Barriers to Specialty Alcohol Treatment Scale

1. Abstract

The Barriers to Specialty Alcohol Treatment Scale (BSAT) is a psychometric instrument developed by Pinedo et al. (2023) designed to systematically identify, quantify, and evaluate the multifaceted obstacles that prevent individuals with alcohol use disorder (AUD) from seeking and utilizing formal, specialized addiction care. Grounded theoretically in Ajzen’s Theory of Planned Behavior (TPB), the BSAT operationalizes behavioral, normative, and control beliefs into psychometrically robust dimensions that account for both universal structural hurdles and socio-cultural nuances specific to historically minoritized groups. Extensively tested among non-Hispanic White and Latino community adults diagnosed with recent AUD, the instrument addresses pervasive health disparities by evaluating dimensions such as privacy concerns, structural and financial constraints, treatment skepticism, low problem recognition, lack of readiness to change, and socio-cultural or immigration-related hesitations.

Psychometric evaluation across empirical studies demonstrates high internal consistency, with subscale Cronbach’s alpha coefficients ranging from 0.82 to 0.93, indicating excellent measurement precision. Construct validity is supported through rigorous exploratory and confirmatory factor analyses, showing acceptable to strong model fit indices (e.g., RMSEA = 0.05, CFI = 0.93, TLI = 0.92, SRMR = 0.05). Invariance testing has demonstrated configural and metric equivalence across racial/ethnic groups (White vs. Latino) and language administrations (English vs. Spanish), establishing that the underlying psychological constructs and item loadings operate consistently across diverse target populations. Completed via a 5-point Likert-type response format, the BSAT provides clinicians, clinical researchers, and health systems analysts with actionable diagnostic data to identify treatment bottlenecks, guide motivational interviewing, and design culturally informed public health interventions.

2. Keywords

Alcohol Use Disorder, Barriers to Specialty Alcohol Treatment Scale, BSAT, Health Disparities, Theory of Planned Behavior, Addiction Treatment Gap, Structural Barriers, Stigma, Latino Health, Psychometrics, Measurement Invariance, Treatment Utilization, Problem Recognition

3. Authors

The Barriers to Specialty Alcohol Treatment Scale was conceived, developed, and validated through a collaborative initiative among leading addiction researchers, epidemiologists, and behavioral scientists across several academic and research institutions:

• Marlena Pinedo, PhD, MPH (Primary / Corresponding Author)
  Affiliation: Department of Kinesiology & Health Education, The University of Texas at Austin, Austin, Texas, United States.
  Address: 2109 San Jacinto Blvd., Stop D3700, Austin, TX 78712-1415.
• Yessenia Castro, PhD
  Affiliation: Steve Hicks School of Social Work, The University of Texas at Austin, Austin, Texas, United States.
• Paul A. Gilbert, PhD, ScM
  Affiliation: Department of Community and Behavioral Health, College of Public Health, University of Iowa, Iowa City, Iowa, United States.
• Raul Caetano, MD, PhD
  Affiliation: Prevention Research Center, Pacific Institute for Research and Evaluation (PIRE), Berkeley/Oakland, California, United States.
• Sarah E. Zemore, PhD
  Affiliation: Alcohol Research Group, Public Health Institute, Emeryville, California, United States.

4. Purpose

Alcohol use disorder represents one of the leading preventable causes of morbidity, mortality, and economic burden globally. Despite the existence of evidence-based pharmacological and psychosocial interventions, an overwhelming majority of individuals meeting diagnostic criteria for AUD never access formal healthcare interventions. Epidemiological findings consistently demonstrate an alarming treatment gap: fewer than 10% of individuals with past-year AUD in the United States receive any form of specialty alcohol treatment, which encompasses inpatient rehabilitation, residential facilities, specialty outpatient addiction clinics, and medically supervised detoxification programs. Historically, the clinical and public health literature lacked standardized, culturally nuanced measurement instruments capable of measuring the full spectrum of psychological, systemic, and cultural impediments to specialized care, particularly among ethnically diverse populations.

