Professional Documents
Culture Documents
Bindg Drinking
Bindg Drinking
Bindg Drinking
Many efforts to prevent alcoholrelated harm are aimed at reducing risk drinking. This article
outlines the many conceptual and methodological challenges to defining risk drinking. It summarizes
recent evidence regarding associations of various aspects of alcohol consumption with chronic and
acute alcoholrelated harms, including mortality, morbidity, injury, and alcohol use disorders, and
summarizes the study designs most appropriate to defining risk thresholds for these types of harm. In
addition, it presents an international overview of lowrisk drinking guidelines from more than 20
countries, illustrating the wide range of interpretations of the scientific evidence related to risk
drinking. This article also explores the impact of drink size on defining risk drinking and describes
variation in what is considered to be a standard drink across populations. Actual and standard drink
sizes differ in the United States, and this discrepancy affects definitions of risk drinking and
prevention efforts. KEY WORDS: Alcohol consumption; alcohol use disorder; alcoholrelated harm; alcohol and
other drug use (AOD) use harm reduction; harm minimization; prevention of harm from AOD use; problematic
AOD use; prevention of problematic AOD use; AOD induced risk; attributable risk; risk thresholds; morbidity;
AOD risk mortality; AOD risk injury; standard drink; amount of AOD use; responsible AOD use
P
reventing alcoholrelated harm
does not necessarily require that lenges to arriving at a definition of risk with drinking?
risk drinking be defined. At the drinking. What aspects of alcohol consumption
population level, harm reduction can Perhaps the most essential challenge should be used to define risk drinking?
be achieved through numerous broad lies in determining the threshold that Should these vary according to the
measures that determine the price or discriminates lowrisk and risk type of harm (e.g., drinking volume
availability of beverage alcohol (Babor drinking. Is risk drinking any consump in relation to chronic conditions, and
et al. 2003). Measures such as these affect tion that corresponds to a significantly drinking pattern in relation to acute
drinkers at all consumption levels. higher level of harm than that experi alcoholrelated harm)? Should risk
Although there is inconsistent evidence enced by lifetime abstainers, or does drinking be defined in terms of con
as to whether their impact is greater sumption that reflects current alcohol
the harm have to be of a specified
among heavy or lighttomoderate related problems, as is the case with
magnitude? Given a linear relationship screening for alcohol use disorders
drinkers (Farrell et al. 2003; Gmel et al. between consumption and harm, where
2008; Heeb et al. 2003; Mkel et al. and emergencydepartment studies
is the appropriate cutoff point? Beyond of drinking in relation to the risk of
2008; Manning et al. 1995; Wagenaar this basic question, one must also ask
et al. 2009), such measures have proven injury? Or should it be defined in
what types of harms should be con terms of consumption that increases
to be effective in reducing problems
sidered. Excessive use of alcohol is the risk of developing alcoholrelated
associated with heavy or problem
associated with a wide range of harmful harm in the long term, as is the case
drinking (Wagenaar et al. 2009, 2010).
In contrast to such global approaches, outcomes, including alcohol use dis with prospective studies of alcohol
targeted approaches focus on preventing, orders; mortality and morbidity from related mortality and morbidity?
identifying, and modifying risk drinking chronic medical conditions, such as What types of studies are most
(i.e., drinking at levels or in patterns alcoholic liver disease, and acute causes, appropriate for assessing associations
that increase the risk of alcoholrelated such as vehicular crashes and accidental between different aspects of alcohol
harm). The development and dissemi and intentional injury; and a host of
nation of drinking guidelines that define social and legal problems. Should risk DEBORAH A. DAWSON, PH.D., is a
the limits of lowrisk alcohol consump drinking definitions be keyed more former staff scientist at the Laboratory
tion are one example of this type of closely to those types of harm most of Epidemiology and Biometry, National
prevention effort. Defining risk drink strongly attributable to alcohol use, Institute on Alcohol Abuse and
ing may seem simple compared with or to the most severe harms (i.e., Alcoholism, National Institutes of Health,
preventing it, but in fact there are many mortality or years of life lost) regardless Bethesda, Maryland.
consumption and alcoholrelated harm? on the most appropriate definition. was observed for pooled studies that
To what extent should we account for The following sections describe the excluded occasional and former drinkers
the quality of the consumption data evidence for associations of drinking from the reference group, irrespective
upon which evidence of alcoholrelated volume and pattern with alcoholrelated of the level of adjustment. Gender
harm is based? Many of the large harm, issues surrounding standard specific curves, regardless of the level
prospective studies used to assess drink size, and the conclusions drawn of control or reference group charac
mortality risk collect data on numerous by selected countries in defining risk teristics, showed that the risk of all
putative risk factors, and they often drinking in their national drinking cause mortality started to increase at
contain too few questions on alcohol guidelines. a lower ADV for women (approxi
use to yield estimates of consumption mately 35 g) than men (approximate
that fully capture the contribution of ly 45 g). However, it has been argued
heavy drinking days or multiple bev Association of Drinking that exclusion of studies biased by
erage types. If it is likely that associa Volume With Alcohol misclassification error with respect
tions of consumption with the risk of Related Harm to the abstainer category might have
harm are based on underestimates of yielded lower thresholds for harm
consumption, how should we account Drinking volume, generally character (Fillmore et al. 2007). In a meta
for that fact when using the data to ized in terms of average daily volume analysis of allcause mortality studies
inform definitions of risk drinking? (ADV) of alcohol intake, has been conducted before 2000, Gmel and
Finally, what is the appropriate cut widely studied in association with colleagues (2003) reported a significant
off between enough information and mortality and chronic disease morbid increase in risk relative to lifetime
too much? Should definitions of risk ity in large prospective cohort studies. abstainers at ADVs of 30 g to 50 g
drinking, or, conversely, lowrisk drink Because a full review of this extensive for women (relative risk1 [RR]: 1.40)
ing guidelines, be complex enough to literature lies beyond the scope of this and at ADVs of 40 g to 70 g for men
include volume and patternrelated article, this section will summarize (RR: 1.04), based on studies where
risks and their variation across popu the findings of selected metaanalyses the mean age of the respondents was
lation subgroups or should they be conducted in 2000 or more recently. at least 45 years at baseline.
