Risk Behavior Assessment

Abstract

The National Institute on Drug Abuse (NIDA) Risk Behavior Assessment (RBA) is an interviewer-administered, semi-structured epidemiological and behavioral assessment tool designed to systematically document high-risk behaviors associated with the transmission of the Human Immunodeficiency Virus (HIV), Hepatitis B Virus (HBV), Hepatitis C Virus (HCV), and other sexually transmitted infections (STIs). Originally developed by NIDA’s Community Research Branch in conjunction with principal investigators of the NIDA Cooperative Agreement for HIV/AIDS Community-Based Outreach/Intervention Research program, the RBA evaluates high-risk populations not in drug treatment, particularly injection drug users (IDUs) and crack cocaine users. The instrument comprises 12 distinct sections (Sections P and A through K) capturing sociodemographic characteristics, drug use patterns, injection practices, 48-hour drug recall, treatment history, detailed sexual practices categorized by partner gender, commercial sex work/trading, medical diagnoses, legal/incarceration history, socioeconomic status, and interviewer validity impressions. Administration takes approximately 18 to 60 minutes using absolute frequencies, numerical continuous counts, and categorical skip patterns. Psychometric evaluations using 48-hour test-retest paradigms demonstrate robust reliability across major behavioral domains, with test-retest coefficients ranging from moderate to exceptional (.50 to .92 for sexual risk items; .70 to .95 for drug consumption frequencies). Biological criterion-related validity has been confirmed via urinalysis for recent substance use and serological assays for viral and bacterial pathogens including HIV, hepatitis, and syphilis. As an open-access public domain instrument, the RBA and its longitudinal companion, the Risk Behavior Follow-up Assessment (RBFA), serve as foundational measurement standards in public health surveillance, harm reduction evaluation, and behavioral epidemiology.

Keywords

Risk Behavior Assessment, RBA, HIV risk behavior, injection drug use, sexual risk behavior, NIDA Cooperative Agreement, psychometrics, test-retest reliability, criterion validity, substance use assessment

Authors

The Risk Behavior Assessment was developed through a multi-site collaborative initiative spearheaded by the Community Research Branch of the National Institute on Drug Abuse (NIDA) in conjunction with principal investigators from the Cooperative Agreement for HIV/AIDS Community-Based Outreach/Intervention Research Program. Key academic contributors and principal investigators responsible for the psychometric refinement, standardization, and extensive field validation include:

• Dennis G. Fisher, Ph.D. — Center for Behavioral Research and Services, California State University, Long Beach, CA (Correspondence: [email protected]).
• Lucy E. Napper, Ph.D. — California State University, Long Beach, CA and Lehigh University.
• Grace L. Reynolds, Ph.D. — Center for Behavioral Research and Services, California State University, Long Beach, CA.
• Mark E. Johnson, Ph.D. — Center for Behavioral Health Research and Services, University of Alaska, Anchorage, AK.
• Richard H. Needle, Ph.D., M.P.H. — Formerly Chief, Community Research Branch, Division of Clinical and Services Research, National Institute on Drug Abuse, Bethesda, MD.
• Collaborating NIDA Cooperative Agreement Investigators: Including Robert E. Booth, Ph.D. (University of Colorado Health Sciences Center), Dale D. Chitwood, Ph.D. (University of Miami), Harvey Cesari, M.S. (NIDA), and Fen Rhodes, Ph.D. (California State University, Long Beach).

Purpose

The primary purpose of the Risk Behavior Assessment (RBA) is to provide an objective, standardized, and clinically rigorous assessment of behaviors that expose individuals to infectious blood-borne and sexually transmitted pathogens. Conceived during the height of the HIV/AIDS epidemic among out-of-treatment drug-using populations, the instrument addresses the pressing need for a cross-site, standardized measurement battery capable of quantifying behavioral frequencies over specific recall windows (predominantly 30 days and 48 hours).

