Update on Alcohol and Health

www.alcoholandhealth.orgwww.alcoholandhealth.org 11

Update on Update on Alcohol and HealthAlcohol and Health

Alcohol and Health: Current EvidenceAlcohol and Health: Current Evidence

May–June 2005May–June 2005


Studies on Studies on Alcohol and Alcohol and

Health OutcomesHealth Outcomes


Alcohol May Increase Alcohol May Increase the Risk of Colon the Risk of Colon

CancerCancer

Su LJ, et al. Su LJ, et al. Nutr and Cancer.Nutr and Cancer. 2004;50(2):111–119. 2004;50(2):111–119.


Objectives/MethodsObjectives/Methods To examine the association between To examine the association between

alcohol intake and colon canceralcohol intake and colon cancer

Assessment of alcohol use and incident Assessment of alcohol use and incident cases of colon cancer among participants cases of colon cancer among participants of a national health and nutrition surveyof a national health and nutrition survey 10,418 participants followed for 10 years10,418 participants followed for 10 years

Analyses adjusted for colon cancer risk Analyses adjusted for colon cancer risk factorsfactors


Results Results 111 subjects developed colon cancer.111 subjects developed colon cancer.

Drinking an average of >=1 drinks per Drinking an average of >=1 drinks per day, compared with abstinence, day, compared with abstinence, significantly increased the risk of significantly increased the risk of developing colon cancer.developing colon cancer. Relative risk (RR) 1.7 Relative risk (RR) 1.7

The risk of colon cancer increased as The risk of colon cancer increased as alcohol intake increased…alcohol intake increased… from 0 drinks to <1 drink to >=1 drinks per from 0 drinks to <1 drink to >=1 drinks per

day (day (P P for trend=0.04). for trend=0.04).


Results (cont.)Results (cont.)

Regular drinking for >34 years, compared Regular drinking for >34 years, compared with abstinence, also significantly with abstinence, also significantly increased colon cancer risk (RR 1.7). increased colon cancer risk (RR 1.7).

In analyses stratified by type of alcohol, In analyses stratified by type of alcohol, the increase in risk was significant for the increase in risk was significant for liquor, but not beer or wine, consumption.liquor, but not beer or wine, consumption.


Conclusions/CommentsConclusions/Comments The results of this study suggest that alcohol The results of this study suggest that alcohol

may increase the risk of colon cancer. may increase the risk of colon cancer.

Although the authors propose several plausible Although the authors propose several plausible mechanisms, no specific effect of alcohol on mechanisms, no specific effect of alcohol on colon cancer carcinogenesis is known. colon cancer carcinogenesis is known.

Nonetheless, additional studies should be Nonetheless, additional studies should be conducted to confirm the observed association.conducted to confirm the observed association.


No U-Shaped Curve No U-Shaped Curve for Alcohol and for Alcohol and

All-Cause Mortality in All-Cause Mortality in Younger MenYounger Men

Arndt V, et al. Arndt V, et al. Ann Epidemiol.Ann Epidemiol. 2004;14(10):750– 2004;14(10):750–753. 753.


Objectives/MethodsObjectives/Methods To examine whether younger men experience To examine whether younger men experience

the same benefits of moderate drinking as do the same benefits of moderate drinking as do middle-aged menmiddle-aged men

Assessment of alcohol intake, vital status, and Assessment of alcohol intake, vital status, and all-cause mortality in 17,279 male construction all-cause mortality in 17,279 male construction workers in Germany who…workers in Germany who… were aged 25–64 yearswere aged 25–64 years underwent an occupational health exam at baselineunderwent an occupational health exam at baseline were followed for an average of 10 years were followed for an average of 10 years

Analyses adjusted for age, nationality, and Analyses adjusted for age, nationality, and smokingsmoking


ResultsResults During follow-up, 698 subjects died.During follow-up, 698 subjects died.

The relationship between alcohol consumption The relationship between alcohol consumption and all-cause mortality was…and all-cause mortality was…

J or U-shaped for men aged 35–64 years J or U-shaped for men aged 35–64 years (highest risk in nondrinkers and heavier daily (highest risk in nondrinkers and heavier daily drinkers), but drinkers), but

linear for men aged 25–34 years (mortality linear for men aged 25–34 years (mortality increased as intake increased, increased as intake increased, PP for for trend=0.02).trend=0.02).

