Table 11Screening and screening-related assessment procedures used in the brief intervention trials included in this review

StudyScreening/identification proceduresDrinking pattern inclusion criteriaDrinking pattern exclusion criteriaRecruitment yield(s); enrolled/eligible (%);
Curry, 2002 126 Health Psychology 2003 Patients with advance primary care appointments with 23 participating physicians in an HMO clinic were contacted by phone after physicians reviewed lists and applied exclusion criteria. Remaining patients were screened for “at-risk drinking patterns” via a 10–15 minute phone interview about general health behaviors that included the AUDIT, modified Cahalan questions about drinking quantity and frequency, and single questions about frequency of binge drinking and driving after drinking. Alcohol use questions were imbedded with other health behavior items to help mask the focus on drinking.Standard Drink (DR) = 14 g ETOH
1.

AUDIT ≥16 (referred to specialty treatment)

2.

known by physician to be alcoholic

5187 patients identified;
Eligible patients were randomized prior to their routine appointments; those who kept the appointments were enrolled in the study.AUDIT ≤154793 eligible after MD exclusions;
AND3439/4793 completed screening survey;
1 or more of the following drinking patterns: mean ≥ 2 drinks/day in the past month OR ≥ 2 episodes of binge drinking (≥ 5 drinks/occasion) in the past month OR ≥ 1 episode of driving after ≥3 drinks in past month380/3439 screened eligible (11%);
307 enrolled / 380 eligible;
(28 refused to give study information to MD, 19 referred to specialty SA treatment, 26 did not have a primary care visit in study window)
N=307/3439 (8.9%)
Ockene, 1999 127 Archives of Internal Medicine 4 primary care internal medicine practices were randomly assigned to intervention or control conditions. Patients of 46 participating clinicians were screened with the Health Habits Survey by phone, mail, or at a primary care visit to identify “high-risk drinking” or 2+ CAGE responses.Standard drink = 12.8 g ETOH
1.

Current participation in an alcohol intervention program

1760/9772 screened positive for “high-risk drinking” (18%);
Patients who screened positive were invited to a 20–35 minute Lifestyle Interview, either in-person or by phone, which included the Time-Line Follow-Back (TLFB) interview to record alcohol consumption over past 7 days. Study-eligible patients were then invited to enroll. Alcohol use questions were imbedded with other health behavior items to help mask the focus on drinking.“High risk drinkers:”1500/1760 were phoned;
Patients were enrolled if they made a primary care visit within 6 months of the Lifestyle Interview.men > 12 DR/week703/1500 remained eligible after assessment interview;
OR ≥5 DR/occasion ≥once in past month;545/703 made a primary care visit within study window;
women > 9 DR/week530 enrolled / 545 eligible (10 refused and 5 were too ill to participate)
OR ≥4 DR/occasion ≥once in past month;N=530/9772
OR
≥2 positive CAGE questions
Fleming, 1999 120 Journal of Family Practice Participating physicians were recruited from fee-for-service and HMO primary care clinics. Patients 65 or older attending scheduled appointments were asked to complete a modified Health Screening Survey, and those who screened positive for “problem drinking” were invited to a 30-minute, in-person Researcher Lifestyle Interview, which included the Time-Line Follow-Back (TLFB) interview to record alcohol consumption over past 7 days. Study-eligible patients were then randomized. Alcohol use questions were imbedded with other health behavior items to help mask the focus on drinking.Standard drink: 12–14 g ETOH
1.

Alcohol treatment or alcohol withdrawal symptoms in previous year;

2.

MD advice to change alcohol use in previous 3 months;

3.

>50 DR/week

6073/6693 completed screening survey;
(study pop. age ≥65)656/6073 screened positive (11%) & were invited to in-person assessment interview;
Men >11 drinks (>132 g alcohol)/week;396/656 completed assessment interview;
women >8 drinks (>96 g alcohol)/week;180/396 remained eligible
OR158 enrolled / 180 eligible by interview (22 excluded by chart audit or refused)
≥2 positive CAGE questions;N=158
OR
Binge drinking (men: ≥4 DR/occasion; women: ≥3 DR/occasion on ≥2 times in last 3 months).
Fleming, 1997 111 Journal of the American Medical Association 64 participating physicians were recruited from fee-for-service and HMO primary care clinics. Patients attending scheduled appointments were asked to self-administer a Health Screening Survey, and those who screened positive for “problem drinking” were invited to a 30-minute, in-person Research Lifestyle Interview, which included the Time-Line Follow-Back (TLFB) interview to record alcohol consumption over past 7 days. Study-eligible patients were then randomized. Alcohol use questions were imbedded with other health behavior items to help mask the focus on drinking.Standard Drink = 12 g ETOH
1.

