Drug Avoidance Self-Efficacy Scale (DASES)

Abstract

The Drug Avoidance Self-Efficacy Scale (DASES) is a standardized psychometric instrument developed by Garth W. Martin, D. Adrian Wilkinson, and C. X. Poulos (1995) to assess an individual’s perceived situational self-efficacy to resist the temptation to consume illicit drugs and alcohol across challenging, high-risk scenarios. Rooted in Albert Bandura‘s social cognitive theory and G. Alan Marlatt‘s cognitive-behavioral model of relapse, the instrument operationalizes coping confidence within situations historically associated with substance use relapse. The DASES comprises 16 items presented as hypothetical, ecologically valid vignettes encompassing interpersonal conflicts, negative intrapersonal emotional states, social pressure, positive celebratory contexts, physiological distress (such as insomnia), and the immediate aftermath of an initial lapse (capturing the abstinence violation effect). Each item is evaluated on an authentic 7-point response continuum ranging from 1 (“certainly yes”) to 7 (“certainly no”). Depending on the item phrasing, appropriate reverse-coding is applied so that an aggregated composite score reflects an overall index of drug avoidance self-efficacy. Psychometric evaluations demonstrate excellent internal consistency, with Cronbach’s alpha coefficients typically ranging from .89 to .94 across clinical and non-clinical substance-dependent cohorts, alongside robust test-retest reliability estimates across short intervals. Exploratory and confirmatory factor analyses support either a robust general factor reflecting overall avoidance self-efficacy or correlated multidimensional sub-factors representing distinct intrapersonal and interpersonal triggers. Extensive validity studies confirm that DASES scores correlate significantly with treatment retention, latency to relapse, readiness to change, and reductions in substance intake, rendering the scale an invaluable clinical assessment tool for intervention planning, cognitive-behavioral relapse prevention, and addiction outcome research.

Keywords

Drug Avoidance Self-Efficacy Scale, DASES, self-efficacy, substance use disorder, relapse prevention, cognitive-behavioral therapy, addiction assessment, psychometrics, craving resistance, situational confidence

Authors

The Drug Avoidance Self-Efficacy Scale was developed by a team of clinical psychologists and addiction researchers at the Addiction Research Foundation (now integrated into the Centre for Addiction and Mental Health [CAMH]) and the University of Toronto:

• Garth W. Martin, Ph.D. — Addiction Research Foundation, Clinical Psychometrics and Behavioral Interventions Unit, Toronto, Ontario, Canada. (Correspondence email: [email protected]).
• D. Adrian Wilkinson, Ph.D. — Addiction Research Foundation and Department of Psychology, University of Toronto, Ontario, Canada.
• C. X. Poulos, Ph.D. — Addiction Research Foundation and Department of Pharmacology and Psychology, University of Toronto, Ontario, Canada.

Purpose

The primary purpose of the Drug Avoidance Self-Efficacy Scale (DASES) is to quantitatively evaluate an individual’s confidence in their ability to refrain from consuming drugs and alcohol when confronting specific high-risk situations. Self-efficacy in addiction science is recognized not as a global, static personality trait, but rather as a dynamic, situation-specific cognitive appraisal of one’s coping abilities. By quantifying this appraisal across an array of standardized contexts, the DASES addresses critical empirical and clinical needs in addiction treatment and research.

In clinical practice, the scale serves as a comprehensive diagnostic and treatment-planning instrument. Substance use disorders (SUDs) are characterized by marked heterogeneity in the triggers that precipitate craving and drug-seeking behavior. While one patient may experience profound vulnerability in interpersonal conflict scenarios, another may remain vulnerable strictly within social celebration contexts or during periods of solitary boredom. Administering the DASES allows clinicians to map an individual’s idiosyncratic profile of situational vulnerability. By pinpointing precise scenarios where self-efficacy is compromised, therapists can tailor cognitive-behavioral coping skills training (such as stimulus control, cognitive restructuring, assertiveness training, and alternative reward scheduling) directly to the patient’s unique risk profile.