RR 0.9 for 0 drinks per day compared with RR 0.9 for 0 drinks per day compared with occasional drinkingoccasional drinking

RR 2.5 for >=8 drinks per dayRR 2.5 for >=8 drinks per day



Results did not change substantially when Results did not change substantially when analyses were also adjusted for disorders analyses were also adjusted for disorders at baseline that could influence drinking at baseline that could influence drinking behavior and mortality (e.g., liver behavior and mortality (e.g., liver diseases, cancer). diseases, cancer).


Conclusions/CommentsConclusions/Comments This study showed that moderate alcohol This study showed that moderate alcohol

consumption may lower all-cause mortality in consumption may lower all-cause mortality in middle-aged and older men. middle-aged and older men. This is likely due to the protective effects of This is likely due to the protective effects of

moderate consumption on coronary heart disease. moderate consumption on coronary heart disease.

Any benefit from moderate drinking among Any benefit from moderate drinking among younger people, who have a lower risk of younger people, who have a lower risk of heart disease, is not apparent. heart disease, is not apparent.

Thus, the “U-shaped” curve does not apply to Thus, the “U-shaped” curve does not apply to the young. the young.


Drinking Frequency and Drinking Frequency and Amount Have Opposite Amount Have Opposite

Effects on BMIEffects on BMI

Breslow RA, et al. Breslow RA, et al. Am J EpidemiolAm J Epidemiol. 2005;161(4):368–. 2005;161(4):368–376.376.


Objectives/MethodsObjectives/Methods To examine the potential relationship To examine the potential relationship

between drinking patterns and body mass between drinking patterns and body mass index (BMI)index (BMI)

Pooled cross-sectional data from 45,896 Pooled cross-sectional data from 45,896 adults who…adults who… had participated in a US health surveyhad participated in a US health survey were current drinkers who never smokedwere current drinkers who never smoked

Analyses were weighted and adjustedAnalyses were weighted and adjusted


ResultsResults BMI significantly increased in both men BMI significantly increased in both men

and women as alcohol intake increased…and women as alcohol intake increased… from 1 drink to >=4 drinks per drinking day on from 1 drink to >=4 drinks per drinking day on

average (average (P P for trend <0.001). for trend <0.001).

However, BMI significantly decreased in However, BMI significantly decreased in both men and women as drinking both men and women as drinking frequency increased…frequency increased… from 1–11 drinking days in the past year to 2–7 from 1–11 drinking days in the past year to 2–7

drinking days per week (drinking days per week (PP for trend <0.001). for trend <0.001).



Those who consumed the smallest Those who consumed the smallest amounts (1 drink per drinking day) the amounts (1 drink per drinking day) the most frequently were the leanest.most frequently were the leanest.

Those who consumed the greatest Those who consumed the greatest amounts (>=3 drinks per drinking day) amounts (>=3 drinks per drinking day) the least frequently were the heaviest.the least frequently were the heaviest.


Conclusions/CommentsConclusions/Comments Limiting analyses to nonsmokers is a strength of this Limiting analyses to nonsmokers is a strength of this

study; teasing apart the separate effects of drinking study; teasing apart the separate effects of drinking and smoking, which often co-occur, is difficult.and smoking, which often co-occur, is difficult.

Among these nonsmokers, there was a strong Among these nonsmokers, there was a strong

pattern of decreasing BMI with increasing drinking pattern of decreasing BMI with increasing drinking frequency, particularly with consumption of small frequency, particularly with consumption of small amounts. amounts.

As is the case with drinking and many other health As is the case with drinking and many other health consequences, benefits accrue with regular consequences, benefits accrue with regular moderate alcohol use while harm often ensues from moderate alcohol use while harm often ensues from heavy episodic drinking. heavy episodic drinking.


Intimate Partner Intimate Partner Violence: The Role of Violence: The Role of

AlcoholAlcohol

Caetano R, et al. Caetano R, et al. Alcohol Clin Exp Res.Alcohol Clin Exp Res. 2005;29(1):98– 2005;29(1):98–106.106.