simple enough so that drinkers can The focus will be on studies with In terms of diseasespecific morbidity
easily recall them and clinicians can doseresponse curves that can be used and mortality, much of the literature
easily identify risk drinkers based on to inform riskdrinking definitions, concerning coronary heart disease
a single metric? Many guidelines con rather than studies that summarize (CHD), stroke, and type 2 diabetes
tain different limits for men and associations by means of alcohol has centered on the debate concerning
women; others stipulate lower limits attributable fractions (i.e., the propor possible protective effects of moderate
for the youngest and oldest drinkers. tions of deaths from selected causes volumes of intake, but some meta
In addition, some guidelines explicitly attributable to drinking) or years of analyses have reported increased risks
mention groups of individuals for whom life lost (e.g., Gutjarh et al. 2001; Rehm of these diseases at heavy volumes of
any drinking is inadvisable (e.g., women et al. 2003, 2006, 2007, 2009). More consumption. A metaanalysis of 28
who are pregnant or trying to become over, because the focus is on the defi CHD studies by Corrao and colleagues
pregnant, people intending to drive nition of risk drinking, this article will (2004) indicated an increased risk
or operate complex machinery, or not examine levels of drinking volume compared with nondrinkers at an
individuals with medical conditions for which supposedly protective effects ADV of 89 g. Corrao and colleagues
or taking medication). These exceptions (i.e., risk levels lower than those of observed increased risks of hemor
to the general guidelines might also abstainers) have been observed. rhagic stroke at ADVs of 50 g and
include individuals with former alcohol Di Castelnuovo and colleagues higher in the metaanalyses of six
problems or those with a history of (2006) conducted a metaanalysis studies and of ischemic stroke at an
treatment for alcohol use disorders. of 34 prospective studies of allcause ADV of 100 g in the metaanalyses
Is it appropriate to use the same defi mortality and established 56 indepen of three studies. These findings for
nition of risk drinking for all prevention dent risk curves reflecting various stroke risk are consistent with those
efforts, or should the context determine models and population subgroups. of Reynolds and colleagues (2003),
the relative emphasis on different aspects Visual inspection of the pooled adjusted who conducted a metaanalysis of 35
of risk drinking? How should we risk curves for studies that presented cohort and casecontrol studies of
account for variation across beverages both unadjusted and adjusted associa stroke risk. They reported a signifi
and drinkers in drink size and alcohol tions, including confidence intervals, cantly increased risk of all types of
content when defining risk drinking? suggests that the mortality risk began stroke at ADVs of 60 g and higher,
These questions provide some notion to significantly exceed the level of
1
of the complex challenges posed in nondrinkers at an ADV of approxi Relative risk is a ratio of the probability of the event occurring
in the exposed group versus a nonexposed group. For example,
defining risk drinking and illustrate mately 38 g of ethanol (or about 2.7 a relative risk of 10 means that the exposed group is 10 times
why there is no absolute consensus standard drinks). A similar threshold more likely than the nonexposed group to experience the event.
than men (Cherpitel et al. 2006). In related injury as opposed to an unin association with usual consumption
contrast, an Australian casecontrol tentional one were increased by a fac of five or more drinks among U.S.
study of individuals hospitalized for tor of 5.5 at BACs of 0.15 to 0.199 adults, but only among those who
injury paired with community con (Macdonald et al. 2005). Evidence drank this amount less than once a
trol subjects found significantly high of gender differences was mixed. month. In a study of singlevehicle
er risks of injury among women with Associations of drinking in the event motor vehicle crashes, Heng and col
an intheevent intake of more than with violent versus unintentional injury leagues (2006) reported that the risk
60 g of ethanol (Stockwell et al. 2002). were greater for men than women in of fatality was significantly increased
In general, the risk of injury is more Argentina, Belarus, and Spain but even at BACs associated with fairly
strongly associated with drinking in greater for women than for men in low levels of intheevent consump
the event than with regular drinking China. There were no gender differ tion (e.g., at BACs as low as 0.010
patterns. In fact, a metaanalysis of ences with respect to drinking in the to 0.019 for drinkers ages 16 to 20).
