Table 1 Study characteristics and screening tool performance

Bastiaens, 2002 [22]

USA

Prospective cohort

Adult patients presenting to a psychiatric emergency and evaluation centre between 8am-8 pm during study period (n = 215)

RAFFT (3)

Surveys administered by psychiatric triage nurses

Multiple drugs

DSM-IV substance use disorder (MINI)

0.84

0.67

RAFFT (3)

Alcohol

DSM-IV alcohol abuse/dependence (MINI)

0.73

0.71

CAGE (2)

0.90

0.84

Beaudoin, 2016 [23]

USA

Prospective cohort

ED patients ≥ 50 years who have used an opioid prescription within the past 30 days (n = 113)

PDUQp (10)

Self-report on electronic tablet computer

Opioids

Opioid misuse (NESARC)

0.44

0.79

PDUQp (10)

DSM-V opioid use disorder (AUDADIS)

0.38

0.81

PDUQp (10)

DSM-V opioid use disorder (moderate to severe) (AUDADIS)

0.56

0.75

Borges, 2001 [24]

Santa Clara, San Jose (USA) and Pachuca, Mexico

Prospective cohort

Hispanic ED patients ≥ 18 years (n = 586 [Santa Clara]; 1511 [Pachuca])

RAPS (1)

Trained interviewers

Alcohol

ICD-10 and DSM-IV alcohol dependence (CIDI)

0.93

0.79

Optimized screening tool generated from other tools (CAGE, BMAST, AUDIT, TWEAK, TRAUMA)

0.98

0.65

RAPS (1)

ICD-10 and DSM-IV alcohol dependence and harmful drinking/abuse (CIDI)

0.74

0.85

Optimized screening tool generated from other tools (CAGE, BMAST, AUDIT, TWEAK, TRAUMA)

0.89

0.71

Brousse, 2014 [25]

France

Prospective cohort

ED patients 18–80 years presenting with acute alcohol intoxication as principal or additional diagnosis AND blood alcohol level ≥ 0.8 g/L (n = 164)

AUDIT (12 [Male])

Trained interviewers

Alcohol

DSM-IV alcohol abuse (MINI)

PPV: 1.00^a

NPV: 0.42

AUDIT (7 [Female])

PPV: 0.95

NPV: 1.00

RAPS4-QF (2 [Male])

PPV: 0.97

NPV: 1.00

CAGE (2 [Female])

PPV: 0.95

NPV: 1.00

AUDIT (14 [Male])

DSM-IV mild alcohol dependence (MINI)

PPV: 0.93

NPV: 0.55

AUDIT (11 [Female])

PPV: 0.86

NPV: 1.00

CAGE (3 [Female])

PPV: 0.88

NPV: 0.80

RAPS4-QF (3 [Male])

PPV: 0.90

NPV: 0.60

AUDIT (18 [Male])

DSM-IV moderate alcohol dependence (MINI)

PPV: 0.94

NPV: 0.50

AUDIT (18 [Female])

PPV: 0.96

NPV: 0.65

Canagasaby, 2005 [26]

USA

Prospective cohort

ED patients ≥ 18 years presenting within 48 h of acute injury (n = 2517 cases); ED patients ≥ 18 years presenting for care of medical illness, matched by age group, sex, and rural versus urban (n = 1151 medical controls)

SASQ (5 drinks in last 3 months [men])

Trained interviewers

Alcohol

DSM-IV current alcohol use and dependence (DIS)

0.85

0.70

SASQ (4 drinks in last 3 months [women])

0.82

0.77

Quantity-Frequency (8 [men])

0.78

0.78

Quantity-Frequency (6 [women])

0.78

0.78

Quantity (3 [men])

0.81

0.66

Quantity (3 [women])

0.71

0.76

Chalmers, 2019 [27]

USA

Prospective cohort

ED patients ≥ 18 years requesting an opioid prescription refill, with chronic pain ≥ 6 months, or experiencing pain at time of presentation (n = 719)

COMM (9)

Research assistants

Opioids

Any of 10 aberrant behaviours noted in EMR

0.49

0.51

ORT (8)

0.93

0.60

SOAPP-R (18)

