be done to reduce alcohol harm among women?

Action on gendered alcohol marketing

Typically ‘female’ drinks are produced to taste sweeter and have a lower alcohol content; examples include fruit beers, wines, and liqueurs [87]. Features of the product are highlighted that may be expected to appeal to a female audience, such as being “bloat resistant” [88]. There has been a social media response with the hashtag #dontpinkmydrink to draw attention to the use of gender stereotypes by alcohol producers and retailers [89].

A 2019 rapid review conducted for the Institute of Alcohol Studies identified the ways women are targeted by alcohol marketing, including [90]:

• Creation of new products
• Lifestyle messages that are underpinned by gender stereotype, eg slimness/weight, pink, all-female friendships
• Offers of stereotypical feminine accessories (eg makeup) and messages of empowerment
• Interactive techniques (eg competitions, photograph requests) on social media used to involve the public, including women, in content creation and to encourage interaction with and the sharing of brand content on social media platforms, in ways that are gendered, and in ways that create a wider audience reaction

This review also examined how women are represented in alcohol marketing. Gender roles ascribed to women have changed over time, but new representations of women as sexually active and empowered co-exist alongside their sexualisation and objectification in marketing. In the night time environment, nightlife venue marketing on social media uses women’s bodies and sexualities including photographs of female patrons in a way that reproduces the male gaze.

The Advertising Standards Authority introduced new codes prohibiting causing offence through the use of gender stereotypes in 2019 [91], but the effect of these is not yet known.

For more information on alcohol marketing, read our briefing on alcohol marketing.

A 2017 Women and Alcohol seminar series co-hosted by Scottish Health Action on Alcohol Problems and the Institute of Alcohol Studies generated the following recommendations [92]:

Recommendations for research

• Better collaboration between researchers, practitioners, women’s rights groups, and those with lived experience of alcohol-related harm.
• Research should be undertaken to identify interventions including small scale and local activities, which have been implemented in an attempt to reduce alcohol-related harm to women. This research will provide an understanding of the types and content of interventions available and can be used to establish a basis for research where the effectiveness of these interventions can be assessed.
• Rigorous research should be undertaken to improve understanding of how alcohol marketing is used in social media and how this and other forms of new technology could be regulated.
• Research should be undertaken to explore how the French ‘Loi Évin’ (a regulation on alcohol marketing) might be adapted to the UK context.

Recommendations for policy

• Population level policies that restrict price and availability of alcohol are needed.
• Restrictions should be in place for all forms of alcohol marketing, including online, which employ sexualised and disrespectful images and messaging relating to women.
• To combat exploitative marketing within the night-time economy, it may be beneficial to review licensing legislation and enforcement options.
• Drawing on research evidence, legislation comparable to the ‘Loi Évin’ model should be implemented.
• More needs to be done to educate women about the alcohol industry’s aims and how they are using marketing strategies which subvert feminism and manipulate women.
• Ensure that reliable and credible public health information about alcohol is available and accessible to all women. This information should be free from the influence of commercial operators.

Recommendations for service providers

• All alcohol-related services should aim to provide increased availability of/improved access to women only spaces.
• There should be increased availability of residential treatment and recovery support for women and children.
• There should be increased availability of services, including online, where women can access support, while remaining anonymous.

---

* Binge drinkers; exceeded six units on heaviest drinking day (in line with the government’s Alcohol Strategy, men are considered to have binged if they drank more than eight units of alcohol on their heaviest drinking day in the week before interview, and women if they drank more than six units.. If someone drank equally heavily on more than one day, they were asked about the most recent of these days).

** Other drinkers; did not exceed six units on heaviest drinking day.

*** non-opiate and alcohol: people who have problems with both non-opiate drugs and alcohol. alcohol only: people who have problems with alcohol but do not have problems with any other substances. For more details see ‘Figure 1: How people are classified into substance reporting group’ in Adult substance misuse treatment statistics 2018 to 2019: report

**** The Million Women Study was a national study of women’s health, involving more than one million UK women aged 50 and over, who were invited to participate between 1996 and 2001. It was the largest study of its kind in the world. For more information, visit <www.millionwomenstudy.org>