emergencydepartment studies demon event in the United States, but the Although most studies of chronic
strated that pooled attributable risk association between frequent HED disease and allcause mortality have
sizes were 43 percent for drinking and violent injury was greater among focused on the association with volume
in the event compared with just 27 American women than men (OR: of ethanol intake, as described previ
percent for usual drinking pattern 4.52 vs. 1.63) (Wells et al. 2007). ously, a limited number of studies
(Cherpitel et al. 2005). Pooled data Because deaths from external causes have examined associations with
from the ERCAAP showed that indi generally reflect drinking in the event drinking pattern measures, primarily
viduals who tested positive for drinking of a fatal injury, analyses of the role HED. Tolsrup and colleagues (2004)
in the event were more than 50 percent of alcohol in such deaths have focused found that the allcause mortality risk
more likely to present for an injury more strongly on drinking pattern associated with drinking 21 or more
as opposed to a medical problem than average volume of ethanol intake. drinks a week was greater among
(Cherpitel et al. 2003). In a prospective study of Russian men people with infrequent as opposed to
Individual case crossover studies, aged 25 to 64 years, with an average frequent intake (the former implying
in which an individuals selfreported followup of 9.5 years, usual con more drinks per drinking occasion)
regular drinking pattern during some sumption of at least 160 g of alcohol in a prospective Danish cohort study.
specified period is used as the control per occasion increased the risk of On the basis of a Finnish cohort
for his or her selfreported drinking death from external causes by a factor study of men ages 25 to 64, Laatikainen
in the event (Maclure 1991), have of 2.08 compared with a usual con and colleagues (2003) found that the
shown three to fourfold increases in sumption of less than 80 g among prospective risk of allcause mortality
the risk of injury in association with individuals who drank at least once was 57 percent higher among men
drinking in the event (Borges et al. a month. A similar prospective study who had consumed six or more drinks
2004; Vinson et al. 2003a). One study of Finnish men found that those who at a time than among those who had
showed the excess risk increasing directly usually drank six or more bottles of not, even after controlling for volume
with the number of drinks consumed beer per drinking occasion had a far of consumption. Another Finnish
from an OR of 1.8 for one to two higher risk of death from external cohort study reported that men who
drinks to an OR of 17.0 for seven or causes than those who usually con usually drank six or more beers per
more drinks (Vinson et al. 2003b). sumed less than three bottles (RR: occasion had higher risks of allcause
Pooled data from 28 emergency 7.10), even after adjusting for total mortality and fatal myocardial infarc
department studies in 16 countries consumption (Kauhanen et al. 1997). tion than those who usually consumed
showed that the random pooled effect No significant increase in risk was fewer than three beers (RR: 3.01 and
of drinking in the event compared with observed at lower usual levels of intake 6.50, respectively), independent of
usual drinking was an increase of 5.69 (Malyutina et al. 2002). In a study their total volume of consumption
in the likelihood of injury (Borges et of fatal injury that entailed matching (Kauhanen et al. 1997). Mkel and
al. 2006). Associations with drinking death records with data from a series colleagues (2005), who linked Finnish
in the event are even stronger for of Finnish alcohol surveys, consuming alcohol survey participants with death
violencerelated injuries than for all five or more drinks 25 to 52 times records, found an increased risk of
injuries, with a case crossover study per year and more than 52 times per allcause mortality among men in
showing a 34fold increase in the risk year were associated with fatal injury association with a high volume of
for a violencerelated injury associated RRs of 2.63 and 5.78, respectively, ethanol intake consumed on heavy
with drinking in the event relative to relative to never consuming five or drinking occasions, but not in associ
drinking the previous day and a 10 more drinks, even after adjusting for ation with a high volume consumed
fold increase relative to drinking in frequencies of drinking fewer than five on lighterdrinking occasions. This
the previous month (Vinson et al. drinks (Paljrvi et al. 2005). Dawson relationship did not extend to women.
2003a). The ERCAAP data also (2001) also found an increased risk In a 3.8year followup of patients
indicated that the odds of a violence of mortality from external causes in hospitalized for myocardial infarction,
emergency care settings. Several brief drinking among women and the elderly ICAPReports/tabid/75/Default.aspx).
screeners containing both alcohol (Berner et al. 2007; Dawson et al. The various guidelines differ among
consumption and alcohol problem 2005, in press; Kelly et al. 2009). A countries not only in terms of the
items have shown high levels of sensi recent test of a singleitem screener maximum permissible numbers of
tivity (the proportion of individuals based on maximum drinks consumed drinks but also in terms of what types
with the AUD outcome in question likewise found variation across sub of limits are included (daily, weekly,
who screen positive) and specificity groups. By gender, the cut point that or both) and whether there are different
(the proportion of individuals without maximized sensitivity and specificity guidelines for men and women (see
the AUD outcome in question who for any AUD or any AUD/hazardous table). Although most countries
screen negative) for AUD and/or haz drinking was five or more drinks for drinking guidelines are expressed solely
ardous drinking (Berner et al. 2007; men and four or more drinks for in terms of daily limits, Denmark,
Kelley et al. 2009). These include the women (Dawson et al., in press), Finland, Ireland, and South Africa
Alcohol Use Disorders Identification thus providing support for the gender have weekly drinking limits only.
