Table 2 Description of screening procedure and instruments of included trials

BI versus No BI

Baer et al. 2007 [33]

NR

Investigator developed instrument that included questions about binge drinking and past month use of illicit drugs

One or more binge drinking episodes or used illicit ‘street’ drugs four or more times in the past 30 days.

Not validated

Humeniuk et al. 2008 [35]

Questionnaires were either self-administered (Australia, USA) or by trained personnel (Brazil, India) with other demographic questions at primary care clinics in the various sites.

The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST V3.0 [39])

ASSIST score in the moderate risk range (4 to 26) for cannabis, cocaine, amphetamine stimulants, or opioids. Those with scores in the low and high risk (except tobacco) ranges and those who frequently injected drugs were excluded.

The validity of the ASSIST has been assessed in primary healthcare settings and demonstrated good concurrent, construct, predictive, and discriminative validity [40]

Among participants who scored moderate-risk for more than one substance, the focus of the intervention was the highest scoring or the substance of most concern to the participant.

BI versus Written Information

Bernstein et al. 2009 [34]

NR

Investigator developed instrument referred to by authors as the ‘Youth and Young Adult Health and Safety Needs Survey’. Included unspecified risk items from the USA Centers for Disease Control, Youth Behavioral Risk Factor Surveillance Survey (YBRFS; [41]).

Did not report ‘at risk alcohol use’; smoked cannabis ≥3 times in the last 30 days; reported risky behavior associated with cannabis use; reported ‘3 to 5 days per month’ of cannabis use.

Not validated

Zahradnik et al. 2008 [37]

Participants were asked to complete a self-report screening questionnaire. Those meeting screening criteria were given a diagnostic interview.

Self-report questionnaire. Assessed prescription drug intake by asking ‘have you been taking prescription drugs like hypnotics, sedatives, or analgesics regularly within the last four weeks?’ and screened for disorders using:

Participants were included if they:

SDS - scale’s psychometric properties published in Gossop et al. [42] However, psychometric properties of translated/adapted version unknown.

1) met criteria for prescription drug dependence or abuse (3+ points on the adapted SDS and 5+ points on the QPM and were deemed depended via diagnostic interview) or,

1) German translation of the Severity of Dependence Scale (SDS [42, 43]) adapted to assess prescription drug dependence.

2) consumed prescription drugs with addiction potential for at least 60 days in the last 3 months.

QPM - According to authors the QPM was validated; however, this was impossible to verify as results are published in a German study [44].

2) A questionnaire for prescription drug misuse (QPM; [44])

Diagnostic interview. Section E of SCID-I (Structured Clinical Interview for DSM-IV for Axis I Disorders [45]).

Bernstein et al. 2005 [17]

NR

‘Standard substance abuse screening questions for quantity and frequency in the last month’ that were integrated as part of a health needs history. Exact questions not reported.

Current use of drugs (as determined by the screening questions) and ≥3 on the 10-item DAST

‘Standard substance abuse screening questions’ - not validated

Those screening positive were administered the 10-item Drug Abuse Severity Test (DAST-10 [46])

DAST-10 has satisfactory levels of validity and reliability for use in a clinical or research setting [47, 48]