Objectives/MethodsObjectives/Methods To explore the relationship between intimate To explore the relationship between intimate

partner violence and alcohol consumptionpartner violence and alcohol consumption

Interviews of a random sample of Interviews of a random sample of heterosexual couples at baseline and 5 years heterosexual couples at baseline and 5 years later to assess…later to assess…

alcohol use and related problems, andalcohol use and related problems, and

the incidence (new cases) and recurrence the incidence (new cases) and recurrence (cases at both follow-up and baseline) of (cases at both follow-up and baseline) of male-to-female-partner violence (MFPV) and male-to-female-partner violence (MFPV) and female-to-male partner violence (FMPV)female-to-male partner violence (FMPV)


Objectives/Methods Objectives/Methods (cont.)(cont.)

Sample:Sample: 1136 couples1136 couples Included only those who were Included only those who were

cohabitating and/or married at both cohabitating and/or married at both baseline and follow-up baseline and follow-up

Partner violence:Partner violence: A range of violent behaviors, such as A range of violent behaviors, such as

slapping, kicking, forcing sex, and slapping, kicking, forcing sex, and threatening with a gun or knifethreatening with a gun or knife


ResultsResults The incidence of both FMPV and MFPV at follow-up The incidence of both FMPV and MFPV at follow-up

was only 6%. was only 6%.

Recurrence was more common (FMPV 44%, MFPV Recurrence was more common (FMPV 44%, MFPV 39%).39%).

In unadjusted analyses, incidence and/or recurrence In unadjusted analyses, incidence and/or recurrence

of partner violence was significantly associated of partner violence was significantly associated with…with…

greater mean consumption by male perpetrators, greater mean consumption by male perpetrators, heavy drinking (>=5 drinks on an occasion in the heavy drinking (>=5 drinks on an occasion in the

past year) by male and female perpetrators, past year) by male and female perpetrators, and/or and/or

alcohol problems among male and female alcohol problems among male and female perpetrators.perpetrators.



Among these alcohol indicators…Among these alcohol indicators…

only mean consumption (among both only mean consumption (among both female perpetrators and male victims) female perpetrators and male victims) remained significantly associated with remained significantly associated with new or recurrent cases of FMPVnew or recurrent cases of FMPV

……when analyses were adjusted for when analyses were adjusted for potential confounders.potential confounders.


Conclusions/CommentsConclusions/Comments

FMPV was as common as MFPV in this FMPV was as common as MFPV in this population-based sample, whereas male population-based sample, whereas male perpetration of partner violence is often the perpetration of partner violence is often the norm in clinical samples. norm in clinical samples.

Nevertheless, heavy episodic drinking, alcohol Nevertheless, heavy episodic drinking, alcohol problems, and higher average alcohol problems, and higher average alcohol consumption should be considered risk factors consumption should be considered risk factors for partner violence and addressed in for partner violence and addressed in prevention efforts.prevention efforts.


Moderate Drinking and Moderate Drinking and A Longer LifeA Longer Life

Doll R, et al. Doll R, et al. Int J Epidemiol.Int J Epidemiol. 2005;34(1):199–204. 2005;34(1):199–204.

Wannamethee SG. Wannamethee SG. Int J EpidemiolInt J Epidemiol. . 2005;34(1):205–206.2005;34(1):205–206.


Objectives/MethodsObjectives/Methods

To examine the relationship between To examine the relationship between alcohol consumption and mortality among alcohol consumption and mortality among older adultsolder adults

23-year prospective study of approximately 23-year prospective study of approximately 12,000 male British doctors who…12,000 male British doctors who…

were aged 48–78 years were aged 48–78 years drank a mean of about 1.5–2 standard drinks per drank a mean of about 1.5–2 standard drinks per

dayday

Analyses adjusted for age, smoking, and Analyses adjusted for age, smoking, and follow-up durationfollow-up duration


ResultsResults

Cause of DeathCause of Death

(7364 deaths during follow-(7364 deaths during follow-up)up)

Current Drinkers Current Drinkers Compared with Compared with NondrinkersNondrinkers11

Ishcemic heart diseaseIshcemic heart disease RR 0.7RR 0.7

Other vascular diseasesOther vascular diseases RR 0.8RR 0.8

Respiratory diseaseRespiratory disease RR 0.8RR 0.8

Total mortalityTotal mortality RR 0.8RR 0.8

11Includes ex-drinkersIncludes ex-drinkers


Results (cont.)Results (cont.) The relationship between alcohol and The relationship between alcohol and

mortality was U-shaped for all causes except mortality was U-shaped for all causes except alcohol-augmentable causes (i.e., those alcohol-augmentable causes (i.e., those increased by heaving drinking, such as liver increased by heaving drinking, such as liver disease). disease). Risk of death from alcohol-augmentable causes Risk of death from alcohol-augmentable causes

increased as consumption increased from <5 to increased as consumption increased from <5 to >20 drinks per week.>20 drinks per week.