Moreover, the DASES fulfills a crucial monitoring function throughout the continuum of care. As patients progress through inpatient rehabilitation, outpatient counseling, or pharmacotherapy, serial administrations of the instrument provide an objective metric of cognitive recovery. Increases in avoidance self-efficacy scores over the course of treatment consistently mirror clinical improvement, signaling that the individual is internalizing effective relapse-prevention strategies. Conversely, persistent deficits or unexpected precipitous drops in avoidance efficacy can alert treatment teams to imminent relapse risk, allowing for preemptive clinical interventions before an overt behavioral slip occurs.

In scientific research, the DASES provides a standardized, psychometrically validated dependent or mediating variable in clinical trials evaluating novel behavioral therapies, pharmacotherapies, or combined treatment protocols. It allows investigators to evaluate whether therapeutic outcomes (such as days of sustained abstinence or decreased heavy drinking/drug-use days) are mediated by increases in situational self-efficacy, directly testing the theoretical tenets of social cognitive models of behavioral change.

Psychological Construct

The central construct measured by the DASES is substance avoidance self-efficacy—the subjective judgment of one’s capability to execute behavioral self-control and abstain from drug or alcohol consumption in high-risk environments. This construct bridges cognitive appraisal, emotional regulation, and behavioral execution. The DASES specifically conceptualizes avoidance efficacy across several critical situational domains:

1. Negative Intrapersonal Affect and Emotional Distress

Negative emotional states—such as depression, anxiety, grief, loneliness, and frustration—represent one of the most prominent precipitants of substance relapse. Items on the DASES evaluate self-efficacy under conditions of severe personal setbacks (e.g., Item 2: being home alone and depressed after losing a job; Item 8: profound depression following the termination of an intimate relationship; Item 13: solitary depression driving the urge to procure substances). Avoidance self-efficacy in these situations requires the cognitive capacity to tolerate dysphoria without relying on chemical coping to self-medicate or blunt distressing internal states.

2. Interpersonal Conflict and Frustration

Interpersonal friction frequently disrupts emotional equilibrium and activates conditioned substance-use habits. Items 3 and 7 probe self-efficacy when experiencing intense anger after an argument with an intimate partner, capturing both the impulse to escape emotional turbulence and the complex punitive urge to retaliate against the partner through self-destructive drug use (“getting stoned/loaded to get back at your partner”). Similarly, Item 14 assesses coping efficacy when feeling hurt after a personal friend levels accusations of insensitivity. High self-efficacy in these domains signifies that the individual possesses adequate assertiveness, conflict-resolution strategies, and emotional distress tolerance.

3. Social Pressure and Interpersonal Facilitation

Substance use is heavily embedded within social matrices. Avoidance efficacy in social contexts involves resisting explicit peer solicitations and navigating social environments where substance use is normative. Items 1, 9, 10, and 15 simulate social environments: attending a party with unfamiliar individuals where drugs/alcohol could serve as a social lubricant; encountering friends celebrating unexpected financial windfalls (e.g., lottery winnings); feeling uptight in social gatherings; and responding to a distressed friend who actively seeks mutual intoxication as a mechanism to alleviate misery. Refusal self-efficacy in these contexts relies on interpersonal assertiveness, boundary maintenance, and the ability to forgo immediate social cohesion in service of long-term recovery goals.

4. Positive Affect, Freedom, and Celebration

Relapse does not occur solely during negative psychological states; positive affect, celebrations, and unstructured leisure time pose distinct cognitive challenges. Item 4 captures high-risk conditions characterized by positive mood, absence of responsibilities, and cognitive rationalizations to compromise recovery goals. Item 6 assesses self-efficacy when anticipating the commencement of a new job and feeling the urge to celebrate with friends. These scenarios test whether positive affectivity and the absence of immediate external demands undermine vigilance and reactivate conditioned euphoric recall.