In clinical and preventive intervention settings, the RBA serves several vital functions:

• Epidemiological Surveillance: It captures the prevalence and incidence of high-risk practices among vulnerable, marginalized, and underserved groups, including out-of-treatment injection drug users, crack cocaine smokers, sex workers, and individuals experiencing persistent housing instability or homelessness.
• Intervention Outcome Evaluation: Developed directly to measure intervention effects within community outreach initiatives, the baseline RBA pairs systematically with the Risk Behavior Follow-up Assessment (RBFA) administered at 3-month and 6-month post-intervention intervals. This longitudinal alignment allows public health researchers to gauge reductions in high-risk injecting practices (such as multiperson syringe reuse, indirect sharing via rinse water or cookers) and unprotected sexual exposures.
• Bio-Behavioral Triangulation: The RBA integrates biological testing paradigms. By combining self-reported behaviors with toxicological urine screening and serological testing for HIV, hepatitis A/B/C, and bacterial STIs (such as syphilis and gonorrhea), researchers can calibrate self-report validity, estimate biological risk ratios, and establish transmission modeling parameters.
• Harm Reduction Planning: Data derived from the RBA allow public health authorities to tailor local needle and syringe exchange programs, determine targeted naloxone distribution strategies, deploy mobile STI testing units, and design specialized sexual health counseling modules for vulnerable subgroups (e.g., street-based sex workers or individuals with co-occurring psychiatric distress).

Psychological Construct

The RBA does not operationalize a singular latent personality trait or psychological syndrome; rather, it assesses multidimensional behavioral constructs operationalized under the broader framework of Health-Risk Behavior and Infectious Disease Risk Topography. Each section targets discrete, measurable operational behaviors that directly govern pathogen exposure pathways:

1. Drug Consumption and Administration Modality (Sections B & D)

This construct captures the absolute frequency, duration, and specific routes of administration (oral, nasal insufflation, inhalation/smoking, subcutaneous injection, and intravenous injection) across distinct psychoactive drug classes. The measured classes encompass alcohol, cannabis, crack cocaine, powder cocaine, heroin, speedballs (cocaine and heroin combined), nonprescription methadone, other illicit or diverted prescription opiates, and amphetamines. Section D measures acute substance use over the past 48 hours, establishing a baseline for immediate recall that can be directly verified against physiological urine toxicology.

2. Parenteral / Injection Transmission Risk (Section C)

Administered exclusively to respondents who have injected drugs within the preceding 30 days, this dimension evaluates behavioral vectors that facilitate the direct and indirect exchange of infected blood. Specific constructs include:

• The frequency of using syringes previously used by someone else, broken down by familiarity of partner (e.g., sex partners, close friends, strangers).
• Mechanisms of syringe cleaning (e.g., use of full-strength bleach, cold water, hot water, alcohol) and adherence to multi-step decontamination protocols.
• Indirect paraphernalia sharing, including shared cookers, filtration cottons, rinse water, and frontloading or backloading practices.

3. Sexual Transmission Risk Topology (Section F)

Recognizing the profound heterogeneity of human sexual behavior, Section F stratifies respondents into six distinct sexual orientation and partner-gender behavioral subsets: men who have sex exclusively with men (MSM), men who have sex with women (MSW), men who have sex with both men and women (MSMW), women who have sex exclusively with women (WSW), women who have sex with men (WSM), and women who have sex with both men and women (WSMW). The measured constructs focus on precise event counts over the last 30 days:

• Specific anatomical exposures: receptive and insertive vaginal intercourse, receptive and insertive anal intercourse, and receptive and insertive oral intercourse.
• Consistent barrier protection usage: absolute counts of male condom, female condom, or dental dam utilization per anatomical act.
• Partner characteristics: absolute number of casual, primary, commercial, and known injection-drug-using sex partners.
• Substance-associated sexual disinhibition (Question 58): quantifying instances where specific psychoactive substances were ingested immediately prior to or during sexual encounters.

4. Commercial Sex Trading and Survival Economies (Section G)

This dimension operationalizes the economic commodification of sexual practices, capturing frequencies of providing or receiving sexual acts in exchange for monetary compensation, illicit drugs, shelter, food, or safety. It measures vulnerability to coercive sexual dynamics, diminished condom negotiation leverage, and rapid partner turnover.

5. Health Status, Legal Incarceration, and Economic Vulnerability (Sections E, H, I, J)

These auxiliary behavioral and contextual dimensions capture broader social determinants of health:

• Addiction Treatment History (Section E): Quantifies engagement with detoxification, methadone maintenance, residential treatment, and outpatient therapeutic modalities.
• Health-Related Conditions (Section H): Captures prior medical diagnoses, subjective symptom history, and testing history for HIV, viral hepatitis, tuberculosis, and bacterial STIs.
• Incarceration Exposure (Section I): Measures lifetime and past-30-day contact with correctional institutions (jails, state/federal prisons), capturing the epidemiological impact of institutional transitions.
• Socioeconomic Deprivation (Section J): Records sources of legal and illegal income, housing instability, and periods of outright street homelessness.