When recent ex-drinkers were grouped with When recent ex-drinkers were grouped with current drinkers, the association between current drinkers, the association between alcohol and total mortality was attenuated alcohol and total mortality was attenuated (RR 0.9).(RR 0.9).


Conclusions/CommentsConclusions/Comments This study from a leading epidemiologist This study from a leading epidemiologist

confirms the relationship between moderate confirms the relationship between moderate drinking and lowered risk of total and heart drinking and lowered risk of total and heart disease mortality. disease mortality.

However, its finding of decreased risk of However, its finding of decreased risk of respiratory disease mortality is surprising. respiratory disease mortality is surprising.

An editorialist notes that the main results An editorialist notes that the main results may be due partially to a greater risk of may be due partially to a greater risk of death among ex-drinkers (e.g., those who death among ex-drinkers (e.g., those who stopped drinking because of illness) and/or stopped drinking because of illness) and/or insufficient adjustment for confounding insufficient adjustment for confounding factors. factors.


Alcohol and the Risk of Alcohol and the Risk of Renal Cell Cancer in Renal Cell Cancer in

Male SmokersMale Smokers

Mahabir S, et al. Mahabir S, et al. Cancer Epidemiol Biomarkers Prev. Cancer Epidemiol Biomarkers Prev. 2005;14(1):170–175.2005;14(1):170–175.



To examine alcohol’s potential role in the To examine alcohol’s potential role in the carcinogenesis of renal cell cancer (RCC) carcinogenesis of renal cell cancer (RCC)

Assessment of alcohol intake and the Assessment of alcohol intake and the incidence of RCC over 12 years in…incidence of RCC over 12 years in…

27,111 Finnish male smokers aged 50–69 27,111 Finnish male smokers aged 50–69 yearsyears

Analyses adjusted for potential confoundersAnalyses adjusted for potential confounders


ResultsResults

195 cases of RCC developed. 195 cases of RCC developed.

The risk of RCC in those consuming >0.2 The risk of RCC in those consuming >0.2 drinks per day, compared with those drinks per day, compared with those consuming less, decreased as alcohol consuming less, decreased as alcohol consumption increased (consumption increased (P P for for trend=0.005). trend=0.005). RR 0.9 for 0.2 to <1 drink per dayRR 0.9 for 0.2 to <1 drink per day RR 0.9 for 1–2 drinks per dayRR 0.9 for 1–2 drinks per day RR 0.5 for >2 drinks per day RR 0.5 for >2 drinks per day



In analyses adjusted for beverage type, In analyses adjusted for beverage type, risk significantly decreased with increasing risk significantly decreased with increasing beer use and liquor use. beer use and liquor use. There were too few wine drinkers to assess the There were too few wine drinkers to assess the

association between wine intake and RCC. association between wine intake and RCC.

Other risk factors (e.g., age, body mass Other risk factors (e.g., age, body mass index, number of cigarettes smoked per index, number of cigarettes smoked per day) did not modify alcohol’s effect on day) did not modify alcohol’s effect on RCC. RCC.


Conclusions/CommentsConclusions/Comments This large prospective study found that This large prospective study found that

increased alcohol intake is associated with increased alcohol intake is associated with decreased risk of RCC in smokers. decreased risk of RCC in smokers. Drinkers in the highest quartile of consumption (>2 Drinkers in the highest quartile of consumption (>2

drinks per day) had the lowest risk. drinks per day) had the lowest risk.

Because this amount is associated with Because this amount is associated with adverse health consequences that are more adverse health consequences that are more common than RCC, results from this study common than RCC, results from this study should not influence current lower-risk should not influence current lower-risk drinking recommendations.drinking recommendations.

Further, these results—though intriguing—Further, these results—though intriguing—should be prospectively confirmed in other should be prospectively confirmed in other populations before accepted.populations before accepted.