0.57

0.45

COMM (9)

Any of 10 aberrant behaviours noted in EMR OR statewide prescription drug monitoring database OR local medical examiner database

0.50

0.51

ORT (8)

0.38

0.59

SOAPP-R (18)

0.60

0.50

Cherpitel, 2000a [19]

Pachuca, Mexico and Santa Clara, USA

Prospective cohort

Hispanic ED patients ≥ 18 years (n = 1511 [Pachuca, 93% of 1624 patients sampled]; 586 [Santa Clara, 80% of 733 Hispanic patients sampled])

CAGE (2 [Pachuca])

Interviews by trained research assistants

Alcohol

ICD-10 current harmful drinking or alcohol dependence, and DSM-IV current alcohol abuse or dependence (CIDI)

0.76

0.94

CAGE (2 [Santa Clara])

0.88

0.90

BMAST (6 [Pachuca])

0.42

0.99

BMAST (6 [Santa Clara])

0.71

0.97

AUDIT (8 [Pachuca])

0.92

0.93

AUDIT (8 [Santa Clara])

0.95

0.90

TWEAK (8 [Pachuca])

0.90

0.91

TWEAK (3 [Santa Clara])

0.91

0.88

RAPS (1 [Pachuca])

0.92

0.93

RAPS (1 [Santa Clara])

0.95

0.88

Trauma Scale (2 [Pachuca])

0.31

0.95

Trauma Scale (2 [Santa Clara])

0.67

0.81

Breathalyzer (positive [Pachuca])

0.51

0.93

Breathalyzer (positive [Santa Clara])

0.21

0.99

REPORT drinking within 6 h prior to event (yes [Pachuca])

0.45

0.93

REPORT drinking within 6 h prior to event (yes [Santa Clara])

0.45

0.95

5 MONTHLY (yes [Pachuca])

0.85

0.91

5 MONTHLY (yes [Santa Clara])

0.81

0.88

HOLD (5 [Pachuca])

0.96

0.76

HOLD (5 [Santa Clara])

0.89

0.74

Arrests for driving after drinking (yes [Pachuca])

0.07

1.00

Arrests for driving after drinking (yes [Santa Clara])

0.50

0.88

Cherpitel, 2000b [18]

Santa Clara, USA

Prospective cohort

ED patients ≥ 18 years (n = 1952)

RAPS4 (1)

Not reported

Alcohol

ICD-10 and DSM-IV current (last 12 months) alcohol dependence (CIDI)

0.93

0.87

RAPS4 (1)

ICD-10 and DSM-IV current (last 12 months) harmful drinking/alcohol abuse

0.55

0.79

Cherpitel, 2000c [20]

Pachuca, Mexico and Santa Clara, USA

Prospective cohort

ED patients ≥ 18 years (n = 1624 [Pachuca]; 733 [Santa Clara])

Study reports on same population, screening tools (CAGE, BMAST, AUDIT, TWEAK, RAPS, TRAUMA, REPORT, 5 MONTHLY, HOLD), and reference standard as Cherpitel, 2000a

Cherpitel, 2001a [21]

Pachuca, Mexico and Santa Clara, USA

Prospective cohort

Hispanic ED patients ≥ 18 years (up to 65 years in Pachuca, Mexico) (n = 1511 [Pachuca]; 586 [Santa Clara]; not all included in analyses)

Study reports on same population, screening tools (CAGE, TWEAK, HOLD), and reference standard as Cherpitel, 2000a

Cherpitel, 2001b [17]

Jackson & Hinds County, USA

Prospective cohort

ED patients ≥ 18 years (n = 1498)

CAGE (2)

Trained interviewers

Alcohol

ICD-10 or DSM-IV current alcohol dependence (CIDI)

0.89

0.94

BMAST (6)

0.28

0.99

AUDIT (8)

0.93

0.94

TWEAK (3)

0.89

0.92

RAPS (1)

0.97

0.86

HOLD (5)

0.83

0.82

5 MONTHLY (1)

0.74

0.92

Cherpitel, 2003 [28]

USA

Prospective cohort

ED patients ≥ 18 years (n = 412)

AUDIT (8)