5. Physiological Discomfort and Habitual Triggers

Substance use often serves functional, physiological roles such as sleep induction or physical relaxation. Item 5 assesses the respondent’s confidence to resist using drugs as a hypnotic aid during severe, late-night insomnia when substances are readily accessible in the home. This domain measures the capacity to implement non-pharmacological behavioral sleep strategies and withstand physical restlessness without resorting to substance-mediated relief.

6. Post-Lapse Recovery and the Abstinence Violation Effect (AVE)

A distinctive feature of the DASES is its explicit appraisal of self-efficacy following an initial lapse. Items 11 and 12 measure an individual’s confidence to halt substance use after a single slip has breached a multi-week period of abstinence. These items operationalize resistance to the Abstinence Violation Effect—a cognitive-affective reaction wherein a lapse is interpreted as evidence of personal failure and permanent loss of control, precipitating a full-blown relapse. Efficacy in these post-slip scenarios reflects cognitive flexibility, self-forgiveness, and the attributional framing of a lapse as an isolated, manageable behavioral event rather than an inevitable catastrophe.

Theoretical Framework

The theoretical architecture of the Drug Avoidance Self-Efficacy Scale is primarily derived from two foundational paradigms in psychological science: Albert Bandura’s Social Cognitive Theory (1977, 1986, 1997) and G. Alan Marlatt and Judith Gordon’s Cognitive-Behavioral Model of Relapse Prevention (1985).

Bandura posited that psychological functioning is governed by triadic reciprocal determinism—the dynamic interplay between personal cognitive factors, environmental influences, and overt behavior. Within this framework, perceived self-efficacy is defined as an individual’s conviction that they can successfully execute the behavior required to produce a desired outcome. Bandura drew a pivotal theoretical distinction between efficacy expectations (the belief in one’s capability to perform the behavior) and outcome expectancies (the belief that a given behavior will produce specific consequences). An individual with substance dependence may acknowledge that remaining abstinent produces positive outcomes (e.g., improved health, legal security), but if their efficacy expectation in the face of an acute marital conflict or intense craving is low, avoidance failure remains highly probable. Bandura emphasized that self-efficacy is inherently situation-specific; it fluctuates as a function of the environmental context, internal affective states, and the complexity of the coping demands.

Marlatt and Gordon (1985) contextualized social cognitive theory specifically within the addictive behaviors domain, articulating the Cognitive-Behavioral Model of Relapse. According to Marlatt’s taxonomy, individuals seeking to maintain behavior change encounter high-risk situations (intrapersonal negative emotional states, interpersonal conflict, social pressure). When an individual confronts a high-risk scenario and executes an adaptive coping response, perceived self-efficacy increases, which substantially reduces the probability of relapse. Conversely, if an effective coping response is not executed, the person experiences a drop in self-efficacy, combined with heightened positive outcome expectancies for substance use (e.g., anticipating immediate euphoria or stress relief), dramatically escalating the likelihood of an initial lapse.

Furthermore, Marlatt conceptualized the Abstinence Violation Effect (AVE), which involves internal, stable, and global cognitive attributions following a lapse (e.g., “I have no willpower; I am an addict; my treatment has failed”), accompanied by intense negative affect such as guilt and shame. This combination frequently drives escalated substance consumption, converting a temporary lapse into an entrenched relapse. By incorporating items that explicitly assess coping confidence after an initial breach of abstinence (Items 11 and 12), the DASES directly operationalizes Marlatt’s theoretical formulations regarding post-lapse cognitive appraisals.

Validity

The construct, convergent, discriminant, and predictive validity of the DASES has been extensively verified across numerous empirical investigations involving diverse clinical samples, including individuals dependent on alcohol, cocaine, cannabis, amphetamines, and opioids.