Studies on Studies on InterventionsInterventions


Brief Counseling at RegularBrief Counseling at Regular

Office Visits Reduces DrinkingOffice Visits Reduces Drinking

for 1 Yearfor 1 Year

Reiff-Hekking S, et al. Reiff-Hekking S, et al. J Gen Intern Med.J Gen Intern Med. 2005;20(1):7–13.2005;20(1):7–13.


Objectives/MethodsObjectives/Methods To examine whether 5–10 minutes of counseling To examine whether 5–10 minutes of counseling

during a regularly scheduled primary care visit during a regularly scheduled primary care visit can decrease weekly consumption at 12 monthscan decrease weekly consumption at 12 months

Data from 445 adults who had been drinking Data from 445 adults who had been drinking risky amountsrisky amounts11 and participated in a controlled and participated in a controlled trial of brief intervention delivered during trial of brief intervention delivered during regular primary care visits regular primary care visits

46 participating clinicians from 4 different 46 participating clinicians from 4 different primary care practices; 19 received training in primary care practices; 19 received training in and provided brief interventionand provided brief intervention

Analyses adjusted for potential confounders Analyses adjusted for potential confounders (e.g., sex, baseline level of consumption)(e.g., sex, baseline level of consumption)

11Defined in this study as >12 standard drinks per week or bingeing on 1 or more occasions in the previous Defined in this study as >12 standard drinks per week or bingeing on 1 or more occasions in the previous month for men;month for men;

>9 standard drinks per week or bingeing on 1 or more occasions in the previous month for women>9 standard drinks per week or bingeing on 1 or more occasions in the previous month for women


ResultsResults

Brief Brief Intervention Intervention

GroupGroup

Usual Care Usual Care GroupGroup

Mean decrease in Mean decrease in weekly alcohol weekly alcohol intakeintake

-5.7 drinks-5.7 drinks11 -3.2 drinks-3.2 drinks

Mean decrease in # Mean decrease in # of binge drinkingof binge drinking22 episodesepisodes

-2.0 episodes-2.0 episodes11 -1.6 episodes-1.6 episodes

Likely to adhere to Likely to adhere to safe drinking limits safe drinking limits during follow-upduring follow-up

54%54%33 49%49%

1 1 Significant finding Significant finding

22>=5 drinks on 1 occasion for men; >=4 drinks on 1 occasion for women>=5 drinks on 1 occasion for men; >=4 drinks on 1 occasion for women

33A borderline significant findingA borderline significant finding

Outcomes at 12 Outcomes at 12 monthsmonths



This study demonstrates that a single, 5–10 This study demonstrates that a single, 5–10 minute brief intervention during a regular minute brief intervention during a regular primary care visit can produce modest, yet primary care visit can produce modest, yet lasting, reductions in alcohol intake among lasting, reductions in alcohol intake among people who drink risky amounts. people who drink risky amounts.

Additional research is needed to determine Additional research is needed to determine whether brief counseling over multiple visits whether brief counseling over multiple visits in the context of a long-term patient-clinician in the context of a long-term patient-clinician relationship can further reduce risky drinking.relationship can further reduce risky drinking.


Opiate Antagonists for Opiate Antagonists for Alcohol Dependence: Alcohol Dependence:

Do They Work?Do They Work?

Srisurapanont M, et al. Srisurapanont M, et al. Cochrane Database Syst Rev. Cochrane Database Syst Rev. 2005;Volume 1.2005;Volume 1.


Objectives/MethodsObjectives/Methods To summarize the findings of studies that To summarize the findings of studies that

have examined the role of opiate have examined the role of opiate antagonists in treating alcohol dependence antagonists in treating alcohol dependence

Systematic review of all relevant randomized Systematic review of all relevant randomized controlled trials (RCTs) of opiate antagonistscontrolled trials (RCTs) of opiate antagonists

27 RCTs of naltrexone, 2 RCTs of 27 RCTs of naltrexone, 2 RCTs of nalmefenenalmefene

Most trials also involved psychosocial Most trials also involved psychosocial treatmenttreatment


ResultsResults Most studies of short-term (up to 12 weeks) Most studies of short-term (up to 12 weeks)

treatment and outcomes found that naltrexone, treatment and outcomes found that naltrexone, compared with placebo, decreased the risks of…compared with placebo, decreased the risks of… returning to heavy drinking (RR 0.6, a significant returning to heavy drinking (RR 0.6, a significant

finding) and finding) and returning to any drinking (RR 0.9, a borderline returning to any drinking (RR 0.9, a borderline

significant finding). significant finding).