Research assistants

Alcohol

DSM-IV alcohol abuse (CIDI)

0.79

0.95

RAPS4 (1)

0.82

0.93

RAPS4-QF (1 or both QF questions positive)

0.98

0.83

AUDIT (8)

DSM-IV alcohol dependence (CIDI)

0.89

0.92

RAPS4 (1)

0.89

0.90

Cherpitel, 2005 [29]

Poland

Prospective cohort

ED patients ≥ 18 years (n = 1492)

RAPS4 (1 [male])

Trained interviewers

Alcohol

ICD-10 or DSM-IV alcohol dependence (CIDI)

0.92

0.83

RAPS4 (1 [female])

1.00

0.97

CAGE (2 [male])

0.66

0.94

CAGE (2 [female])

0.90

0.99

AUDIT (8 [male])

0.94

0.81

AUDIT (8 [female])

1.00

0.98

RAPS4 (1 [male])

ICD-10 or DSM-IV alcohol abuse/harmful drinking (CIDI)

0.57

0.82

RAPS4 (1 [female])

0.42

0.97

RAPS4-QF (1 [male])

0.92

0.43

RAPS4-QF (1 [female])

0.84

0.86

CAGE (2 [male])

0.28

0.92

CAGE (2 [female])

0.21

0.96

AUDIT (8 [male])

0.59

0.80

AUDIT (8 [female])

0.37

0.97

RAPS4 (1 [male])

ICD-10 or DSM-IV alcohol dependence or abuse/harmful drinking (CIDI)

0.70

0.88

RAPS4 (1 [female])

0.59

0.98

RAPS4-QF (1 [male])

0.94

0.46

RAPS4-QF (1 [female])

0.9

0.87

CAGE 2 (2 [male])

0.42

0.97

CAGE 2 (2 [female])

0.41

0.99

AUDIT (8 [male])

0.71

0.86

AUDIT (8 [female])

0.56

0.97

Cremonte, 2010 [30]

Mar del Plata, Argentina, Pachuca, Mexico and Santa Clara, USA

Prospective cohort

ED patients ≥ 18 years who reported having consumed ≥ 1 drink in last 12 months (current drinkers) (n = 662 [Buenos Aires, Argentina]; 559 [Pachuca, Mexico]; 884 [Santa Clara, USA])

AUDIT (8 [Argentina])

Trained interviewers

Alcohol

DSM-IV alcohol dependence (CIDI)

0.93

0.8

AUDIT (8 [Mexico])

0.92

0.98

AUDIT (8 [USA])

0.94

0.81

CAGE (1 [Argentina])

0.75

0.87

CAGE (1 [Mexico])

0.92

0.64

CAGE (1 [USA])

0.96

0.68

RAPS4 (1 [Argentina])

0.89

0.87

RAPS4 (1 [Mexico])

0.92

0.98

RAPS4 (1 [USA])

0.95

0.75

TWEAK (2 [Argentina])

0.98

0.67

TWEAK (2 [Mexico])

0.9

0.98

TWEAK (2 [USA])

0.91

0.81

Cremonte, 2008 [31]

Mar de Plata, Argentina

Prospective cohort

ED patients ≥ 18 years who reported having consumed ≥ 1 drink in last 12 months (current drinkers) (n = 779 [92% completion]; 643 current drinkers analyzed)

AUDIT (8)

Trained interviewers

Alcohol

DSM-IV alcohol abuse (CIDI)

0.65

0.77

CAGE (1)

0.48

0.82

BMAST (6)

0.04

0.95

RAPS4 (1 or both quantity and frequency questions)

0.86

0.64

TWEAK (2)

0.82

0.65

AUDIT (8)

ICD-10 harmful drinking (CIDI)

0.68

0.77

CAGE (1)

0.51

0.82

BMAST (6)

0.04

0.95

RAPS4 (1 or both quantity and frequency questions)

0.88

0.68

TWEAK (2)

0.66

0.62

BMAST (6)

DSM-IV or ICD-10 alcohol dependence (CIDI)

0.44

0.99

Study also reports on same population, certain screening tools (AUDIT, CAGE, RAPS4, TWEAK), and reference standard of alcohol dependence as Cremonte, 2010 – results not repeated)