Alcohol treatment or alcohol withdrawal symptoms in previous year;

2.

MD advice to change alcohol use in previous 3 months;

3.

>50 DR/week

2925/17,695 screened positive (16.5%);
Men >14 DR/week;2450/2925 willing to participate in assessment interview;
Women >11 DR/week1705/2450 completed assessment interview;
OR852/1705 remained eligible
≥2 positive CAGE questions;774 enrolled / 852 eligible by interview (78 excluded by chart audit or refused)
ORN=774
Binge drinking (men: ≥5 DR/occasion; women: ≥4 DR/occasion on ≥4 times in last 30 days).
Manwell, 2000 121 Alcoholism: Clinical and Experimental Research [This trial examined a subset of 5979 female patients age 18–40 included in Fleming 1997; recruitment methods were identical.]Standard Drink = 12 g ETOH
1.

Alcohol treatment or alcohol withdrawal symptoms in previous year;

2.

MD advice to change alcohol use in previous 3 months;

3.

>56 DR/week

730/5979 screened positive for problem drinking (12%);
64 participating physicians were recruited from fee-for-service and HMO primary care clinics. Patients attending scheduled appointments were asked to self-administer a Health Screening Survey, and those who screened positive for “problem drinking” were invited to a 30-minute, in-person Research Lifestyle Interview, which included the Time-Line Follow-Back (TLFB) interview to record alcohol consumption over past 7 days. Study-eligible patients were then randomized. Alcohol use questions were imbedded with other health behavior items to help mask the focus on drinking.(Women only, age 18–40)454/730 completed assessment interview;
>11 DR/week205 enrolled / 454 interviewed (249 ineligible after interview)
ORN=205
> 4 DR/occasion
OR
≥2 positive on CAGE
Senft, 1997 124 American Journal of Preventive Medicine. Patients attending primary care visits with 47 clinicians at 3 HMO clinics were asked to self-administer a screening questionnaire, which included the AUDIT instrument, in waiting rooms. Those scoring in the “hazardous drinking range” were asked to enroll, and acceptors were randomized prior to seeing the clinician.Standard drink approx. 0.5 oz ETOH
1.

AUDIT >21

620/8017 screened positive for hazardous drinking (7.7%);
No attempt was made to shield the study's focus on drinking and health. Recruitment and intervention occurred at a single, routine primary care visit.AUDIT 8–21 (85% of sample)516 enrolled / 620 eligible (20 refused, 84 were missed)
ORN=516
sum of 2 AUDIT quantity-frequency item scores was ≥5
OR
≥6 DR per occasion at least weekly
Anderson & Scott, 1992 116 British Journal of Addiction 8 group practices participated. Patients self-administered the Health Survey Questionnaire, which was distributed either by mail or at primary care visits.Standard Drink = 10g ETOH (unit)
1.

Weekly consumption ≥1050g (105 DR);

2.

Received advice during the previous year to cut down on alcohol use

524/8483 met weekly consumption criterion (6.2%);
Men meeting the consumption criterion were invited to an in-person assessment interview, which included a one-week drinking diary. Eligible men were randomized after interview.(Men only)419 invited for assessment interview;
Weekly consumption ≥ 350g (35 DR)205/419 completed interview;
194/205 met consumption criterion in past week;
154 enrolled / 194 interviewed
(40 met exclusion criterion)
N=154
Nilssen, 1991 125 Preventive Medicine All adults within an age range in the city of Tromso were invited to attend a comprehensive health screening exam. At the end of the exam, participants were asked to complete and return by mail a questionnaire that included questions about alcohol consumption. Eligibility was determined by GGT levels and reported alcohol consumption.Standard Drink NR
1.

Medical records reviewed to exclude persons with alcoholism or hepato-biliary diseases

21647/27198 screened;
Elevated GGT (50–200 U/l men, 45–200 U/l women);776/21647 with GGT in range;
AND381/776 eligible by questionnaire;
Alcohol drinking at least 2–3 times per week;338 enrolled / 381 surveyed (43 excluded for medical reasons)
ORN=338
Single occasion drinking equivalent to 1 bottle of wine 1–2 times per month.
Wallace, 1988 128 British Medical Journal 47 group practices participated. Patients self-administered the Health Survey Questionnaire, which was distributed either by mail or at primary care visits.Standard drink = 1
1.

Received medical advice about drinking during the previous year

2.