In the few studies of short-term treatment and In the few studies of short-term treatment and medium-term (>12 weeks to 52 weeks) outcomes, medium-term (>12 weeks to 52 weeks) outcomes, naltrexone…naltrexone… significantly decreased the risk of returning to significantly decreased the risk of returning to

heavy drinking (RR 0.8) but not returning to any heavy drinking (RR 0.8) but not returning to any drinking.drinking.


Results (cont.)Results (cont.) Studies that directly compared naltrexone with Studies that directly compared naltrexone with

other drugs (e.g., acamprosate, disulfiram) other drugs (e.g., acamprosate, disulfiram) were inconclusive but generally favored were inconclusive but generally favored naltrexone.naltrexone.

The few placebo-controlled studies of The few placebo-controlled studies of nalmefene found that short-term treatment nalmefene found that short-term treatment with the drug significantly decreased the short-with the drug significantly decreased the short-term risk of returning to any drinking (RR 0.6).term risk of returning to any drinking (RR 0.6).

Data on returning to heavy drinking, medium-term Data on returning to heavy drinking, medium-term treatment, and medium-term outcomes were not treatment, and medium-term outcomes were not reported. reported.


Conclusions/CommentsConclusions/Comments High-quality RCTs support the short-term use High-quality RCTs support the short-term use

of naltrexone for treating alcohol dependence. of naltrexone for treating alcohol dependence.

Nalmefene shows promise but is only Nalmefene shows promise but is only supported by 2 small studies. supported by 2 small studies.

Longer-term studies, including comparisons Longer-term studies, including comparisons with other biological and behavioral with other biological and behavioral treatments, are needed to determine the treatments, are needed to determine the appropriate medium- and long-term use of appropriate medium- and long-term use of opiate antagonists in treating alcohol opiate antagonists in treating alcohol dependence.dependence.


AUDIT and CRAFFT AUDIT and CRAFFT Outperform CAGE in Outperform CAGE in

Screening Young Screening Young PeoplePeople

Cook RL, et al. Cook RL, et al. J Gen Intern MedJ Gen Intern Med. 2005;20(1):1–6.. 2005;20(1):1–6.



To determine the most accurate and To determine the most accurate and efficient method for screening young efficient method for screening young people for alcohol problems people for alcohol problems

Cross-sectional study that compared the Cross-sectional study that compared the performance of 3 common screening performance of 3 common screening instruments against a gold-standard instruments against a gold-standard diagnostic interview in 358 people who…diagnostic interview in 358 people who… were aged 15–24 years were aged 15–24 years were attending a sexually transmitted disease were attending a sexually transmitted disease

clinic clinic reported alcohol use in the past yearreported alcohol use in the past year


ResultsResults One-third met DSM-IV criteria for a current alcohol disorder.One-third met DSM-IV criteria for a current alcohol disorder.

SensitivitySensitivity SpecificitySpecificity

AUDITAUDIT11 (at cut-off of 9) (at cut-off of 9) 76%76% 79%79%

CRAFFTCRAFFT22 (at cut-off of 2) (at cut-off of 2) 94%94% 33%33%

CAGECAGE3 3 (at cut-off of 1)(at cut-off of 1) 69%69% 63%63%

11Alcohol Use Disorders Identification TestAlcohol Use Disorders Identification Test22A mnemonic standing for A mnemonic standing for CCar,ar, R Relax, elax, AAlone, lone, FForget, orget, FFriends,riends, T Troublerouble33A mnemonic standing for A mnemonic standing for CCut-down, ut-down, AAnnoyed, nnoyed, GGuilty, uilty, EEye-openerye-opener

The AUDIT and the CRAFFT performed better than the CAGE among men, women and blacks. Among whites, the The AUDIT and the CRAFFT performed better than the CAGE among men, women and blacks. Among whites, the AUDIT performed best.AUDIT performed best.


Conclusions/CommentsConclusions/Comments This study affirms that the CAGE alone is a This study affirms that the CAGE alone is a

limited screening tool for younger people.limited screening tool for younger people.

The AUDIT, though an excellent tool, is The AUDIT, though an excellent tool, is likely too long and awkward to score for likely too long and awkward to score for widespread clinical use. widespread clinical use.

