CRAFFT Screening Test (CRAFFT 2.1)

People who want a structured way to consider alcohol- or drug-related experiences can use this substance use screening to organize concerns that may otherwise be difficult to name. The CRAFFT Screening Test (CRAFFT 2.1) is intended for respondents reflecting on their own use and its consequences. Its results summarize endorsed situations and estimate substance-use-disorder likelihood, giving self-reflection a focused starting point.

What This Assessment Measures

The CRAFFT questionnaire examines situations involving risk, impaired judgment, coping-motivated use, solitary use, memory loss, social concern, and trouble linked to alcohol or drugs. The questions ask whether particular experiences have occurred, so the result summarizes patterns of exposure and consequence rather than a broad personality profile.

The result profile contains two displays:

• CRAFFT Total Score (0-6) reflects the number of positive responses to items related to substance use. Low concern covers 0-1, while Elevated concern covers 2-6.
• DSM-5 Substance Use Disorder Probability (0-100) assesses the likelihood of meeting DSM-5 criteria for a substance use disorder. Low likelihood covers 0-33, Moderate likelihood covers 34-66, and High likelihood covers 67-100.

Together, these displays distinguish endorsed experiences from estimated disorder probability. The profile can identify concerns that merit a closer, contextual look.

Methodology and Evidence Base

Each item uses a direct Yes-or-No response format. The first displayed result counts positive answers, while the probability result is a separate estimate; because the conversion formula is not specified, item-level weights or an exact calculation should not be inferred. This keeps the scoring description limited to what the questionnaire reports.

As a brief substance use disorder screening approach, the questionnaire is conceptually consistent with CRAFFT (Mitchell et al., 2014), whose cut-points were reevaluated against DSM-5 criteria. Earlier work by Knight et al. (1999, 2002) addressed the brief screen’s development and validity, while Harris et al. (2016) examined computer self-administration alongside clinician administration. It is not presented as an adaptation of those studies.

Interpretation is strongest when answers are considered alongside the respondent’s pattern of use, safety context, and feedback from others. This self-report format can be affected by memory, willingness to disclose, and current circumstances, and the cited studies do not independently standardize this online version. The probability display is therefore best treated as a prompt for fuller assessment, not a diagnosis.