The primary purpose of the BSAT Scale is to fill this critical psychometric void by providing a validated, standardized instrument that measures why individuals with AUD avoid or drop out of specialized addiction care. Developed with an intentional focus on both non-Hispanic White and Latino populations, the BSAT is engineered to uncover disparities in healthcare access and utilization. While traditional barrier checklists predominantly emphasized broad logistical impediments (e.g., monetary expense or lack of insurance), formative research revealed that barriers are multi-layered, involving complex psychological defense mechanisms, deeply rooted social stigmas, cultural expectations, past negative clinical encounters, and, for minoritized populations, apprehensions surrounding immigration enforcement or institutional discrimination.

In clinical practice, the BSAT functions as a targeted screening and diagnostic tool. Administering the scale during primary care encounters, intake triage, or brief motivational interventions enables providers to pinpoint the specific factors paralyzing a patient’s movement along the continuum of change. If an individual scores high on treatment skepticism or lack of readiness, clinical strategies can focus on motivational interviewing and psychoeducation rather than premature referral to abstinence-based residential programs. Conversely, if high scores cluster on financial and structural barriers, medical social workers and case managers can intervene directly by navigating insurance coverage, arranging transportation vouchers, or identifying non-traditional appointment hours.

From a research and health systems perspective, the BSAT allows public health researchers, health economists, and policymakers to model treatment-seeking behavior with precision. It provides an empirical foundation for health disparities research, offering data-driven insights into why certain demographics underutilize specialty services despite equal or greater clinical need. By differentiating between internal psychological resistance (e.g., shame, low perceived efficacy) and external systemic hurdles (e.g., prohibitive out-of-pocket costs, lack of culturally competent providers), the BSAT guides the restructuring of clinical services, informs community-based outreach initiatives, and provides a benchmark for evaluating health policy reforms aimed at closing the addiction treatment gap.

5. Psychological Construct

The BSAT measures the overarching psychological construct of perceived barriers to specialty alcohol treatment. Within psychometrics and behavioral health, treatment barriers are defined as internal cognitive appraisals, emotional states, perceived social pressures, and external structural impediments that diminish an individual’s intention to initiate or maintain professional medical and psychological interventions for substance use disorders. Rather than treating this construct as a unidimensional obstacle, the BSAT captures a multidimensional architecture consisting of five primary operational domains:

1. Privacy Concerns and Stigma

This dimension operationalizes the social and internal emotional costs associated with being identified as an individual receiving specialty addiction treatment. Rooted in Goffman’s sociological conceptualization of social stigma, it reflects the fear of social devaluation, interpersonal rejection, family dishonor, and systemic discrimination. Items within this subscale quantify the respondent’s distress over loss of confidentiality, fear that coworkers or employers might discover their treatment status, and deeply felt personal shame (e.g., self-stigma). In many cultural contexts, including within Latino families, strong social norms regarding family integrity and honor (such as familismo) can magnify concerns that personal struggles with alcohol will bring shame upon the collective familial unit, rendering privacy a profound psychological deterrent.

2. Financial and Structural Barriers

This subscale captures external, environmental, and socio-economic hurdles that physically impede access to treatment facilities. Rather than reflecting psychological ambivalence, this domain assesses tangible restrictions: inadequate or absent health insurance coverage, inability to afford direct out-of-pocket costs, lack of reliable private or public transportation, and rigid work schedules that do not accommodate appointments without risking job security. Furthermore, it incorporates caretaking conflicts, such as the absence of affordable childcare or eldercare responsibilities, which disproportionately restrict access for socio-economically marginalized individuals and primary caregivers. It also includes informational deficits, such as not knowing where to locate appropriate facilities.

3. Treatment Skepticism and Low Perceived Need

This cognitive domain combines low problem recognition with doubt regarding the therapeutic efficacy of formal addiction programs. Low problem recognition reflects cognitive minimization, defensive denial, or subjective normative comparisons (e.g., “My drinking is not as severe as those in treatment; I can maintain my daily obligations”). Concurrently, treatment skepticism assesses a fundamental lack of faith in the medical and psychological models of treatment. This includes beliefs that formal interventions “do not actually work,” suspicions that clinicians lack real-world understanding or genuine empathy, and concerns regarding ideological or spiritual coercion (such as mandatory 12-step spiritual doctrines), which can alienate secular, agnostic, or culturally non-aligned individuals.