Several features make the CRAFFT Several features make the CRAFFT attractive: its brevity, easy scoring, attractive: its brevity, easy scoring, relevance to young people, and high relevance to young people, and high sensitivity even among diverse sensitivity even among diverse populations.populations.


Conclusions/Comments Conclusions/Comments (cont.)(cont.)

Epidemiologic studies suggest that adding Epidemiologic studies suggest that adding questions about typical consumption questions about typical consumption patterns, especially binge use, would patterns, especially binge use, would increase the performance of these increase the performance of these instruments to detect alcohol use disorders instruments to detect alcohol use disorders in young populations. in young populations.

Further research to clarify whether these Further research to clarify whether these instruments can also identify those who are instruments can also identify those who are at risk for or have experienced alcohol-at risk for or have experienced alcohol-related problems is warranted. related problems is warranted.


Studies onStudies onSpecial PopulationsSpecial Populations


The Prevalence of Drinking The Prevalence of Drinking Among Women Who Are or Among Women Who Are or

Might Become PregnantMight Become Pregnant

Tsai J, et al. Morb Mortal Wkly Rep 2004:53(5);1178–1181.



To characterize alcohol use among To characterize alcohol use among women who are pregnant or might women who are pregnant or might become pregnant (i.e., are sexually become pregnant (i.e., are sexually active, able to become pregnant, and do active, able to become pregnant, and do not use birth control)not use birth control)

Data from women aged 18–44 years who Data from women aged 18–44 years who

had participated in a US behavioral risk had participated in a US behavioral risk survey survey


ResultsResultsPregnant Pregnant Women Women

(n=2689)(n=2689)

Women Who Women Who Might Become Might Become

Pregnant Pregnant (n=4404)(n=4404)

Reported alcohol useReported alcohol use11 10%10% 55%55%

Reported binge Reported binge drinking or frequent drinking or frequent drinkingdrinking22

2%2% 12%–13%12%–13%

11At least 1 drink on 1 occasion in the past monthAt least 1 drink on 1 occasion in the past month

22>=5 drinks on 1 occasion in the past month or >=7 drinks per week>=5 drinks on 1 occasion in the past month or >=7 drinks per week

Among women who might become pregnant, binge drinking Among women who might become pregnant, binge drinking was more common in 18–34 year olds, non-Hispanic whites, was more common in 18–34 year olds, non-Hispanic whites, current smokers, unmarried women, and women who had current smokers, unmarried women, and women who had driven while alcohol-impaired.driven while alcohol-impaired.


Conclusions/CommentsConclusions/Comments Drinking among pregnant women may have been Drinking among pregnant women may have been

underreported since self-report of alcohol use is subject underreported since self-report of alcohol use is subject to social desirability bias (i.e., interviewees’ desire to to social desirability bias (i.e., interviewees’ desire to report abstinence from alcohol during pregnancy). report abstinence from alcohol during pregnancy).

Drinking among women who might become pregnant was Drinking among women who might become pregnant was also likely underestimated because this group did not also likely underestimated because this group did not include women using ineffective birth control methods. include women using ineffective birth control methods.

Nonetheless, this study’s findings suggest a need for Nonetheless, this study’s findings suggest a need for continued efforts to…continued efforts to…

screen all women of childbearing age for alcohol use screen all women of childbearing age for alcohol use and and

counsel them—well before they choose to become counsel them—well before they choose to become pregnant—about the adverse effects of alcohol on pregnant—about the adverse effects of alcohol on pregnancy. pregnancy.


Predicting DrinkingPredicting Drinking Patterns Among Patterns Among

Aging AdultsAging Adults

Moore AA, et al. Moore AA, et al. Am J Public HealthAm J Public Health. 2005;95(3):458–465.. 2005;95(3):458–465.


Objectives/MethodsObjectives/Methods To examine how drinking patterns change To examine how drinking patterns change

as adults age as adults age

Data on alcohol use from 14,105 adults Data on alcohol use from 14,105 adults (mean age 47 years at baseline) who…(mean age 47 years at baseline) who… had participated in a US health and nutrition had participated in a US health and nutrition

survey survey were interviewed at baseline and at 4 additional were interviewed at baseline and at 4 additional

times over 20 years times over 20 years

Analyses adjusted for birth year, per capita Analyses adjusted for birth year, per capita consumption, and various demographic consumption, and various demographic factors (e.g., sex, race)factors (e.g., sex, race)


ResultsResults 43% consistently (i.e., at each interview) 43% consistently (i.e., at each interview)

identified themselves as drinkers. identified themselves as drinkers.

