The adolescent counterpart to the AUDIT and DAST - a brief, structured screen for alcohol, cannabis and other drug use in young people, and for the risk behaviours that surround it.
The CRAFFT screens for substance-related risk in adolescents aged roughly 12 to 21. It opens with three past-twelve-month frequency questions - days of alcohol use, days of cannabis use, days of any other drug use - and then, depending on the answers, asks up to six further questions. The acronym is a mnemonic for the topics those six questions cover: riding in a Car with an impaired driver, using to Relax, using Alone, Forgetting what happened, Family or Friends saying to cut down, and getting into Trouble while using.
The instrument is not a measure of quantity or dependence. It is a risk screen: it asks about the situations and consequences that most reliably distinguish adolescents whose use warrants a fuller assessment from those whose does not. The car question is treated separately from the score, because riding with an impaired driver is a safety issue in its own right regardless of the young person's own use.
02 - Scoring & cutoffs
How scoring works.
Each of the six CRAFFT questions scores one point if answered yes, giving a total from 0 to 6. A score of two or more is the standard threshold for high risk and indicates that a fuller assessment is warranted. Adolescents who report no substance use at all in the past twelve months answer only the car question; a yes there is addressed as a safety concern and does not by itself generate a CRAFFT score. The threshold is a screening trigger, not a diagnosis: what follows a positive screen is a conversation and a clinical assessment, not a label. Aisel does not host the CRAFFT items or a scoring calculator - see the licensing note below.
03 - Validation evidence
How well it performs.
Knight and colleagues validated the CRAFFT in 538 adolescent clinic patients aged 14 to 18 against a structured diagnostic assessment. A cutoff of two or more identified any problem use with 76% sensitivity and 94% specificity (positive predictive value 0.83, negative predictive value 0.91), and any substance use disorder with 80% sensitivity and 86% specificity. The 2.1 revision replaced the earlier opening questions with the three past-year frequency items, which improved the routing of who is asked the full set. Performance figures below come from the 2002 validation of the original version; clinicians should note that the revision has not been re-validated to the same depth.
76% / 94%
Sensitivity / specificity - any problem use (score >=2)
Knight et al., 2002 (n = 538)
0.83 / 0.91
Positive / negative predictive value
Knight et al., 2002
80% / 86%
Sensitivity / specificity - any substance use disorder
Knight et al., 2002
12-21 years
Validated age range
Boston Children's Hospital, CRAFFT 2.1 manual
04 - How it compares
How it compares to the alternatives.
Instrument
Items
Time
When to reach for it
CRAFFT
6 + 3 frequency
2-3 min
The reference screen for 12-21s; covers alcohol, cannabis and other drugs together. Items not reproduced here for licensing reasons.
Fastest adult alcohol gate; explicitly not for adolescents.
S2BI
3
~1 min
A shorter adolescent frequency screen developed by the same group; no Aisel page yet.
05 - When to use it
Right tool, wrong tool.
Reach for it when
-Routine screening of 12-21 year olds in primary care, paediatrics, school health and adolescent mental health
-Covering alcohol, cannabis and other drugs in one pass rather than running separate adult screens
-Opening a conversation about substance-related risk behaviours - driving, using alone, forgetting - as well as use itself
-Settings where the young person can complete the screen privately; the developers recommend self-administration for candour
Reach for something else when
-Adults - use the AUDIT or DAST-10; the CRAFFT was developed and validated in adolescents
-Diagnosing a substance use disorder; a positive screen indicates assessment is needed, nothing more
-Measuring severity or tracking change over time; the CRAFFT is a binary risk gate with no severity bands
-Any situation where the young person's answers could be seen by a parent - this destroys the validity of the screen and the developers warn against it
06 - Confidence & precision
Reading the score with care.
As a binary screen the CRAFFT has no standard error of measurement and no meaningful-change threshold; precision is best read through predictive values. In the validation sample roughly one in four adolescents with problem use screened negative, so a negative CRAFFT should not override clinical concern. It is a gate, not an outcome measure.
07 - Limitations
What it cannot tell you.
Sensitivity of 76% means around a quarter of adolescents with problem use are missed on a single screen.
Self-report of illegal behaviour by minors is vulnerable to under-disclosure, particularly where confidentiality is not clearly established or a parent is present.
The score does not distinguish substances or grade severity; a score of 2 from cannabis alone and a score of 2 from polysubstance use look identical.
The 2.1 revision's frequency items have not been validated to the same depth as the original six questions.
Validation was conducted in US adolescent clinic populations; performance in other health systems and cultures is less well established.
See how Aisel removes friction where it costs most. A 20-minute walkthrough tailored to your clinic.
We value your privacy
We use cookies to analyse site usage and improve your experience. Analytics and embedded media (e.g. YouTube) only load if you accept. Read our cookie policy.