Test (AUDIT; Saunders et al. 1993) specific five or more/four or more Canada, New Zealand, Poland, the
and the Rapid Alcohol Problems drinks definition of risk drinking that United Kingdom, and the United
ScreenQuantity Frequency (RAPS4 has come into common use in U.S. States include both daily and weekly
QF; Cherpitel 2002). The AUDITC, surveys of drinking practices and limits. Australia and Slovenia provide
containing only the three AUDIT problems (Wechsler and Nelson limits for consumption on any day
consumption questions on overall 2001; Wechsler et al. 1995). (which when multiplied times seven
frequency of drinking, usual quantity roughly correspond to weekly limits),
of drinks consumed on drinking days, with higher limits for any single
and frequency of heavy drinking, has International LowRisk drinking occasion. The Australian
proven nearly as effective as the full Drinking Guidelines guidelines explicitly state that the
10question AUDIT in screening for former are targeting the risk of
AUDs and risk drinking in the general Perhaps the best illustration of the chronic conditions and the latter the
population and subpopulations such complexity of defining risk drinking acute consequences of heavy drink
as veterans and patient samples can be obtained by comparing interna ing. The United States, Canada, Italy,
(Aertgeerts et al. 2001; Bradley et tional drinking guidelines. A number and South Africa all have definitions
al. 2003; Bush et al. 1998; Dawson of countries have systematically formu of moderate drinking that are more
et al. 2005b; Gordon et al. 2001; lated lowrisk drinking guidelines with restrictive than their thresholds for
Gual et al. 2002; Rumpf et al. 2002). the input of expert committees of lowrisk drinking, and Spains guide
More recently, studies of a singleitem researchers who have conducted exten lines include regional variations (i.e.,
screening instrument based solely on sive reviews of the scientific literature. considerably higher limits in the
the frequency of heavy drinking also By providing the upper limits of low Basque country and Catalonia than
have reported high levels of sensitivity risk drinking, these guidelines implic in the overall national guidelines). In
and specificity in screening for AUDs itly reveal various countries definitions general, lowrisk drinking guidelines
and risk drinking in trauma center of risk drinking (i.e., consumption stipulate the upper limit of lowrisk
and emergencydepartment samples beyond the lowrisk limits). Two reports consumption, but a few countries
(Canagasaby and Vinson 2005; describing recent changes to the Australian (i.e., Japan, Portugal, Romania, Spain
Dawson et al. 2010; Seale et al. 2006; and Canadian drinking guidelines [Catalonia], the United Kingdom,
Smith et al. 2009; Stewart et al. 2008; provide the rationales used for setting and the United States [Dietary
Taj et al. 1998; Williams and Vinson the guidelines in those countries. They Guidelines moderate drinking defini
2001). In a sample of U.S. adults, illustrate the broad range of evidence tion for men]) provide a range of
drinking five or more drinks (men) typically considered in establishing these acceptable values.
or four or more drinks (women) at guidelines and the diverse approaches Many of the differences across
least once in the preceding year resulted (e.g., relative versus absolute risk) that countries in the specific numbers of
in a sensitivity and specificity of 86.7 may be applied to the interpretation of drinks comprising daily or weekly
percent and 82.1 percent, respectively, this evidence (National Health and limits reflect variation in the standard
in predicting DSMIV alcohol abuse Medical Research Council, http://www. drink size used to express the daily
and/or dependence (Dawson et al. nhmrc.gov.au/_files_nhmrc/file/public and/or weekly limits. The standard
2010). Many of the studies of the ations/synopses/ds10alcohol.pdf; drink size assumed by the U.S. drink
AUDIT, AUDITC, and other brief Stockwell et al., in press). ing guidelines (0.6 oz or approximately
screening instruments have noted Lowrisk drinking guidelines vary 14 g of ethanol) is almost twice as
differential performance across sub substantially across countries, as is large as the standard drink size of 8 g
groups of the general U.S. population, evident in an online listing of these used by the United Kingdom. A stan
often supporting lower screening guidelines that is continually updated dard drink size of 10 g is used in
score thresholds for detecting problem (see http://www.icap.org/Publications/ Australia and most of Europe other
but sensitivity was accordingly reduced The Effect of Drink Size Unfortunately, all evidence suggests
(30.8 percent to 70.6 percent). In on Definitions of Risk that this is not the case. In an exhaus
multivariate models adjusted for Drinking tive review of 32 studies related to
sociodemographic characteristics, the drink size published through 2007,
odds of all of the outcomes except for Despite the abundant evidence of DevosComby and Lange (2008)
peptic ulcer were significantly increased chronic and acute alcoholrelated harm found that drinkers often were
among drinkers who exceeded the at various levels of average daily or per unaware of how standard drinks were
weekly or daily limits, regardless of occasion ethanol intake, converting risk defined in their countries and that
the frequency of the latter. It is inter thresholds into a comprehensible defi actual drink sizes (or attempts to
esting to note that the odds ratios nition of risk drinking must ultimately pour a standard drink) often exceeded
virtually were identical when consider confront the issue of drink size. The standard drink sizes. The magnitude
ing only whether the drinkers exceed amount of ethanol contained in an of the discrepancy varied substantially
ed the daily limits. That is, little alcoholic drink varies considerably across studies and was associated with
additional information on risk was depending on the type of alcohol (e.g., study design, drinker characteristics,
obtained by considering the weekly beer, wine, or spirits) and on the size of type of beverage, and vessel size. More