GGT > 150 IU/l

4454 were deemed eligible for interview (denominator not reported);
Patients meeting consumption criterion were invited to an in-person “lifestyle and health survey interview,” which included a past-week “systematic history” of alcohol consumption. Eligible patients were randomized after interview.English unit (NR, but elsewhere=10 g ETOH4203/4454 invited to assessment interview (251 excluded for medical reasons, death, moved away);
Men ≥35 units/week;2571/4203 attended assessment interview;
Women ≥21 units/week929 eligible based on past week consumption
OR909 enrolled/929 eligible (20 met exclusion criteria)
Alcohol consumption not exceeding these limits, but ≥2 positive CAGE questionsN=909
OR
self-assessed drinking problem
Maisto, 2001 122 Journal of Studies on Alcohol Patients at 12 primary care clinics self-administered screening questionnaires (a “lifestyle survey”), which included the AUDIT, in waiting rooms. Eligible patients were invited to attend an assessment interview within 2 weeks of the screening visit.Standard Drink = 0.6 oz ETOH
1.

No drinking in past year

2.

Substance abuse treatment in past year

3.

Acute alcohol withdrawal symptoms

1388/13273 screened positive at primary care visits (10.5%);
The AUDIT, additional alcohol quantity-frequency questions, and a last 30-day history of consumption using the TLFB were administered during this in-person interview. Those who remained eligible after interview were invited to enroll.AUDIT ≥ 8343/1388 consented to assessment interview
OR301 enrolled / 343 assessed (42 met exclusion criteria)
Men ≥ 16 DR/week;N=301
Women ≥ 12 DR/week
WHO Brief Intervention Study Group, 1996 119 American Journal of Public Health Patients were recruited during visits to a variety of health care settings in 8 nations using an initial screening interview to identify those with alcohol consumption patterns that raised risk of chronic disease or acute alcohol-related problems.Standard Drink = 1.5 cL ETOH (14g or 0.5 oz)
1.

Prior treatment for liver damage or alcohol dependence

2.

Warning from MD or other health professional to abstain

3.

Past or recent morning drinking

4.

Recent very high consumption (men 120g/day or women 80g/day)

1559 enrolled / screened & interviewed (number not reported)
Those selected were administered a 20-minute health interview (baseline assessment), and those who remained eligible were asked to enroll. Alcohol use questions were imbedded with other health behavior items to help mask the focus on drinking.Men >50 g/day; or > Women 32 g/day;N=1559
OR
≥6 DR/occasion)
Richmond, 1995 117 Addiction 119 GPs from 40 group practices participated. Patients were enrolled in a weekly sequential process (non-random assignment to study conditions). Patients attending primary care visits were asked to self-administer a 3-minute Health & Fitness Questionnaire in the waiting room. Eligible patients were asked participate in a research study about lifestyles; consenting patients provided blood samples and were given a 15-minute in-person assessment interview, which included a past-week drinking diary. Those remaining eligible were enrolled.Standard Drink = 10 g ETOH
1.

Severe alcohol dependence (Ph score > 10)

2.

Severe alcohol related problems (MAST > 20)

3.

Previous or current alcohol treatment

4.

Any alcohol consumption medically contraindicated

13017/14725 completed screening questionnaires;
Men >35 DR/week;894/13017 met drinking inclusion criteria (6.5%);
Women >21 DR/week713/894 invited to enroll in study (181 leaving area or missed at visit);
467/713 agreed to participate;
378 enrolled / 467 (89 met exclusion criteria or did not attend assessment interview)
N=378
Scott & Anderson, 1990 118 Drug & Alcohol Review 8 group practices participated. Patients self-administered the Health Survey Questionnaire, which was distributed either by mail or at primary care visits.Standard Drink = 10 g ETOH (unit)
1.

Weekly consumption of ≥71 units;

2.

Received advice during previous year to cut down alcohol use

384/11521 met weekly consumption criterion (3.3%);
Women meeting the consumption criterion were invited to an in-person assessment interview, which included a one-week drinking diary. Eligible women were randomized after interview.(Women only)352/384 invited for assessment interview;
21–71 units at least weekly165/352 attended assessment interview;
81/165 met consumption criteria in past week;
72 enrolled/81 assessed (9 met exclusion criteria)
N=72

Notes:

“Assessment” is usually used in these trials to indicate more extensive questioning about alcohol use than is included at the “screening” phase; it differs from diagnostic assessment methods used to triage patients to levels of specialty alcohol treatment, such as those based on DSM-IV criteria for alcohol abuse and dependence.

The Time Line Follow Back (TLFB) is a procedure for ascertaining detailed daily descriptions of alcohol consumption, and is considered by some researchers to be the gold standard measuring alcohol consumption; it may, however, be too time-consuming for use during routine primary care visits.

From: 3, Results

Cover of Behavioral Counseling Interventions in Primary Care to Reduce Risky/Harmful Alcohol Use
Behavioral Counseling Interventions in Primary Care to Reduce Risky/Harmful Alcohol Use [Internet].
Systematic Evidence Reviews, No. 30.
Whitlock EP, Green CA, Polen MR, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.