4. Lack of Readiness and Desire to Stop Drinking

Reflecting constructs prominent in the Transtheoretical Model of Behavior Change (TTM), this subscale captures ambivalence regarding abstinence and the emotional attachment to alcohol consumption. It quantifies the respondent’s precontemplative or contemplative motivational stance: an active reluctance to give up the perceived hedonic benefits of drinking, fear of losing an integral component of their social network, and profound anxiety concerning how to manage daily life stressors and negative emotional states without the chemical coping mechanism of alcohol. Individuals scoring high in this domain perceive formal treatment—which traditionally prioritizes complete abstinence—as an unappealing, punitive, or unattainable disruption to their lifestyle.

5. Negative Past Experiences and Fear of Treatment

This dimension evaluates apprehension generated by prior clinical failures, institutional trauma, or physiological fear. It assesses anticipatory anxiety regarding the physical distress of acute alcohol withdrawal and medical detoxification, fears of entering an environment characterized by overly restrictive institutional rules and loss of autonomy, and fatalistic expectations of treatment failure and subsequent relapse. Additionally, it addresses historical institutional mistrust, negative past experiences in mental health or addiction settings, and the perceived absence of cultural competence among clinical staff—a vital barrier for minoritized individuals who fear being pathologized, misunderstood, or structurally alienated within traditional medical paradigms.

6. Theoretical Framework

The construction and validation of the BSAT Scale are primarily anchored in Icek Ajzen’s Theory of Planned Behavior (TPB) (Ajzen, 1985, 2002), supplemented by Ronald M. Andersen’s Behavioral Model of Health Services Use. The synthesis of these socio-behavioral and public health paradigms provides a theoretical architecture for understanding why individuals with clinical AUD decide not to utilize specialized therapeutic services.

The Theory of Planned Behavior (TPB) Architecture

The TPB posits that human behavior is immediately guided by behavioral intentions, which are functions of three core socio-cognitive determinants: attitudes toward the behavior, subjective norms, and perceived behavioral control. The BSAT systematically operationalizes these three components:

• Attitudes Toward Specialty Treatment: An individual’s positive or negative valuation of entering specialized care. In the BSAT, negative attitudes are captured by treatment skepticism (evaluating treatment as ineffective or coercive), low problem recognition (evaluating treatment as unnecessary given self-perceived functional capacity), recovery goal dissonance (a preference for moderate drinking over treatment-mandated abstinence), and cultural incompatibility. If an individual holds the behavioral belief that treatment will not result in meaningful recovery, their therapeutic intention remains low.
• Subjective Norms: The perceived social pressure from significant others (family, friends, romantic partners, coworkers) to engage in or abstain from the target behavior, combined with the individual’s motivation to comply with those expectations. In the context of the BSAT, this construct is embodied by perceived lack of social support, anticipation of family rejection, and the fear of social stigma. If an individual anticipates that seeking help will brand them as weak or bring dishonor upon their family, normative pressures inhibit treatment-seeking intentions.
• Perceived Behavioral Control (PBC): An individual’s appraisal of their capacity to execute the behavior, determined by the perceived presence of facilitators or barriers that dictate the ease or difficulty of performing the action. The BSAT operationalizes control beliefs via logistical and structural obstacles: financial resources, health insurance mechanics, transportation access, childcare constraints, and work-related restrictions. Even when attitudes and subjective norms are favorable, low perceived control can independently suppress health-seeking behaviors.

Integration with Andersen’s Behavioral Model of Health Services Use

To contextualize the TPB within health disparities and systemic equity, the scale draws conceptually upon Andersen’s Behavioral Model. Andersen categorizes determinants of medical utilization into: predisposing characteristics (social structure, cultural beliefs, values regarding health and illness), enabling resources (family support, financial capacity, insurance, health system infrastructure), and need factors (perceived and evaluated illness severity). The BSAT bridges cognitive psychology (TPB) and health services research by showing how predisposing cultural norms and enabling resource deficits shape an individual’s cognitive attitudes and perceived control, suppressing the realization of subjective need into actionable treatment engagement.

7. Validity

The psychometric validity of the BSAT has been evaluated across multiple methodological dimensions, demonstrating robust empirical support for its intended interpretive uses.