Friedmann, 2001 [32]

USA

Prospective cohort

ED patients ≥ 18 years (n = 504 approached; 395 completed screening questionnaire [35 unresponsive or overlooked]; 250 received CIDI gold standard interview)

CAGE (1)

Research assistants

Alcohol

Alcohol abuse OR dependence (prior 12 months) (CIDI)

0.69

0.86

CAGE (2)

0.49

0.90

Augmented CAGE (2)

0.72

0.85

Quantity-Frequency (> 14 drinks/week [Male]; > 7 drinks/week [Female])

0.25

0.94

Maximum on an occasion (> 4 drinks/occasion [Male]; > 3 drinks/occasion [Female])

0.46

0.90

Quantity-Frequency OR Maximum on an occasion (as per thresholds above)

0.48

0.89

NIAAA Strategy (CAGE, Quantity-Frequency, or Maximum on an occasion as per thresholds above)

0.83

0.84

CAGE (1)

Alcohol abuse OR dependence (lifetime) (CIDI)

0.69

0.83

CAGE (2)

0.48

0.87

Augmented CAGE (2)

0.73

0.81

Quantity-Frequency (> 14 drinks/week [Male]; > 7 drinks/week [Female])

0.24

0.92

Maximum on an occasion (> 4 drinks/occasion [Male]; > 3 drinks/occasion [Female])

0.41

0.87

Quantity-Frequency OR Maximum on an occasion (as per thresholds above)

0.43

0.87

NIAAA Strategy (CAGE, Quantity-Frequency, or Maximum on an occasion as per thresholds above)

0.81

0.8

Galicia, 2016 [33]

Spain

Retrospective cohort

ED patients presenting due to recent cocaine use from Jan 1-Dec 31, 2010 (“recent” not defined) (n = 933)

MARRIED-cocaine (210)

Trained reviewer

Cocaine

Outcome: ED revisit (timeframe not specified)

0.46

0.83

Geneste, 2012 [34]

France

Prospective cohort

ED patients 18–80 years presenting with acute alcohol intoxication as principal or additional diagnosis AND blood alcohol level ≥ 0.8 g/L (n = 164)

AUDIT (12)

Trained interviewers

Alcohol

DSM-IV alcohol abuse/harmful drinking or alcohol dependence (MINI)

0.88

0.88

CAGE (3)

0.84

0.76

RAPS4 (1)

0.95

0.65

RAPS4-QF (3)

0.89

0.77

AUDIT (18)

DSM-IV alcohol dependence (MINI)

0.80

0.83

CAGE (3)

0.89

0.61

RAPS4 (2)

0.64

0.84

RAPS4-QF (3)

0.92

0.53

Giguere, 2017 [35]

Canada

Prospective cohort

Adults with psychiatric complaints in an emergency setting (n = 912)

DAST

Self-report on electronic tablet computer

Multiple drugs

Any ICD-10 substance use disorder (medical record review)

Men AUC = 0.794; optimal cut-off = 4 (sensitivity and specificity at this cut-off not provided)

Women AUC = 0.748; optimal cut-off = 2 (sensitivity and specificity at this cut-off not provided)

Kelly, 2004 [36]

USA

Prospective cohort

ED patients ≥ 12 and ≤ 20 years; only patients ≥ 18 years analyzed (n = 246; 191 consented for follow-up)

AUDIT (10)

Patient questionnaire “self-report”

At follow-up: primary investigator or Master’s-level clinical assessors

Alcohol

DSM-IV alcohol abuse or dependence (SCID)

0.82

0.78

CAGE (1)

0.66

0.58

CAGE (2)

0.53

0.78

CRAFFT (3)

0.82

0.67

RAPS-QF (3)

0.82

0.55

Kelly, 2009 [37]

USA

Prospective cohort

ED patients ≥ 12 and ≤ 20 years; only patients ≥ 18 years analyzed (n = 419 participated in ED study; 340 [81%] consented for follow-up)

AUDIT-C (6)

Patient questionnaire

Alcohol

DSM-IV-TR alcohol use disorder (SCID)

