ASI Addiction Severity Index

The Addiction Severity Index (ASI) is a semi-structured clinical interview developed in 1980 by McLellan and colleagues that provides a multidimensional assessment of substance-related and life problems across seven functional domains: medical status, employment/support, drug use, alcohol use, legal status, family/social relationships, and psychiatric status. Unlike single-dimension screening tools, the ASI recognizes that addiction impacts multiple life areas and requires comprehensive evaluation for effective treatment planning. Each domain receives two types of scores: Interviewer Severity Ratings (ISR) on a 0-9 scale, where 0 indicates no problem and 9 represents extreme problem requiring intensive intervention, and Composite Scores (CS) ranging 0.00-1.00 calculated from specific items within each domain, with higher scores indicating greater severity. The interview takes 45-60 minutes and includes both lifetime and past 30-day assessments, providing both historical context and current problem severity. For example, a patient might show alcohol CS = 0.73 (severe current alcohol problem), drug CS = 0.15 (minimal drug use), employment CS = 0.82 (unemployment and financial instability), and psychiatric CS = 0.65 (significant depression and anxiety requiring treatment). This profile reveals that while alcohol is the primary substance concern, successful treatment must also address employment instability and mental health comorbidities. The ASI demonstrates strong psychometric properties with test-retest reliability of 0.83-0.97 across domains and has been validated in over 200 studies across diverse populations including inpatient, outpatient, criminal justice, and homeless settings.

ASI composite scores provide standardized severity metrics for each of the seven life problem areas, calculated using weighted algorithms based on past 30-day items within each domain. The drug composite score incorporates days of use for different drug classes (opioids, cocaine, cannabis, etc.), route of administration (injection carries higher weight), and self-reported problem severity. The alcohol composite includes drinking days, amount consumed, and problems experienced. The medical composite weighs chronic conditions, hospitalizations, and impairment days. Employment composite reflects unemployment days, income sources, and financial adequacy. Legal composite accounts for charges, incarcerations, and probation/parole status. Family/social composite includes relationship conflicts and social support quality. Psychiatric composite weights depression, anxiety, cognitive problems, and prior mental health treatment. Score interpretation follows general guidelines: 0.00-0.16 indicates no significant problem requiring intervention; 0.17-0.32 represents slight problem that may warrant monitoring or brief intervention; 0.33-0.50 suggests moderate problem requiring treatment consideration; 0.51-0.70 indicates considerable problem necessitating intensive treatment; and 0.71-1.00 represents extreme problem requiring immediate and comprehensive intervention. For instance, a patient presenting with drug CS = 0.48, alcohol CS = 0.12, psychiatric CS = 0.68, and employment CS = 0.55 shows moderate drug use, minimal alcohol involvement, severe psychiatric distress requiring immediate mental health treatment, and considerable employment problems. This profile suggests integrated dual-diagnosis treatment with vocational rehabilitation is needed. Treatment programs use these scores to match patients to appropriate care intensity: outpatient counseling for CS <0.35, intensive outpatient for CS 0.35-0.60, and residential treatment for CS >0.60 in primary problem domains.

