Table 1.

Comparison of Screening and Diagnostic Tools for Opioid Dependence and Abuse.

Instrument (Author, Year)	Type of Measure	Content	Validity	Reliability	Administration Time
MINI (Sheehan et al., 1998)	Clinician- (MINI-CR) and patient-rated versions (MINI-PR) 120 questions (16 sections) Binary (yes/no) response Scoring: Computer algorithm makes 17 Axis I diagnoses including drug abuse and drug dependence	Based on DSM-IV criteria; validated against the SCID-P	MINI-CR against SCID Sensitivity = .45 Specificity = .96 PPV = .50 NPV = .95 MINI-PR against SCID Sensitivity = .39 Specificity = .92 PPV = .35 NPV = .93	MINI·CR inter-rater reliability for drug abuse and dependence: κ ≥ .88 MINI-CR test– retest reliability for drug abuse and dependence: κ ≥ .89	15 to 21 Minutes on average
SCID (Spitzer et al., 1992)	Administered > 500 questions (33 sections) Binary (yes/no) response Scoring: Computer algorithm makes 33 Axis I and Axis II diagnoses including drug abuse and drug dependence	Based on DSM-IV criteria	Gold standard instrument	Cronbach’s α = .98 (for opiates) Inter-rater reliability for drug dependence: κ = .95	30 Minutes on average
CIDI-2 (also known as Diagnostic Interview Schedule) (Robins et al., 1988)	Administered > 500 questions (22 sections) Binary (yes/no) response Scoring: Computer algorithm makes 22 Axis I and Axis II diagnoses including substance use disorders other than alcohol	Based on DSM-III/IV and ICD-10 criteria	Good concordance with DSM-III diagnosis of substance use disorders: κ = .83	Test–retest reliability for drug abuse and dependence: κ = .73	~2 Hours (drug section takes ~20 minutes)
LDS (Raistrick et al., 1994)	Self-completed 10 questions 4-point Likert-type scale (never, sometimes, often, and nearly always) Scoring: Summed total score Continuous scale Maximum score of 30 Cutoff for dependence = 2	Based on a psychological view of dependence	Opiate LDQ sample scores correlated with scores on: SODQ (r = .30) GHQ (r = .33) SFQ (r = .27)	Cronbach’s α = .94 Test–retest reliability: κ = .95	10 Minutes
SODQ (Sutherland et al., 1986)	Self-completed 21 questions (5 sections) 4-point Likert-type Scale (never or almost never, some- times, often, and nearly always)–scale different in 2 questions Scoring: Summed total score Continuous scale Maximum score of 63 No cutoff	Based on the psychobiological nature of dependence	Correlations between SODQ total score and: Need for more of the drug (r = .50) Drug use being out of control (r = .49) Missing a fix (r = .48) Worrying about drug use (r = .43) Wish to stop using (r = .29) Average fixes per day (r = .50)	Section 1 Cronbach’s α = .81 Section 2 Cronbach’s α = .88 Section 3 Cronbach’s α = .86 Section 4 Cronbach’s α = .70	Not reported–Modeled on the SADQ, which takes 5 minutes to complete
SDS (Gossop et al., 1995)	Self-completed 5 questions (represents final section of SODQ) 4-point Likert-type scale Scoring: Summed total score Continuous scale Maximum score of 15 No cutoff	Based on a psychological view of dependence	SDS against CIDI Sensitivity = .86 Specificity = .83 PPV = .86 NPV = .83	Cronbach’s α = .81 to .92	≤ 1 Minute
CAGE-AID (adapted to include drugs) (Brown et al., 1995)	Administered or self-completed 4 questions Binary (yes/no) response Scoring: Summed total score Maximum score of 4 Cutoff for drug dependence = 2	Based on a psychological view of dependence	CAGE-AID against CIDI Sensitivity = .70 Specificity = .85	X	~ 1 Minute
SMAST-AID (Brown et al., 1995)	Administered or self- completed 13 questions Binary (yes/no) response Scoring: Summed total score Maximum score of 13 Cutoff for drug dependence = 3	Based on a psychological, social, and legal view of dependence	SMAST-AID against DIS-R Sensitivity = .40 Specificity = .95	X	~2 Minutes
ASI (McLellan et al., 1980)	Administered or self-completed 200 questions for initial assessment; 133 questions for follow-up Items are binary (yes/no), multiple-choice, and open- ended response Scoring: Computer algorithm determines severity index	Based on quantitative, psychological, social, and legal views of dependence	Correlations between ASI drug severity and: CAGE-AID (r = .64) DAST (r = .73) Clinical Use, Abuse and Dependence Scale (r = .70)	Cronbach’s α = .58 to .79 Test–retest reliability for drug dependence: κ ≥ .92 Inter-rater reliability for drug dependence: κ = .89	45 to 60 Minutes
DAST (Skinner, 1982; Yudko et al., 2007)	Self-completed 28 questions Binary (yes/no) response Scoring: Summed total score Maximum score of 28 Cutoff for drug abuse = 5	Based on a psychological, social, and legal view of dependence	DAST against MAST: Sensitivity = .78 to .96 Specificity = .71 to .94	Cronbach’s α = .92	≤ 5 Minutes
RODS (Wickersham et al., 2015)	Clinician or nonclinician administered 8 questions Binary (yes/no) response Scoring: Summed score “Yes” response to ≥ 3 questions indicates opioid dependence	Based on DSM-IV criteria; validated against MINI	RODS against MINI Sensitivity = .97 Specificity = .76 PPV = .69 NPV = .98 κ = .67	Cronbach’s α = .92 Item-total correlations (range: .66 to .87)	≤ 2 Minutes

Note. MINI = Mini International Neuropsychiatric Interview; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, Fourth edition; SCID = Structured Clinical Interview for DSM Disorders; SCID-P = patient-rated SCID; PPV = positive predictive value; NPV = negative predictive value; CIDI = Composite International Diagnostic Interview; ICD-10 = International Classification of Diseases; LDS = Leeds Dependence Questionnaire; SODQ = Severity of Opiate Dependence Questionnaire; GHQ = General Health Questionnaire; SFQ = Social Functioning Questionnaire; SADQ = Severity of Alcohol Dependence Questionnaire; SDS = Severity of Dependence Scale; SMAST-AID = Short Michigan Alcoholism Screening Test–Adapted to Include Drugs; DIS-R = Diagnostic Interview Schedule–Revised; ASI = Addiction Severity Index; DAST = Drug Abuse Screening Test; MAST = Michigan Alcoholism Screen Test; RODS = Rapid Opioid Dependence Screen; LDS = Leeds Dependence Questionnaire.