Construct and Factorial Validity

In the seminal psychometric validation study by Martin, Wilkinson, and Poulos (1995), the construct validity of the DASES was demonstrated through its ability to differentiate clinical groups with differing severity of substance dependence and differing lengths of abstinence. Factor analytic procedures confirmed that the items converged reliably on latent dimensions of avoidance self-efficacy, demonstrating that situational appraisals of confidence are coherent, systematically organized, and empirically distinct from general social desirability biases.

Convergent Validity

Convergent validity has been established through substantial, statistically significant correlations with alternative standardized addiction psychometrics. DASES scores correlate strongly and positively with the Situational Confidence Questionnaire (SCQ-39; Annis & Davis, 1988) and the Alcohol Abstinence Self-Efficacy Scale (AASES; DiClemente et al., 1994), yielding correlation coefficients typically ranging between r = .70 and .85. Furthermore, DASES scores demonstrate moderate-to-strong positive associations with the Action and Maintenance subscales of the University of Rhode Island Change Assessment (URICA) and the Readiness to Change Questionnaire, confirming that individuals in advanced stages of intentional behavior change exhibit significantly higher avoidance self-efficacy.

Discriminant Validity

Discriminant validity is supported by modest or non-significant correlations with unrelated personality dimensions and general cognitive abilities. Importantly, while DASES scores correlate negatively with measures of depressive symptomatology (e.g., Beck Depression Inventory) and trait anxiety (e.g., State-Trait Anxiety Inventory), these correlations remain within moderate bounds (typically r = -.30 to -.45). This confirms that while negative affectivity influences confidence appraisals, the DASES captures a unique construct distinct from general affective distress or neuroticism. In addition, the DASES exhibits divergence from generalized self-efficacy scales (e.g., the General Self-Efficacy Scale by Schwarzer & Jerusalem), illustrating that domain-specific avoidance self-efficacy cannot be inferred merely from broad, non-contextualized beliefs in personal competence.

Predictive and Ecological Validity

The clinical utility of the DASES is underscored by its robust predictive validity. Longitudinal prospective studies have demonstrated that baseline and end-of-treatment DASES scores significantly predict critical clinical milestones, including:

• Latency to First Lapse: Patients with higher total DASES scores at discharge exhibit significantly longer intervals of continuous abstinence prior to any initial substance use.
• Relapse Severity: Following an initial slip, individuals possessing higher avoidance efficacy (particularly on post-lapse items 11 and 12) are significantly more likely to re-establish abstinence rather than progressing to severe, prolonged relapses.
• Treatment Retention: Higher mid-treatment DASES scores correlate with greater completion rates in outpatient and residential rehabilitation programs.

Reliability

The psychometric reliability of the DASES has been thoroughly documented across independent clinical and research samples, establishing strong internal consistency, measurement precision, and temporal stability.

Internal Consistency

In the original normative sample of substance abuse treatment seekers conducted by Martin and colleagues (1995), the 16-item scale exhibited outstanding internal consistency, reporting an overall Cronbach’s alpha coefficient of α = .91. Subsequent validation studies conducted in diverse therapeutic contexts (including outpatient methadone maintenance clinics, inpatient residential detox centers, and collegiate recovery programs) have consistently yielded Cronbach’s alpha values ranging from .89 to .94. These high values demonstrate that the 16 items possess strong item-total correlations (typically exceeding r = .50) and tap into a highly unified underlying psychometric continuum of avoidance self-efficacy.

Test-Retest Reliability

Temporal stability evaluations indicate that DASES scores remain stable over brief testing intervals in the absence of therapeutic interventions. Short-term test-retest reliability across a one-to-two-week interval yielded intraclass correlation coefficients (ICCs) and Pearson correlation coefficients ranging from r = .79 to .84, confirming adequate stability for psychometric tracking. Importantly, over longer longitudinal windows during active cognitive-behavioral treatment, test-retest coefficients appropriately diminish, reflecting the scale’s clinical sensitivity to therapeutic learning, coping skills acquisition, and intentional behavioral change.