The ASI employs targeted questions within each domain to quantify problem severity objectively. The Medical Status section asks about chronic medical conditions, current medications, hospitalizations in the past 6 months and lifetime, days experiencing medical problems in past 30 days, and importance of medical treatment (0-4 scale). The Employment/Support domain queries education years completed, vocational training, employment pattern past 3 years, days worked in past 30 days, money received from employment versus welfare/disability/illegal activity, and satisfaction with employment status. The Alcohol Use section documents age of first intoxication, years of regular use, route of administration if relevant, days of alcohol use in past 30 days, money spent on alcohol, and days experiencing alcohol problems. The Drug Use domain covers 10 substance categories (heroin, methadone, other opioids, barbiturates, sedatives, cocaine/crack, amphetamines, cannabis, hallucinogens, inhalants) with detailed assessment of route (oral, nasal, smoking, non-IV injection, IV injection), years of use, and past 30-day use frequency for each category. The Legal Status section inventories lifetime arrests for specific charges (shoplifting, drug charges, forgery, weapons, assault, etc.), months incarcerated lifetime, current charges pending, probation/parole status, and days of illegal activity for profit in past 30 days. The Family/Social domain assesses current living situation, relationship satisfaction, days of serious conflicts with family in past 30 days, history of abuse (physical/sexual), and quality of social support network. The Psychiatric Status section screens for depression, anxiety, hallucinations, concentration difficulties, violent urges, suicide attempts (lifetime and recent), days experiencing psychological problems in past 30 days, prior psychiatric treatment, and current medications. For example, a 34-year-old patient reporting daily heroin injection for past 8 years, 22 days of use in past month, $1,200 spent on drugs, 18 days of illegal activity, 3 prior incarcerations for possession, living situation instability, 25 days of family conflict, and 28 days of depression would generate composite scores of drug CS = 0.78, legal CS = 0.62, family/social CS = 0.71, and psychiatric CS = 0.73, indicating severe multidimensional problems requiring residential treatment with integrated mental health services.

The ASI occupies a distinct niche in the substance use assessment hierarchy, functioning as a comprehensive clinical evaluation tool rather than a brief screening instrument, with important implications for appropriate use in different clinical contexts. CAGE (4 items), AUDIT (10 items), and DAST-10 (10 items) are brief screening tools designed for rapid case identification in primary care, emergency departments, or population screening, completed in 2-5 minutes, focusing exclusively on substance use symptoms without assessing life context. These tools answer the question 'Does this person have a substance problem?' with sensitivity/specificity optimized for detecting disorders requiring referral. In contrast, the ASI is a 45-60 minute comprehensive assessment interview that addresses 'What are the nature, severity, and scope of this person's substance-related problems across life domains?' and 'What level and type of treatment is needed?' The ASI provides multidimensional severity profiles that guide treatment matching, while screening tools yield binary or categorical results (low/moderate/high risk). The ASI assesses temporal patterns with both lifetime and 30-day windows, whereas screening tools typically evaluate past year or past 3 months. Importantly, the ASI captures problems beyond substance use that impact treatment outcomes: employment instability predicts treatment retention, legal involvement affects program compliance, psychiatric comorbidity necessitates integrated treatment, and family support influences recovery success. For example, two patients might both screen positive on AUDIT (scores 22 and 24, indicating severe alcohol use disorder), but ASI reveals dramatically different profiles: Patient A shows alcohol CS = 0.68, employment CS = 0.15, family/social CS = 0.22, psychiatric CS = 0.18, and legal CS = 0.09, suggesting primary alcohol disorder with intact functioning in other domains, appropriate for outpatient treatment with medical management. Patient B shows alcohol CS = 0.71, employment CS = 0.74, family/social CS = 0.68, psychiatric CS = 0.77, and legal CS = 0.58, indicating severe multidimensional impairment requiring residential treatment with vocational rehabilitation and intensive mental health services. Clinical workflow typically employs screening tools (CAGE, AUDIT, DAST) for initial case finding, followed by ASI administration for positive screens requiring detailed assessment and treatment planning.