Theoretical Framework

The architecture of the Risk Behavior Assessment is underpinned by integrated paradigms from behavioral epidemiology, harm reduction, and Social Cognitive Theory (Bandura, 1986). Rather than viewing risky behavior as an irrational manifestation of psychopathology, the RBA conceptualizes behavior as the direct interface between individual agency, drug-related cognitive alterations, physiological dependence, and structural socioeconomic constraints.

Several theoretical tenets direct the instrument’s operational structure:

• The Triadic Reciprocal Causation Model: Rooted in Bandura’s social cognitive model, behavioral patterns are seen as emerging from continuous reciprocal interactions between environmental stressors (poverty, unstable housing, incarceration), internal biological/psychological states (craving, physical withdrawal, affective dysregulation), and behavioral competencies (skills in condom negotiation or sterile injection preparation). The RBA gathers data across all three facets to enable structural equation and path-analytic modeling of risk pathways.
• The Harm Reduction Continuum: The RBA rejects binary moralistic classifications of drug use (i.e., absolute abstinence versus active addiction). Rooted in harm reduction principles, it assumes that risk exists on a graded continuum. Even within continuous drug administration, significant shifts along the continuum—such as transitioning from injecting to snorting, switching to single-use sterile syringes, or consistently using bleach solutions—mark substantial clinical progress and lowered transmission rates.
• Cognitive Recall and Event-Specific Memory Paradigms: Drawing upon cognitive psychology and survey methodology (Tourangeau, Rips, & Rasinski, 2000), the RBA relies on bounded episodic memory retrieval. Standard global self-evaluations (e.g., “How often do you practice safe sex?”) introduce massive measurement bias, social desirability distortion, and cognitive heuristics. Instead, the RBA forces autobiographical retrieval anchored to defined calendar windows (the past 30 days and past 48 hours) and decomposes complex behavioral repertoires into anatomical and substance-specific act frequencies.

Validity

The construct, criterion-related, convergent, and discriminant validities of the Risk Behavior Assessment have been rigorously analyzed through dozens of multi-site field studies across diverse geographical cohorts within the United States.

1. Criterion-Related and Biological Validity

A persistent methodological challenge in behavioral substance research involves the veracity of self-reported drug administration. To validate self-report, researchers within the NIDA Cooperative Agreement instituted systematic urinalysis protocols that compared self-reported recent drug consumption (Section D: past 48 hours) against biological drug metabolite assays (Dowling-Guyer et al., 1994; Weatherby et al., 1994). Findings revealed high rates of concordance:

• Cocaine / Crack: Concordance rates between self-reported use within 48 hours and urine enzyme multiplied immunoassay technique (EMIT) results consistently exceeded 80% to 90%, demonstrating high sensitivity and specificity among street-recruited drug users.
• Heroin / Opiates: Validation studies confirmed congruent concordance (85–92%), proving that out-of-treatment injection users accurately disclose recent opiate use when guaranteed strict research confidentiality and protected under federal Certificates of Confidentiality.
• Amphetamines: Napper, Fisher, Johnson, and Wood (2008) established the criterion validity of self-reported methamphetamine and amphetamine use against toxicological markers, verifying robust concordance even in street-entrenched populations.

2. Diagnostic and Serological Validity

The validity of Section H (Health-Related Conditions) has been corroborated using clinical diagnostic records and serological markers:

• Viral Hepatitis (HBV & HCV): Fisher, Kuhrt-Hunstiger, Orr, and Davis (1999) and Schlicting et al. (2003) evaluated the accuracy of self-reported hepatitis infection against serological assays (e.g., anti-HCV, HBsAg, anti-HBc). Self-reports demonstrated acceptable specificity (exceeding 85%), although sensitivity was constrained by historical underdiagnosis and asymptomatic carriage among street-recruited populations, emphasizing the importance of coupling behavioral data with routine clinical serology.
• Syphilis & Gonorrhea: Fisher, Reynolds, Creekmur, Johnson, and Deaugustine (2007) evaluated the criterion-related validity of self-reported syphilis history compared with rapid plasma reagin (RPR) and Treponema pallidum hemagglutination (TPHA) test results, reporting high specificity (93%). Paschane, Fisher, Cagle, and Fenaughty (1998) likewise affirmed the epidemiological validity of gonorrhea self-report within cross-regional comparisons.
• HIV Serostatus Disclosure: Fisher, Reynolds, Jaffe, and Johnson (2007) analyzed the validity, sensitivity, and specificity of self-reported HIV status, confirming that individuals living with HIV highly reliably reported their serostatus (sensitivity > 90%, specificity > 98%) when safe, supportive research conditions were maintained.