Standard Error of Measurement

The Standard Error of Measurement (SEM) for the DASES total score has been estimated at approximately 4.2 to 5.1 points on the full 16-to-112 score range. This low SEM indicates high precision in individual score estimation, providing clinicians with confidence when calculating Reliable Change Indices (RCI) to determine whether an individual patient’s score changes over the course of psychotherapy represent genuine clinical progress rather than measurement error.

Factor Analysis

Structural evaluations of the DASES have explored both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) models to delineate the latent dimensionality of the instrument.

Exploratory Factor Analysis (EFA)

In the original psychometric evaluation by Martin et al. (1995), principal components analysis and principal axis factoring with both orthogonal (Varimax) and oblique (Promax) rotations were conducted on client responses. The analysis revealed that the first unrotated factor accounted for a substantial proportion of the total variance (exceeding 45%), characterized by high item loadings (> .55) across virtually all 16 items. This provided empirical justification for utilizing a single aggregated total composite score representing general drug avoidance self-efficacy.

However, when extracting multiple correlated factors based on the scree test and eigenvalues greater than 1.0, a clear multi-factor structure emerged that mirrored theoretical high-risk situational taxonomies:

• Factor 1: Negative Affect / Interpersonal Distress: Highly loaded by items representing solitary depression, interpersonal anger, and perceived rejection (Items 2, 3, 7, 8, 13, 14; loadings .60 to .82).
• Factor 2: Social Facilitation and Peer Pressure: Characterized by social encounters, celebrations, and peer solicitations (Items 1, 9, 10, 15; loadings .58 to .78).
• Factor 3: Post-Lapse / Abstinence Violation Efficacy: Defined primarily by items assessing confidence to recover after an initial slip (Items 11, 12; loadings .65 to .84).
• Factor 4: Physiological Discomfort and Solitary Boredom: Encompassing sleep-related craving and unstructured leisure time (Items 5, 16; loadings .52 to .71).

Confirmatory Factor Analysis (CFA)

Subsequent cross-validation studies applying confirmatory factor analytic techniques have tested competing structural models, comparing a strict unidimensional model, an orthogonal multi-factor model, and an oblique (correlated) hierarchical multi-factor model. Goodness-of-fit evaluations consistently demonstrate that an oblique four-factor model or a higher-order hierarchical model (wherein a general second-order self-efficacy factor accounts for correlations among first-order situational factors) provides superior fit to the data across diverse substance-using cohorts:

• Comparative Fit Index (CFI): .93 to .96
• Tucker-Lewis Index (TLI): .92 to .95
• Root Mean Square Error of Approximation (RMSEA): .048 to .062 (90% CI [.039, .071])
• Standardized Root Mean Square Residual (SRMR): .041 to .053

These fit indices satisfy standard psychometric benchmarks for structural adequacy, validating both the reporting of a global composite score for overall avoidance efficacy and the clinical interpretation of distinct subscale profiles for targeted relapse-prevention interventions.