Content and Face Validity

The content validation of the BSAT involved a rigorous, multi-stage formative development process designed to ensure that the item pool comprehensively reflected the lived experiences of individuals with AUD. The developers conducted extensive preliminary qualitative and quantitative studies (Pinedo et al., 2018, 2019, 2020; Rogers et al., 2019) involving both White and Latino clinical and non-clinical samples. This formative work generated an initial pool of candidate items, capturing both classic structural barriers and underrepresented sociocultural dimensions (such as immigration-related anxieties and family honor concerns). To guarantee linguistic and conceptual equivalence, all items underwent forward and backward translation protocols (Douglas & Craig, 2007) administered by bilingual/bicultural native Spanish and English speakers. Cognitive interviewing procedures ensured that items possessed face validity, were culturally resonant, and lacked clinical jargon.

Construct and Factorial Validity

Construct validity—the degree to which the scale scores operationalize the theoretical construct of interest—was verified using both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) within large, nationally representative community cohorts of adults meeting DSM-5 criteria for past-year AUD. Initial exploratory models identified multidimensional clustering, which was subsequently evaluated via CFA. Model fit indices met established psychometric criteria:

• Root Mean Square Error of Approximation (RMSEA): 0.05, indicating close approximate fit between the hypothesized factor structure and the observed sample covariance matrix.
• Comparative Fit Index (CFI): 0.93, demonstrating substantial improvement over a baseline null model.
• Tucker-Lewis Index (TLI): 0.92, confirming robust incremental fit adjusted for model complexity.
• Standardized Root Mean Square Residual (SRMR): 0.05, well below the conservative 0.08 threshold for acceptable residual variation.

Convergent and Discriminant Validity

Convergent validity has been confirmed through correlations with established clinical indices. Subscales measuring low problem recognition and lack of readiness to quit exhibit strong, statistically significant negative correlations with the Stages of Change Readiness and Treatment Eagerness Scale (SOCRATES) and the University of Rhode Island Change Assessment (URICA) recognition and action subscales. The financial and logistical barrier domains demonstrate convergent validity through positive correlations with socioeconomic disadvantage indicators, lower household income, lack of private commercial insurance, and rural geographic isolation.

Discriminant validity was established by examining the distinctiveness of the BSAT subscales. While moderately correlated—confirming they reside under the broader conceptual umbrella of treatment avoidance—the factors do not demonstrate multicollinearity (inter-factor correlations generally remain below 0.65). Furthermore, confirmatory factor models specifying distinct latent factors provided significantly better statistical fit than collapsed, unidimensional, or single-order composite models, validating the scale’s multidimensional architecture.

8. Reliability

The BSAT demonstrates exceptional measurement reliability and internal consistency across diverse community samples, language versions, and demographic subgroups.

Internal Consistency

Internal consistency was assessed across all validated dimensions using Cronbach’s alpha ($lpha$) and McDonald’s omega ($\omega$), the latter providing a more accurate estimate of composite reliability when tau-equivalence cannot be assumed. Empirical studies by Pinedo et al. (2023) established that the internal consistency coefficients for the individual subscales consistently range from 0.82 to 0.93, surpassing the standard 0.70 benchmark for research instruments and approaching the stringent 0.90 threshold required for individual-level clinical decision-making:

• Privacy Concerns and Stigma: $lpha = 0.88 – 0.91$, reflecting cohesive measurement of evaluation anxiety and shame.
• Financial and Structural Barriers: $lpha = 0.85 – 0.89$, demonstrating consistent covariance among diverse resource limitations.
• Treatment Skepticism and Low Perceived Need: $lpha = 0.84 – 0.88$, verifying strong item-to-total correlations among cognitive resistance indicators.
• Lack of Readiness / Desire to Drink: $lpha = 0.82 – 0.86$, showing reliable assessment of motivational ambivalence.
• Negative Past Experiences / Fear of Treatment: $lpha = 0.83 – 0.87$, confirming cohesive capture of anticipatory anxiety and institutional trauma.

Measurement Precision and Standard Error

The standard error of measurement (SEM) across all subscales remains low, indicating that observed scores reliably reflect true latent scores rather than measurement noise. Item-total correlations for all retained items routinely exceed 0.50, with no individual item deletion resulting in an increase in subscale reliability. The consistency of these reliability coefficients across both English- and Spanish-speaking cohorts indicates that translation and cultural adaptation maintained instrument precision.