3. Construct and Predictive Validity

The RBA demonstrates robust construct and predictive validity by consistently mirroring known epidemiological gradients. Longitudinal analyses show that higher baseline injecting risk scores (e.g., frequent needle sharing, unsterile water source utilization) prospectively predict incident seroconversion for HCV and HBV. Similarly, frequency counts of unprotected insertive/receptive anal sex recorded in Section F reliably predict acquisition of bacterial STIs and HIV across multi-year cohort observations (Needle et al., 1995).

Reliability

The reliability of the RBA has been extensively evaluated across multiple centers using rigorous 48-hour test-retest paradigms. Because behavioral frequencies are subject to dynamic real-world shifts over extended timeframes, a 48-hour test-retest window allows evaluation of measurement stability before true behavioral rates undergo meaningful secular change.

1. Test-Retest Reliability of Sexual Risk Items

Dowling-Guyer, Johnson, Fisher, and Needle (1994) conducted definitive psychometric evaluations of the revised RBA sexual risk variables across diverse cohorts. The 48-hour test-retest reliability coefficients exhibited domain-dependent variation:

• Number of Sexual Partners: The reliability coefficient for reporting the total number of distinct sexual partners in the past 30 days reached r = .92 (intraclass correlation / Pearson correlation), reflecting high precision.
• Vaginal Intercourse Frequency: Reporting of penile-vaginal intercourse frequency was highly stable, with coefficients ranging from .80 to .87 across cohorts.
• High-Frequency / Stable Behaviors: Overall, four core sexual risk behavior items demonstrated reliability coefficients greater than .80; four items scored between .70 and .79; two items yielded coefficients between .60 and .69; and individual items scored between .40 and .59.
• Low-Base-Rate Sexual Behaviors: Infrequently practiced or barrier-protected oral sex behaviors demonstrated lower test-retest coefficients (e.g., .07 for male reporting of cunnilingus with barrier protection). Methodological investigations indicate that such attenuated coefficients reflect extremely low baseline variance and floor effects rather than fundamental measurement unreliability (Dowling-Guyer et al., 1994).

2. Test-Retest Reliability of Drug Use and Paraphernalia Sharing

Initial studies by Needle et al. (1995) and Weatherby et al. (1994) revealed that while general substance use frequencies were exceptionally stable (coefficients consistently between .75 and .95), early iterations of syringe-sharing items showed variability due to ambiguous wording regarding direct versus indirect sharing. The revised RBA corrected these ambiguities by defining distinct paraphernalia elements (syringes, cookers, cotton, rinse water). Re-evaluation confirmed that 48-hour test-retest reliability for injecting frequency, frequency of using clean needles, and specific cooker sharing stabilized with coefficients exceeding .70 to .85 (Dowling-Guyer et al., 1994).

3. Reliability Across Auxiliary Domains

Subsequent psychometric investigations have established the temporal stability of the RBA’s supplemental modules:

• Alcohol Intake: Johnson, Pratt, Neal, and Fisher (in press) reported robust 48-hour test-retest reliability for past-30-day alcohol consumption measures, frequency of intoxication, and alcohol consumption immediately prior to sexual intercourse.
• Addiction Treatment History: Edwards, Fisher, Johnson, Reynolds, and Redpath (2007) confirmed high reliability for self-reported lifetime and recent treatment episodes (intraclass correlation coefficients > .80).
• Economic and Housing Variables: Johnson, Fisher, and Reynolds (1999) validated economic measures (legal income, illegal earnings, daily expenditures), and Klahn, Fisher, Wood, Reynolds, and Johnson (2004) proved the longitudinal stability of self-reported housing and homelessness chronicity.
• Longitudinal Assessment Stability (RBFA): Johnson et al. (2000) evaluated the 6-month follow-up version (RBFA) among out-of-treatment drug users, reporting 48-hour test-retest reliability coefficients ranging from .57 (number of drug-injecting sexual partners) to .87 (male reporting of vaginal sex frequency).