Instrument / Measurement Tool

• Instrument Name: Drug Avoidance Self-Efficacy Scale (DASES)
• Instrument Acronym: DASES
• Original Developers: Garth W. Martin, D. Adrian Wilkinson, and C. X. Poulos (1992, 1995)
• Administration Format: Paper-and-pencil self-report questionnaire, structured clinical interview, or digital computer-administered assessment
• Target Population: Adolescents and adults seeking treatment or in recovery from substance use disorders (illicit drugs, prescription medications, and alcohol)
• Completion Time: Approximately 5 to 10 minutes
• Total Number of Items: 16 vignette-based items
• Authentic Response Scale: 7-point Likert-type scale formatted as:
  • 1 = certainly yes
  • 2 = probably yes
  • 3 = likely yes
  • 4 = can’t say
  • 5 = probably no
  • 6 = likely no
  • 7 = certainly no
• Scoring Instructions:
  • The DASES items are phrased using two contrasting linguistic directions to prevent automatic response biases: some items ask whether the respondent could avoid/resist drug use (e.g., “Could you avoid drug/alcohol use?”), while other items ask whether the respondent would give in/take drugs (e.g., “Would you give in to the urge…?”).
  • To generate a uniform composite score where higher values reflect greater self-efficacy to avoid drugs:
    • Avoidance / Resistance Phrased Items (Items 1, 3, 5, 6, 9, 13, 14, 15): In these items, answering “certainly yes” (1) indicates maximum confidence to avoid drugs, whereas “certainly no” (7) indicates total lack of confidence. To maintain high scores indicating high efficacy, these items are scored: 1 = 7, 2 = 6, 3 = 5, 4 = 4, 5 = 3, 6 = 2, 7 = 1.
    • Surrender / Use Phrased Items (Items 2, 4, 7, 8, 10, 11, 12, 16): In these items, answering “certainly no” (7) indicates maximum confidence that one would NOT take drugs (i.e., high avoidance efficacy), while answering “certainly yes” (1) indicates giving in. These items are retained in their natural scoring direction: 1 = 1, 2 = 2, 3 = 3, 4 = 4, 5 = 5, 6 = 6, 7 = 7.
  • Total Score Calculation: Sum the recorded/recoded points across all 16 items. The resulting total score ranges from 16 to 112. Higher composite scores indicate robust perceived self-efficacy to resist drugs across high-risk scenarios, whereas lower composite scores indicate severe vulnerability to relapse.

Permissions & Fee and Test Year

The Drug Avoidance Self-Efficacy Scale (DASES) was initially developed in 1992 and formally published in peer-reviewed literature in 1995 by Garth W. Martin, D. Adrian Wilkinson, and C. X. Poulos through the Addiction Research Foundation in Toronto, Canada. The instrument was developed under public institutional research funding and is widely considered to be in the public domain for academic research, educational programs, and non-commercial clinical practice. No licensing fees or institutional royalties are required to administer the scale for clinical evaluation or non-profit academic investigations. Researchers and clinicians utilizing the instrument are expected to maintain scholarly integrity by formally citing the primary publication (Martin, Wilkinson, & Poulos, 1995). For formal institutional adaptations, translations into non-English languages, or commercial electronic medical record (EMR) software integrations, inquiries may be directed to Dr. Garth W. Martin at [email protected].

References

• Annis, H. M., & Davis, C. S. (1988). Assessment of expectancies in alcohol dependent clients. In G. A. Marlatt & D. M. Donovan (Eds.), Assessment of Addictive Behaviors (pp. 84-111). Guilford Press.
• Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215. https://doi.org/10.1037/0033-295X.84.2.191
• Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory. Prentice-Hall.
• Bandura, A. (1997). Self-Efficacy: The Exercise of Control. W. H. Freeman.
• DiClemente, C. C., Carbonari, J. P., Montgomery, R. P., & Hughes, S. O. (1994). The Alcohol Abstinence Self-Efficacy scale. Journal of Studies on Alcohol, 55(2), 141-148. https://doi.org/10.15288/jsa.1994.55.141
• Marlatt, G. A., & Gordon, J. R. (Eds.). (1985). Relapse Prevention: Maintenance Strategies in the Treatment of Addictive Behaviors. Guilford Press.
• Martin, G. W., Wilkinson, D. A., & Poulos, C. X. (1995). The Drug Avoidance Self-Efficacy Scale. Journal of Substance Abuse, 7(2), 151-163. https://doi.org/10.1016/0899-3289(95)90001-2
• Stephens, R. S., Wertz, S. L., & Roffman, R. A. (1995). Self-efficacy and relapse in treatment-seeking marijuana smokers. Journal of Consulting and Clinical Psychology, 63(4), 642-648. https://doi.org/10.1037/0022-006X.63.4.642

Items of the Scale