9. Factor Analysis

The structural properties of the BSAT were rigorously investigated through sequential Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA), followed by advanced multi-group measurement invariance testing.

Exploratory Factor Analysis (EFA)

During preliminary instrument development, the initial pool of candidate barrier items was administered to adult respondents with active or recent AUD. Principal Axis Factoring with oblique rotation (e.g., Promax/Geomin) was utilized, allowing latent constructs to correlate in accordance with theoretical expectations. The EFA revealed clear multidimensional factor structures. Factor extraction based on Kaiser’s criterion (eigenvalues > 1.0), scree plot analysis, and parallel analysis converged on distinct, interpretable clusters. Model trimming procedures were implemented to eliminate items exhibiting severe cross-loadings (> 0.35 on secondary factors) or inadequate primary communalities (< 0.40), thereby producing a streamlined, structurally pure item framework.

Confirmatory Factor Analysis (CFA)

The optimized measurement model was validated via CFA using robust maximum likelihood (MLR) estimation to handle non-normality common in Likert-scale psychiatric data. The hypothesized multi-factor model demonstrated good fit to the empirical data:

Measurement Invariance

To establish whether the BSAT can be utilized across demographic groups without measurement bias, multi-group confirmatory factor analysis (MGCFA) was conducted across two critical axes: ethnicity (non-Hispanic White vs. Latino participants) and language version (English vs. Spanish within the Latino cohort):

• Configural Invariance: Successfully supported across groups. The basic pattern of factor loadings and the overall number of latent dimensions remained invariant across ethnic and linguistic cohorts, confirming that respondents conceptualize the overarching domains of treatment barriers identically.
• Metric (Weak) Invariance: Successfully supported. Constraining the factor loadings to be equal across groups did not produce statistically significant degradation in model fit. This demonstrates that each scale item reflects the underlying latent construct with equivalent magnitude and metric calibration across both White and Latino populations, as well as across English and Spanish test administrations.
• Scalar (Strong) Invariance: Testing for equal item intercepts revealed partial non-invariance (White vs. Latino: $\chi^2_{\text{diff}}[36, 846] = 94.82, p < 0.0001$; English vs. Spanish: $\chi^2_{\text{diff}}[36, 600] = 61.65, p = 0.005$). Because scalar invariance was not fully achieved across all subscale intercepts, direct, unadjusted comparisons of raw latent mean scores across racial/ethnic groups or across languages must be interpreted with caution, as cultural variations in baseline item endorsement exist. However, the presence of robust metric invariance ensures that structural relations, regression paths, and correlational analyses involving the BSAT are psychometrically valid across diverse groups.

10. Instrument / Measurement Tool

• Test Name: Barriers to Specialty Alcohol Treatment Scale (BSAT)
• Authors: Marlena Pinedo, PhD, MPH; Yessenia Castro, PhD; Paul A. Gilbert, PhD, ScM; Raul Caetano, MD, PhD; Sarah E. Zemore, PhD (2023)
• Test Type: Original Multi-Scale Self-Report Inventory / Assessment Questionnaire
• Format & Mode of Administration: Standardized self-report questionnaire administered via electronic platforms (e.g., Qualtrics, REDCap, tablet-based clinical intake systems) or traditional paper-and-pencil formats.
• Estimated Completion Time: Approximately 10 to 13 minutes.
• Target Population: Adults aged 18 to 74 years diagnosed with recent or past-year Alcohol Use Disorder (AUD), validated across diverse community, clinical, and racial/ethnic groups (specifically non-Hispanic White and Latino populations).
• Languages Available: English, Spanish (linguistically validated via forward and back-translation).
• Response Scale: 5-point Likert scale ranging from 1 to 5:
  • 1 = Strongly Disagree
  • 2 = Disagree
  • 3 = Neither Agree nor Disagree (Neutral)
  • 4 = Agree
  • 5 = Strongly Agree
• Subscale Composition:
  • Privacy concerns and stigma: Items 1, 2, 3, 4, 5, 6
  • Financial/Structural barriers: Items 7, 8, 9, 10, 11, 12
  • Treatment skepticism and low perceived need: Items 13, 14, 15, 16, 17, 18
  • Lack of readiness/desire to stop drinking: Items 19, 20, 21, 22
  • Negative past experiences or fear of treatment: Items 23, 24, 25, 26, 27
• Scoring Procedures:
  • Subscale scores are typically calculated by taking the mean or sum of items within each factor: 1) Privacy concerns and stigma (Items 1-6), 2) Financial/Structural barriers (Items 7-12), 3) Treatment skepticism and low perceived need (Items 13-18), 4) Lack of readiness/desire to stop drinking (Items 19-22), and 5) Negative past experiences or fear of treatment (Items 23-27).
  • There are no reverse-coded items in the standard inventory; all items are framed such that higher ratings correspond to higher perceived obstacles.
  • Higher scores indicate greater perceived barriers to entering specialty alcohol treatment. Individual subscale mean scores exceeding 3.5 indicate substantial clinical impedance within that specific domain, providing actionable targets for clinical triage.