as well as daily limits. A more recent the drink. Within the major categories recent U.S. research confirmed that
evaluation of the U.S. guidelines con of alcoholic beverages, there are significant vessel size was more important than
sidered pastyear alcohol consumption variations according to beverage subtype. shape in determining the size of
relative to multiple concurrent and Malt liquor, with a typical ethanol con drink pours (Kerr et al. 2009a) and
prospective harms and found that the tent of at least 6.0 percent alcohol by that largerthanstandard drinks were
thresholds that optimized prediction volume (ABV), is far stronger than light common even in bar and restaurant
of concurrent harm (i.e., the upper beer, with a typical ABV of about 4.2 drinks (Kerr et al. 2008). Although
limits of what might be considered percent. Likewise, fortified wines have largerthanstandard drink sizes are
lowrisk drinking) consisted of 4 drinks an ABV that is about 50 percent greater the major concern in the prevention
a day for men and 3 drinks a day for than that of regular table wine or of risk drinking, it should be noted
women (4/3), alone or in combina champagne, approximately 18 percent that a significant proportion of
versus 12 percent (Kerr et al. 2006a). drinkers consume smallerthanstandard
tion with weekly limits of 21 drinks
Moreover, spirits such as whiskey, vodka, drinks. In fact, Kerr and colleagues
for men and women. Prospective
and gin have an ABV that is greater (2005, 2009b) have shown that there
harms were best predicted by weekly
than that of cordials and liqueurs and is a great deal of dispersion in the
limits of 14/7 (men and women, far greater than that of prepackaged distribution of actual drink sizes and
respectively), 14/14 and 10/10 drinks, cocktails (Kerr et al. 2006b). To account that the degree of dispersion varies by
all combined with daily limits of 4 for differences such as these, drinking beverage (smallest for beer and larger
drinks for both men and women guidelines often provide examples of for wine and spirits) and by demo
(Dawson et al., in press). Using a what constitutes a standard drink. The graphic characteristics (smaller for
prospective framework, Batty and National Institute on Alcohol Abuse men and Whites and larger for women
colleagues (2009) recently examined and Alcoholism (NIAAA) lowrisk and minorities). Thus, it must be
the effect of exceeding the daily and drinking guidelines for the United understood that many drinkers will
weekly U.K. drinking limits on the States, for example, define a standard interpret drinking guidelines in terms
occurrence of various harms over the drink containing 0.6 oz or 14 g of of numbers of drinks that correspond
course of a 3.6year followup inter ethanol as the equivalent of 12 oz of to levels of intake that are smaller
val. They reported that exceeding the beer or wine cooler; 8 oz to9 oz of malt or larger than those intended by the
daily limits was associated with an liquor; 5 oz of table wine; 3 oz to 4 oz standard drink definitions included
increased risk of hypertension, whereas of fortified wine; 2 oz to 3 oz of cor in the guidelines. In light of this, it
exceeding the weekly limits was asso dial, liqueur, or aperitif; and 1.5 oz of might be argued that standard drink
ciated with an increased risk of financial brandy, whiskey, vodka, etc. (NIAAA sizes for any given country should
problems. Of interest is a nearsignificant 2005; see also http://rethinkingdrinking. reflect the most common container
association with accidents occurred niaaa.nih.gov/). They also provide or serve sizes in that country, even
with respect to exceeding the weekly information on the number of standard if this leads to lack of comparability
rather than daily limits, suggesting drinks in commonly sold container across countries. That is, the standard
that the intheevent levels of con sizes such as a 40oz can of malt liquor, drink definitions that maximize pre
sumption typically associated with a 750mL bottle of table wine, and a vention efforts may not be those best
injuries did not significantly increase fifth of spirits. suited for comparative research pur
the risk of accidents unless they were Does this ensure that drinkers poses. Research addressing how
consumed often enough to yield a understand the concept of a standard guidelines are understood by drinkers
volume of intake that exceeded the drink and know how many standard who typically pour nonstandard
weekly drinking limits. drinks they typically consume? drinks might help to improve the
Regarding the perplexing and chal American Psychiatric Association. Diagnostic and BUSH, K.; KIVLAHAN, D.R.; MCDONELL, M.B.; ET
Statistical Manual of Mental Disorders, Fourth AL. The AUDIT alcohol consumption questions
lenging issue of drinkers inability to (AUDITC): An effective brief screening test for
Edition. Washington, DC: American Psychiatric
accurately gauge their consumption Association, 1994. problem drinking. Ambulatory Care Quality
in standard drinks, the most obvious Improvement Project (ACQUIP). Alcohol Use
solution lies in the approach that has BABOR, T.F.; CAETANO R.; CASSWELL, S.; ET AL. Disorders Identification Test. Archives of Internal
Alcohol: No Ordinary Commodity. Oxford, United Medicine 158:17891795, 1998. PMID: 9738608
been adopted by a number of Western Kingdom: Research and Public Policy, 2003.
countries, in which alcoholic beverage CAETANO, R.; TAM, T.; GREENFIELD, T.; ET AL.
containers explicitly state how many BAGNARDI, V.; ZATONSKI, W.; SCOTTI, L.; ET AL. DSMIV alcohol dependence and drinking in the
Does drinking pattern modify the effect of alcohol U.S. population: A risk analysis. Annals of Epidemiology
standard drinks (units) they contain. on the risk of coronary heart disease? Evidence 7:542549, 1997. PMID: 9408550
Even in the absence of such labeling, from a metaanalysis. Journal of Epidemiology and
it has been argued that if risks attributed Community Health 62:615619, 2008. PMID: CAIRNEY, S.; CLOUGH, A.; JARAGBA, M.; AND
to drinking five or more drinks are 18559444 MARUFF, P. Cognitive impairment in Aboriginal
people with heavy episodic patterns of alcohol use.