The ASI serves as an excellent longitudinal assessment instrument for tracking treatment response, with composite score changes demonstrating sensitivity to clinical improvement across treatment modalities and strong associations with long-term outcomes. Research demonstrates that composite score reductions of 0.10-0.15 points represent clinically meaningful improvement, while reductions exceeding 0.20 points indicate substantial treatment response. Studies of outpatient, intensive outpatient, and residential programs consistently show mean composite score reductions of 0.15-0.35 across domains from admission to 3-month follow-up, with continued improvement through 12 months post-treatment. For example, a patient entering residential treatment with baseline scores of drug CS = 0.73, employment CS = 0.68, family/social CS = 0.64, and psychiatric CS = 0.71 might show 90-day scores of drug CS = 0.22, employment CS = 0.48, family/social CS = 0.38, and psychiatric CS = 0.45, representing robust response with sustained abstinence (0.51-point drug CS reduction), moderate employment improvement (0.20-point reduction), good family relationship recovery (0.26-point reduction), and significant psychiatric symptom reduction (0.26-point reduction). The persistent moderate employment score (0.48) suggests need for enhanced vocational services. Predictive validity studies demonstrate that baseline ASI composite scores strongly predict treatment retention, completion, and long-term abstinence. Higher psychiatric composite scores (>0.50) predict premature dropout (odds ratio 2.3-3.1) unless integrated mental health treatment is provided. Legal composite scores >0.60 predict better retention when criminal justice mandates exist but higher relapse rates due to environmental stressors. Employment composite scores >0.50 predict poorer long-term outcomes unless vocational rehabilitation is integrated. Family/social composite scores serve as protective factors: scores <0.30 indicating strong support predict 40-50% higher 1-year abstinence rates compared to scores >0.60 reflecting conflicted relationships. Serial ASI administration at admission, 30 days, 90 days, 6 months, and 12 months allows clinicians to identify domains requiring intensified intervention. For instance, a patient showing drug CS reduction from 0.68 to 0.18 by 30 days (good substance use response) but psychiatric CS remaining elevated at 0.63 (inadequate mental health improvement) prompts psychiatric consultation and possible medication adjustment. The ASI-Lite, a 20-minute abbreviated version, facilitates more frequent monitoring while preserving composite score validity.

While the ASI represents a gold-standard comprehensive assessment, several methodological limitations and potential biases require consideration to ensure valid interpretation and clinical utility. Self-report bias constitutes the primary limitation: patients may minimize substance use to avoid legal consequences, satisfy probation requirements, maintain child custody, or present favorably to treatment providers; conversely, some patients may exaggerate problems to obtain disability benefits, access medication-assisted treatment, or avoid incarceration. Studies comparing self-report to biological testing (urinalysis, breathalyzer) or collateral informant reports reveal underreporting in 20-35% of cases, particularly for illegal drug use and criminal activity. Interviewers can mitigate this through rapport-building, emphasizing confidentiality limits, using Timeline Follow-Back methods for substance use, and corroborating with biological testing and collateral sources when possible. For example, a patient reporting 3 days of heroin use in past 30 days who tests positive for fentanyl and has injection track marks likely underreports actual use frequency, requiring interviewer clinical judgment to adjust severity ratings appropriately. Interviewer training and reliability represent another concern: Interviewer Severity Ratings involve subjective judgment, with inter-rater reliability varying from 0.65-0.89 depending on domain and interviewer experience. Programs should conduct regular inter-rater reliability assessments and provide ongoing calibration training to maintain scoring consistency. Composite scores demonstrate better reliability (0.83-0.97) due to objective item scoring but still depend on accurate data entry and algorithm application. Cultural and linguistic factors affect validity: the ASI was developed and validated primarily with English-speaking American populations, requiring cultural adaptation and validation for diverse groups. Translation issues, cultural norms around substance use disclosure, and culturally-specific substance use patterns may affect score validity. The ASI also shows limited sensitivity to change in acute treatment settings (first 2-3 weeks) since the 30-day window includes pre-treatment use, requiring interpretation that early reassessments may not yet reflect treatment effects. Time burden (45-60 minutes) limits feasibility in busy clinical settings, leading some programs to use abbreviated versions (ASI-Lite, 20 minutes) that preserve composite scores but omit detailed history. For instance, in an emergency department where a patient with acute alcohol intoxication shows initial confusion and poor cooperation, the interviewer should defer ASI administration until medical stabilization and then corroborate responses with family members and medical records showing 3 prior detoxification admissions. Clinical judgment integrating ASI scores with collateral information, biological testing, and observable behavior produces the most accurate assessment.