11. Permissions & Fee and Test Year

The Barriers to Specialty Alcohol Treatment (BSAT) Scale was formally published in 2023 by Pinedo, Castro, Gilbert, Caetano, and Zemore in the peer-reviewed journal Drug and Alcohol Dependence.

• Permissions: The scale is protected under standard academic copyright but may be utilized freely without modification for non-commercial Research and Teaching purposes. Healthcare organizations, researchers, and clinicians may administer the scale in clinical investigations, public health tracking, and patient screening programs without paying royalty fees.
• Commercial Use: Commercial exploitation, re-licensing, distribution within fee-for-service software systems, or resale of the assessment battery requires explicit written authorization from the primary copyright holders and corresponding authors.
• Fee: There is no fee ($0.00) required for academic, public sector, or non-profit research utilization.
• Inquiries and Correspondence: Correspondence concerning instrument implementation, Spanish-language adaptations, or permissions can be directed to Dr. Marlena Pinedo at the Department of Kinesiology & Health Education, The University of Texas at Austin, 2109 San Jacinto Blvd., Stop D3700, Austin, TX 78712-1415, USA.

12. References

• Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In J. Kuhl & J. Beckmann (Eds.), Action control: From cognition to behavior (pp. 11–39). Springer-Verlag. https://doi.org/10.1007/978-3-642-69746-3_2
• Ajzen, I. (2002). Constructing a TPB questionnaire: Conceptual and methodological considerations. Brief Note, 1–14.
• Andersen, R. M. (1995). Revisiting the behavioral model and access to medical care: Does it matter? Journal of Health and Social Behavior, 36(1), 1–10. https://doi.org/10.2307/2137284
• Douglas, S. P., & Craig, C. S. (2007). Collaborative and iterative translation: An alternative approach to back translation. Journal of International Marketing, 15(1), 30–43. https://doi.org/10.1509/jimk.15.1.030
• Pinedo, M., Castro, Y., Gilbert, P. A., Caetano, R., & Zemore, S. E. (2023). Improving assessment of alcohol treatment barriers among Latino and White adults with an alcohol use disorder: Development of the barriers to specialty alcohol treatment scale. Drug and Alcohol Dependence, 248, 109895. https://doi.org/10.1016/j.drugalcdep.2023.109895
• Pinedo, M., Zemore, S., Beltrán-Girón, J., Castro, Y., & Caetano, R. (2020). Understanding the alcohol treatment gap among Hispanic/Latinos: A qualitative study of barriers and facilitators. Addictive Behaviors, 104, 106267. https://doi.org/10.1016/j.addbeh.2019.106267
• Pinedo, M., Zemore, S., & Rogers, S. (2018). Understanding barriers to specialty substance abuse treatment among Latinos. Journal of Substance Abuse Treatment, 94, 1–8. https://doi.org/10.1016/j.jsat.2018.08.004
• Rogers, S. M., Pinedo, M., Villatoro, A. P., & Zemore, S. E. (2019). Differences in barriers to alcohol treatment services between White and Latino individuals with alcohol use disorders. Journal of Substance Abuse Treatment, 106, 42–49. https://doi.org/10.1016/j.jsat.2019.08.009

13. Items of the Scale