0.74

0.77

FAST (3)

0.83

0.73

RAPS4-QF (3)

0.79

0.72

RUFT-Cut (3)

0.80

0.68

CRAFFT (3)

0.69

0.73

DSM-IV-2 (1)

0.88

0.90

Lee, 2019 [38]

South Korea

Prospective cohort

ED patients ≥ 18 and ≤ 90 years presenting post suicide attempts, “suicide attempters” (n = 621)

AUDIT-C (4 [Male])

Self-report questionnaire

Alcohol

DSM-IV-TR alcohol use disorder (chart review)

0.97

0.90

AUDIT-C (8 [Female])

0.77

0.91

Meneses-Gaya, 2010a [39]

Brazil

Prospective cohort

Patients ≥ 18 years from a Psychosocial Care Center for Alcohol and Drugs and from the ED (n = 530; 449 ED patients)

AUDIT (9)

Assistant psychologist with extensive training in use of rating scales

Alcohol

DSM-IV alcohol abuse (SCID)

0.88

0.87

AUDIT-3 (2)

0.84

0.84

AUDIT-4 (7)

0.91

0.84

AUDIT-C (6)

0.9

0.83

AUDIT-QF (5)

0.89

0.81

AUDIT-PC (6)

0.84

0.89

CAGE (1)

0.78

0.81

FAST (2)

0.92

0.81

Five-Shot (2)

0.93

0.79

AUDIT (13)

DSM-IV alcohol dependence (SCID)

0.87

0.94

AUDIT-3 (3)

0.83

0.9

AUDIT-4 (10)

0.86

0.92

AUDIT-C (8)

0.88

0.88

AUDIT-QF (6)

0.86

0.87

AUDIT-PC (8)

0.9

0.91

CAGE (1)

0.85

0.8

FAST (6)

0.82

0.94

Five-Shot (3)

0.86

0.9

AUDIT (9)

DSM-IV alcohol abuse AND dependence (SCID)

0.88

0.92

AUDIT-3 (2)

0.85

0.89

AUDIT-4 (7)

0.91

0.89

AUDIT-C (7)

0.91

0.85

AUDIT-QF (5)

0.89

0.86

AUDIT-PC (7)

0.84

0.94

CAGE (1)

0.78

0.85

FAST (2)

0.92

0.85

Five-Shot (2)

0.93

0.83

Meneses-Gaya, 2010b [40]

Brazil

Prospective cohort

Study reports on same population, screening tools (AUDIT, FAST), and reference standard as Meneses-Gaya, 2010a

Neumann, 2004 [41]

Germany

Prospective cohort

Trauma patients ≥ 18 years (n = 1927)

AUDIT (8 [Male])

Self-administered on laptop computer

Alcohol

ICD-10 alcohol dependence, harmful alcohol use, or high-risk drinking (face-to-face diagnostic interviews)

0.75

0.84

AUDIT (5 [Female])

0.84

0.81

Neumann, 2009 [42]

Germany

Prospective cohort

ED patients ≥ 18 years presenting with acute injury (n = 1233)

AUDIT (8 [Male])

Self-administered on laptop computer

Alcohol

ICD-10 alcohol dependence, harmful alcohol use, or high-risk drinking (face-to-face diagnostic interviews)

0.75

0.84

AUDIT (5 [Female])

0.79

0.85

Sattler, 2019 [43]

USA

Prospective cohort

ED patients ≥ 18 and ≤ 91 years (n = 259)

Computerized dynamic screening instrument (Optimized cut-off)

Self-administered with clinical psychology graduate student researchers nearby to answer questions, clarify meaning of items upon request, or read items aloud and record participants' responses if they were incapable of doing so themselves

Multiple drugs

Drug abuse (CIDI)

0.75

0.98

Computerized dynamic screening instrument (Optimized cut-off)

Alcohol

Alcohol abuse (CIDI)

0.50

1.00

Seale, 2018 [44]

USA

Prospective cohort

ED patients ≥ 18 years (n = 221)

Q1. Validated Single Drug Screening Question – How many times in the past year have you used an illegal drug or used a prescription medication for nonmedical reasons (for example, because of the experience or feeling it caused)? (Yes)