based on scientific evidence relying BAILEY, B.A., AND SOKOL, R.J. Pregnancy and alco Addiction 102:909915, 2007. PMID: 17523985
on actual as opposed to standard drink hol use: Evidence and recommendations for prena
CANAGASABY, A., AND VINSON, D.C. Screening for
sizes, coupled with other sources of tal care. Clinical Obstetrics and Gynecology 51:436
hazardous or harmful drinking using one or two
consumption underreporting, then 444, 2008. PMID: 18463472
quantityfrequency questions. Alcohol and
drinking less than five drinks, irre BALIUNAS, D.O.; TAYLOR, B.J.; IRVING, H.; ET AL. Alcoholism 40:208213, 2005. PMID: 15797883
spective of how closely they correspond Alcohol as a risk factor for type 2 diabetes: A sys CARLSSON, S.; HAMMAR, N.; AND GRILL, V.
to standard drink size, will reduce tematic review and metaanalysis. Diabetes Care Alcohol consumption and type 2 diabetes: Meta
harm in the aggregate. That is, if one 32:21232132, 2009. PMID: 19875607 analysis of epidemiological studies indicates a
assumes that relative risks associated BATTY, G.D.; LEWARS, H.; EMSLIE, C.; ET AL. shaped relationship. Diabetologia 48:10511054,
with various consumption levels are Internationally recognized guidelines for sensible 2005. PMID: 15864527
overstated because of underreporting alcohol consumption: Is exceeding them actually CHERPITEL, C.J. Alcohol and injuries: A review of
of consumption, then adherence to detrimental to health and social circumstances? international emergency room studies since 1995.
Evidence from a populationbased cohort study. Drug and Alcohol Review 26:201214, 2007.
lowrisk drinking limits should prove Journal of Public Health (Oxford) 31:360365, PMID: 17364856
effective even for individuals whose 2009. PMID: 19574275
actual drink sizes are larger than stan CHERPITEL, C.J. Screening of alcohol problems in
BECKER, U.; DEIS, A.; SRENSON, T.I.; ET AL. the U.S. general population: Comparison of the
dard. Hence, publicizing lowrisk Prediction of risk of liver disease by alcohol intake, CAGE, RAPS4 and RAPS4QF by gender, ethnic
drinking limits should play an impor sex, and age: A prospective population study. ity and services utilization. Alcoholism: Clinical and
tant role in any activities aimed at Hepatology 23:10251029, 1996. PMID: 8621128 Experimental Research 26:16861691, 2002.
preventing alcoholrelated harm. BERNER, M.M.; KRISTON, L.; BENTELE, M.; AND
PMID: 12436057
HARTER, M. The Alcohol Use Disorders Identification CHERPITEL, C.J.; BOND, J.; AND YE, Y. Alcohol and
Test for detecting atrisk drinking: A systematic review injury: A risk function analysis from the Emergency
and metaanalysis. Journal of Studies on Alcohol and Room Collaborative Alcohol Analysis Project
Acknowledgements (ERCAAP). European Addiction Research 12:4252,
Drugs 68:461473, 2007. PMID: 17446987
2006. PMID: 16352902
This research was supported in part by BORGES, G.; CHERPITEL, C.; AND MITTLEMAN, M.
the Intramural Program of the National Risk of injury after alcohol consumption: A case CHERPITEL, C.J.; YE, Y.; AND BOND, J. Attributable
Institutes of Health, National Institute crossover study in the emergency room. Social Science risk of injury associated with alcohol use: Cross
& Medicine 58:11911200, 2004. PMID: 14723913 national data from the Emergency Room Collaborative
on Alcohol Abuse and Alcoholism. Alcohol Analysis Project. American Journal of Public
BORGES, G.; CHERPITEL, C.J.; OROZCO, R.; ET AL. Health 95:266272, 2005. PMID: 15671463
Acute alcohol use and the risk of nonfatal injury in
CHERPITEL, C.J.; BOND, J.; YE, Y.; ET AL. A cross
sixteen countries. Addiction 101:9931002, 2006.
Financial Disclosure PMID: 16771891
national metaanalysis of alcohol and injury: Data
from the Emergency Room Collaborative Alcohol
The author declares that she has no BRADLEY, K.A.; BUSH, K.R.; EPLER, A.J.; ET AL. Analysis Project (ERCAAP). Addiction 98:1277
competing financial interests. Two brief alcoholscreening tests from the Alcohol 1286, 2003. PMID: 12930215
Use Disorders Identification Test (AUDIT): CHO, E.; SMITHWARNER, S.A.; RITZ, J.; ET AL.
Validation in a female Veterans Affairs patient pop Alcohol intake and colorectal cancer: A pooled
ulation. Archives of Internal Medicine 163:821829, analysis of 8 cohort studies. Annals of Internal
References 2003. PMID: 12695273 Medicine 140:603613, 2004. PMID: 15096331
AERTGEERTS, B.; BUNTIX, F.; ANSOMS, S.; AND BREITMEIER, D.; SEELANDSCHULZE, I.; HECKER, CHONG, E.W.; KREIS, A.J.; WONG, T.Y.; ET AL.