Weekly alcohol consumption at age 57 (the Weekly alcohol consumption at age 57 (the reference age) was greater in…reference age) was greater in…

men,men, whites,whites, smokers,smokers, those who were unmarried, andthose who were unmarried, and those with higher educational and those with higher educational and

income levels.income levels.


Results (cont.)Results (cont.) Alcohol consumption decreased as age Alcohol consumption decreased as age

increased. increased.

This decline was lower in more recent birth This decline was lower in more recent birth cohorts and greater in…cohorts and greater in…

men, men, nonwhites, nonwhites, smokers,smokers, those who were unmarried, and those who were unmarried, and those with a lower educational level. those with a lower educational level.



These high-quality data reveal patterns These high-quality data reveal patterns (such as a lower age-associated decline (such as a lower age-associated decline in intake among more recent birth in intake among more recent birth cohorts) that may prove to have public cohorts) that may prove to have public health consequences. health consequences.

Comparable patterns among unhealthy Comparable patterns among unhealthy drinkers would be of particular interest. drinkers would be of particular interest.


Comorbidity Doesn’t ExplainComorbidity Doesn’t Explain the Prevalence ofthe Prevalence of

Alcoholism in Alcoholism in Native AmericansNative Americans

Gilder DA, et al. Alcohol Clin Exp Res. 2004;28(12):1805–1813.


Objectives/MethodsObjectives/Methods To determine the prevalence of co-occurring To determine the prevalence of co-occurring

anxiety or affective disorders and alcohol anxiety or affective disorders and alcohol dependence in Native Americans disordersdependence in Native Americans disorders

Interviews of 483 Southwest California Interviews of 483 Southwest California Indians (known as Mission Indians) who…Indians (known as Mission Indians) who…

were at least one-sixteenth Native Americanwere at least one-sixteenth Native American lived on contiguous reservationslived on contiguous reservations had a mean age of 30 yearshad a mean age of 30 years were demographically similar to Mission Indians were demographically similar to Mission Indians

as a whole, according to US census figuresas a whole, according to US census figures


ResultsResults 66% of the men and 53% of the women 66% of the men and 53% of the women

had current or past alcohol dependence. had current or past alcohol dependence.

The most common anxiety and affective The most common anxiety and affective disorders independent of substance use disorders independent of substance use included…included… major depressive disorder (12%), major depressive disorder (12%),

agoraphobia (7%), and social phobia agoraphobia (7%), and social phobia (7%).(7%).

The prevalences of these disorders did The prevalences of these disorders did not differ significantly between people not differ significantly between people with and without dependence.with and without dependence.


Results (cont.)Results (cont.) The same was true for the prevalences of The same was true for the prevalences of

agoraphobia and social phobia overall (i.e., agoraphobia and social phobia overall (i.e., independent disorders and those independent disorders and those concurrent with substance abuse).concurrent with substance abuse).

However, major depressive disorders However, major depressive disorders overall (independent disorders and those overall (independent disorders and those concurrent with substance abuse) were…concurrent with substance abuse) were…

significantly more common in people significantly more common in people with alcohol dependence than in people with alcohol dependence than in people without dependence (22% versus 11%).without dependence (22% versus 11%).



At least in this Native American population, At least in this Native American population, anxiety and affective disorders independent anxiety and affective disorders independent of substance abuse were not more common of substance abuse were not more common in people with alcohol dependence. in people with alcohol dependence.

And even though major depressive And even though major depressive disorders overall were more prevalent in disorders overall were more prevalent in those with dependence…those with dependence…

they were much less common than was they were much less common than was dependence. dependence.


Conclusions/Comments Conclusions/Comments (cont.)(cont.)

Affective and anxiety disorders remain Affective and anxiety disorders remain important considerations in the important considerations in the treatment of Native Americans with treatment of Native Americans with alcohol dependence. alcohol dependence.

However, they do not appear to explain However, they do not appear to explain the high prevalence of alcoholism in this the high prevalence of alcoholism in this group.group.

Update on Alcohol and Health

Documents

Transcript of Update on Alcohol and Health