Research assistants

Illegal drugs or, non-medical use of prescription medications

Illicit drug use (MINI)

0.65

0.99

Q2. Single Drug Screening Question (SDSQ) – In the last twelve months, did you smoke pot (marijuana), use another street drug, or use a prescription painkiller, stimulant, or sedative for a non-medical reason? (Yes)

Marijuana, street drugs, or non-medical use of prescription medications

0.68

1.00

Q3. Single Drug Screening Question (SDSQ) – In the last twelve months, did you smoke pot (marijuana), use another street drug, or use a prescription medication ‘recreationally’ (just for the feeling, or using more than prescribed)? (Yes)

Marijuana, street drugs, or recreational use of prescription medications

0.71

0.99

Q4. Single Drug Screening Question (SDSQ) – In the last twelve months, on how many days did you smoke pot (marijuana), use another street drug, or use a prescription medication ‘recreationally’ (just for the feeling, or more than prescribed)? (Yes)

Marijuana, street drugs, or recreational use of prescription medications

0.68

0.99

Singh, 2015 [45]

India

Prospective cohort

Adult emergency medical services patients who consumed any alcohol in the previous year (n = 211; 100 [Cohort 1], 111 [Cohort 2])

Hindi 29-item alcohol screening questionnaire: 5 items predictive (3)

Physicians on duty

Alcohol

ICD-10 alcohol dependence (MINI)

0.78

0.89

van der Westhuizen, 2016 [46]

South Africa

Prospective cohort

Emergency centre patients ≥ 18 years presenting with acute injury due to assault or unintentional causes (falls, burns, etc.) (n = 200)

ASSIST (42)

Trained interviewers

Multiple drugs

ICD-10 and DSM-IV total substance abuse and dependence (MINI)

0.64

0.61

ASSIST (14.5)

ICD-10 and DSM-IV alcohol abuse and dependence (MINI)

0.6

0.6

ASSIST (18)

ICD-10 and DSM-IV illicit substance abuse and dependence (MINI)

0.9

0.87

Vinson, 2007 [47]

USA

Prospective cohort

ED patients with acute injury (n = 2517)

Alcohol questions form

Diagnostic Interview Schedule, 2 constructs predictive (recurrent drinking in situations where physically hazardous, and drinking in larger amounts or over a longer period than intended)

Research staff

Alcohol

DSM-IV alcohol use disorder (AUDA)

0.96

0.85

Williams, 2001 [48]

USA

Prospective cohort

ED patients ≥ 18 years presenting within 48 h of acute injury (n = 2199 during ED shifts; 358 additional “missed” admitted patients; total 2517 individual patients)

Single screening question "When was the last time you had more than 4 (women)/5 (men) drinks containing alcohol?" (3 months)

Trained interviewers

Alcohol

Past-month hazardous drinking (≥ 4 drinks in 1 day or ≥ 14 in 1 week [men]; ≥ 3 in 1 day or ≥ 7 in 1 week [women]) OR past-year DSM-IV alcohol abuse/dependence (DIS)

0.86

0.86

Hazardous drinking (≥ 4 drinks in 1 day or ≥ 14 in 1 week [men]; ≥ 3 in 1 day or ≥ 7 in 1 week [women]) (DIS)

0.94

0.83

DSM-IV alcohol abuse/dependence (DIS)

0.83

0.72

Hazardous drinking (≥ 4 drinks in 1 day or ≥ 14 in 1 week [men]; ≥ 3 in 1 day or ≥ 7 in 1 week [women]) OR DSM-IV alcohol abuse/dependence (DIS)

0.97

0.70

Wilson, 2020 [49]

USA

Prospective cohort

ED patients ≥ 18 years reporting pain persisting in the same body part ≥ 6 months even if unrelated to the current ED visit, or requesting a refill for opioids regardless of pain duration (n = 154 enrolled after initial inclusion/exclusion criteria, and 82 who were on opioids at time of ED presentation)

COMM (9)

Self-completed survey (combining COMM and two other survey instruments)

Opioids

Objective documentation of drug-aberrant behaviors on EMR and/or medical examiner databases

0.45

0.55