FEVERY, J. Screening properties of questionnaires H.; AND SCHNEIDER, U. The influence of blood Alcohol consumption and the risk of agerelated
and laboratory tests for the detection of alcohol alcohol concentrations of around 0.03% on neu macular degeneration: A systematic review and
abuse of dependence in a general practice popula ropsychological functions: A doubleblind, placebo metaanalysis. American Journal of Ophthamology
tion. British Journal of General Practice controlled investigation. Addiction Biology 145:707715, 2008. PMID: 18242575
51:206217, 2001. PMID: 11255902 12:183189, 2007. PMID: 17508991
CONNOR, J.; YOU, R.; AND CASSWELL, S. Alcohol
ALTIERI, A.; GARAVELLO, W.; BOSETTI, C.; ET AL. BREWER, R.D., AND SWAHN, M.H. Binge drinking related harm to others: A survey of physical and sex
Alcohol consumption and risk of laryngeal cancer. and violence. JAMA: Journal of the American Medical ual assault in New Zealand. New Zealand Medical
Oral Oncology 41:956965, 2005. PMID: 15927525 Association 294:616618, 2005. PMID: 16077057 Journal 122:1020, 2009. PMID: 19851416
Clinical and Experimental Research 30:516522, 1988 US national sample. Addiction REHM, J.; TAYLOR, B.; ROERECKE, M.; AND PATRA,
2006a. PMID: 16499493 91:14391456, 1996. PMID: 8917925 J. Alcohol consumption and alcoholattributable
burden of disease in Switzerland, 2002. International
KERR, W.C.; PATTERSON, D.; KOENEN, M.A.; AND MUKAMAL, K.J.; MACLURE, M.; MULLER, J.E.; AND Journal of Public Health 52:383392, 2007. PMID:
GREENFIELD, T.K. Alcohol content variation of bar MITTLEMAN, M.A. Binge drinking and mortality 18369001
and restaurant drinks in Northern California. after acute myocardial infarction. Circulation
Alcoholism: Clinical and Experimental Research 112:38393845, 2005. PMID: 16365208 RIVARA, F.P.; GARRISON, M.M.; EBEL, B.; ET AL.
32:16231629, 2008. PMID: 18616674 Mortality attributed to harmful drinking in the
MURRAY, R.P.; CONNETT, J.E.; TYAS, S.L.; ET AL. United States, 2000. Journal of Studies on Alcohol
KERR, W.C.; PATTERSON, D.; KOENEN, M.A.; AND Alcohol volume, drinking pattern, and cardiovascu
lar disease morbidity and mortality: Is there a
65:530536, 2004. PMID: 15376828
GREENFIELD, T.K. Large drinks are no mistake:
Glass size, not shape, affects alcoholic beverage shaped function? American Journal of Epidemiology RUMPF, H.J.; HAPKE, U.; MEYER, C.; AND JOHN,
drink pours. Drug and Alcohol Review 28:360365, 155:242248, 2002. PMID: 11821249 U. Screening for alcohol use disorders and atrisk
2009a. PMID: 19594789 drinking in the general population: Psychometric
National Institute on Alcohol Abuse and
performance of three questionnaires. Alcohol and
KOPPES, L.L.; DEKKER, J.M.; HENDRIKS, H.F.; ET Alcoholism (NIAAA). Drinking in the United
Alcoholism 37:261268, 2002. PMID: 12003915
AL. Moderate alcohol consumption lowers the risk States: Main Findings from the 1992 National
of type 2 diabetes: A metaanalysis of prospective Longitudinal Alcohol Epidemiologic Survey (NLAES). SAHA, T.D.; STINSON, F.S.; AND GRANT, B.F. The
observational studies. Diabetes Care 28:719725, U.S. Alcohol Epidemiologic Data Reference Manual, role of alcohol consumption in future classifications
2005. PMID: 15735217 Vol. 6, 1st Edition. Bethesda, MD: NIAAA, 1998. of alcohol use disorders. Drug and Alcohol
Dependence 89:8292, 2007. PMID: 17240085
LAATIKAINEN, T.; MANNINEN, L.; POIKOLAINAN, NIAAA. Helping Patients Who Drink Too Much: A
K.; AND VARTIAINEN, E. Increased mortality related Clinicians Guide, Updated 2005 Edition. Bethesda, SAUNDERS, J.B.; AASLAND, O.G.; BABOR, T.F.; ET
to heavy alcohol intake pattern. Journal of Epidemiology MD: NIAAA, 2005. AL.Development of the Alcohol Use Disorders
and Community Health 57:379384, 2003. PMID: Identification Test (AUDIT): WHO Collaborative
12700224 NIAAA. Rethinking Drinking: Alcohol and Your
Health. Available at: http://rethinkingdrinking. Project on Early Detection of Persons with
LANE, S.D.; CHEREK, D.R.; PIETRAS, C.J.; AND niaaa.nih.gov/. Accessed December 7, 2009. Harmful Alcohol Consumption, II. Addiction
TCHEREMISSINE, O.V. Alcohol effects on human 88:791804, 1993. PMID: 8329970
risk taking. Psychopharmacology 172:6877, 2004. NEALE, D.J., AND FROMME, K. Eventlevel covaria
tion of alcohol intoxication and behavioral risks SEALE, J.P.; BOLTRI, J.M.; SHELLENBERGER, S.; ET
PMID: 14647967 AL.Primary care validation of a single screening
during the first year of college. Journal of Consulting
LIVINGSTON, M., AND ROOM, R. Variations by age and Clinical Psychology 75:294306, 2007. PMID: question for drinkers. Journal of Studies on Alcohol
and sex in alcoholrelated problematic behavior per 17469887 67:778784, 2006. PMID: 16847548
drinking volume and heavier drinking occasion. SMITH, P.C.; SCHMIDT, S.M.; ALLENWORTH
Drug and Alcohol Dependence 101:169175, 2009. OKORO, C.A.; BREWER, R.D.; NAIMI, T.S.; ET AL.
Binge drinking and healthrelated quality of life: DAVIES, D.; AND SAITZ, R. Primary care validation
PMID: 19233572 of a singlequestion alcohol screening test. Journal
Do popular perceptions match reality? American
MACDONALD, S.; CHERPITEL, C.J.; BORGES, G.; ET Journal of Preventive Medicine 26:230233, 2004. of General Internal Medicine 24:783788, 2009.
AL. The criteria for causation of alcohol in violent PMID: 15026103 PMID: 19247718
injuries based on emergency room data from six STEWART, S.H.; BORG, K.T.; AND MILLER, P.M.
countries. Addictive Behaviors 30:103113, 2005. PLJARIV, T.; MKEL, P.; AND POIKOLAINEN, K.
Pattern of drinking and fatal injury: A population Prevalence of problem drinking and characteristics
PMID: 15561452 of a singlequestion screen. Journal of Emergency
based followup study of Finnish men. Addiction
MKEL, P.; BLOOMFIELD, K.; GUSTAFSSON, N.K.; 100:18511859, 2005. PMID: 16367986 Medicine 39(3):291295, 2010. PMID: 18534807
ET AL. Changes in volume of drinking after changes
PLETCHER, M.J.; VAROSY, P.; KIEFE, C.I.; ET AL. STICKLEY, A., AND PRIDEMORE, W.A. The effects of
in alcohol taxes and travellers allowances: Results binge drinking and social capital on violent victimi
from a panel study. Addiction 103: 181191, 2008. Alcohol consumption, binge drinking, and early
coronary calcification: Findings from the Coronary sation: Findings from Moscow. Journal of Epidemiology
PMID: 18028522
Artery Risk Development in Young Adults (CAR and Community Health 64(10):902907, 2010.
MKEL, P.; PALJRVI, T.; AND POIKOLAINEN, K. DIA) Study. American Journal of Epidemiology PMID: 19822905
Heavy and nonheavy drinking occasions, allcause 161:423433, 2005. PMID: 15718478
STOCKWELL, T.; BUTT, P.; BEIRNESS, D.; GLIKSMAN,
and cardiovascular mortality and hospitalizations:
REHM, J., AND GMEL, G. Patterns of alcohol con L.; AND PARADIS, C. The basis for Canadas new
A followup study in a population with a low con
sumption and social consequences: Results from an lowrisk drinking guidelines: A relative risk
sumption level. Journal of Studies on Alcohol
8year followup study in Switzerland. Addiction approach to estimating hazardous levels and patterns
66:722728, 2005. PMID: 16459933
94:899912, 1999. PMID: 10665078 of alcohol use. Drug and Alcohol Review, in press.
MALYUTINA, S.; BOBAK, M.; KURILOVITCH, S.; ET
REHM, J.; MONGA, N.; ADLAF, E.; ET AL. School STOCKWELL, T.; MCLEOD, R.; STEVENS, M.; ET AL.
AL. Relation between heavy and binge drinking and
matters: Drinking dimensions and their effects on Alcohol consumption, setting, gender and activity
allcause and cardiovascular mortality in
alcoholrelated problems among Ontario secondary as predictors of injury: A populationbased case
Novosibirsk, Russia: A prospective cohort study.
school students. Alcohol and Alcoholism 40:569 control study. Journal of Studies on Alcohol and
Lancet 360:14481454, 2002. PMID: 12433511
574, 2005. PMID: 16157609 Drugs 63:372379, 2002. PMID: 12086138
MANNING, W.G.; BLUMBERG, L.; AND MOULTON,
L.H. The demand for alcohol: The differential REHM, J.; PATRA, J.; AND POPOVA, S. Alcohol TAJ, N.; DEVERASALES, A.; AND VINSON, D.C.
response to price. Journal of Health Economics attributable mortality and potential years of life lost in Screening for problem drinking: Does a single
14:123148, 1995. PMID: 10154654 Canada 2001: Implications for prevention and policy. question work? Journal of Family Practice
Addiction 101:373384, 2006. PMID: 16499510 46:328335, 1998. PMID: 9564375
MARSHAL, M.P. For better or worse? The effects of
alcohol use on marital functioning. Clinical Psychology REHM, J.; ROOM, R.; GRAHAM, K.; ET AL. The rela TAYLOR, B.; IRVING, H.M.; BALIUNAS, D.; ET AL.
Review 23:959997, 2003. PMID: 14624823 tionship of average volume of alcohol consumption Alcohol and hypertension: Gender differences in
and patterns of drinking to burden of disease: An doseresponse relationships determined through
MIDANIK, L.T.; GREENFIELD, T.K.; AND CAETANO, overview. Addiction 98:12091228, 2003. PMID: systematic review and metaanalysis. Addiction
R. Risk functions for alcoholrelated problems in a 12930209 104:19811990, 2009. PMID: 19804464
Online Newsletter
NIAAA
U.S. Department
of Health and
Human Services
National Institutes
of Health