Factor Analysis

Unlike psychometric scales developed within classical test theory to measure latent psychological constructs (such as neuroticism, depression, or general intelligence), the RBA was constructed as an objective behavioral survey and epidemiological inventory. Because its indicators reflect discrete behavioral counts, the global instrument does not adhere to a single continuous unidimensional factor model. However, extensive exploratory factor analyses (EFA), confirmatory factor analyses (CFA), and structural equation models (SEM) have been performed on individual sections and integrated risk subsets.

1. Structural Factor Models of Drug Administration

Factor analytic treatments of Section B (Drug Use) and Section C (Injection Practices) consistently yield distinct, multi-factorial solutions rather than a generalized “addiction severity” factor. EFAs utilizing principal axis factoring and promax (oblique) rotation typically resolve into three to four robust factors:

• Factor 1: Opiate/Injection Polydrug Spectrum: Marked by high loadings (> .70) for daily heroin injection, speedball injection, frequency of syringe sharing, and cooker/cotton sharing.
• Factor 2: Stimulant/Crack Use Dimension: Characterized by high loadings (> .65) for crack cocaine inhalation, powder cocaine use, and frequency of high-binge administration.
• Factor 3: Non-Injection Substance Cluster: Comprising alcohol frequency, cannabis use, and non-prescribed oral sedative/hypnotic consumption.

2. Factor Structure of Sexual Risk Topology

Factor analyses conducted on Section F (Sexual Risk Behavior) reliably confirm a bifurcated structural pattern dividing risk into two distinct behavioral domains:

• Factor 1: High-Frequency Unprotected Penetrative Sex: Heavy loadings (> .75) for total counts of unprotected vaginal intercourse and unprotected anal intercourse with primary and regular sexual partners.
• Factor 2: Partner Multiplicity and Commercial Sex Trading: Dominated by loadings (> .60) for total number of casual partners, injection-drug-using partners, and exchanges of sex for drugs or monetary compensation (Section G).

Model fit indices for confirmatory factor models examining these risk subdomains in large community cohorts demonstrate acceptable parameters (e.g., Comparative Fit Index [CFI] > .93, Tucker-Lewis Index [TLI] > .91, and Root Mean Square Error of Approximation [RMSEA] < .05), validating the structural division between parenteral risk and sexual transmission risk.

Instrument / Measurement Tool

• Tool Name: Risk Behavior Assessment (RBA)
• Companion Tool: Risk Behavior Follow-up Assessment (RBFA) — Available in standardized 3-month and 6-month follow-up iterations.
• Assessment Type: Semi-structured, interviewer-administered epidemiological survey and behavioral assessment.
• Administration Time: 18 to 60 minutes (moderated by the respondent’s injection drug history, volume of sexual partners, and interviewer familiarity).
• Target Populations: Individuals at elevated risk for HIV, HBV, HCV, and STIs, primarily out-of-treatment injection drug users, crack cocaine users, commercial sex workers, and vulnerable urban populations.
• Structural Sections:
  • Section P: Preliminary Administration Data (Participant ID, interview date, start/end time, site code).
  • Section A: Sociodemographic Indicators (Age, biological sex, race, ethnicity, educational attainment).
  • Section B: Drug Use History (Lifetime and past-30-day frequency across major psychoactive categories).
  • Section C: Injection Drug Practices (Restricted to individuals injecting in the past 30 days; evaluates needle acquisition, multi-person syringe use, bleaching routines, and indirect sharing).
  • Section D: Recent Drug Use (Detailed 48-hour recall designed for bio-toxicological verification).
  • Section E: Substance Abuse Treatment History (Lifetime and recent detoxification, inpatient, outpatient, and pharmacotherapy admissions).
  • Section F: Sexual Risk Behavior (Branching modules stratified by gender and sexual partner orientation: MSM, MSW, MSMW, WSW, WSM, WSMW; records exact counts of unprotected vaginal, anal, and oral sex).
  • Section G: Sex Trading Practices (Commercial exchange of sexual access for money, goods, shelter, or psychoactive substances).
  • Section H: Health-Related Conditions (Historical diagnoses, testing records, and symptoms of HIV, hepatitis, tuberculosis, and bacterial STIs).
  • Section I: Arrest and Legal Incarceration History (Lifetime detention, prison stays, and recent 30-day incarceration episodes).
  • Section J: Economic and Living Circumstances (Legal/illegal sources of financial support, housing status, homelessness).
  • Section K: Interviewer Post-Administration Impressions (Standardized ratings evaluating participant comprehension, honesty, cooperativeness, and overall data accuracy).
• Response Modes:
  • Numerical Continuous Counts: Single-, double-, triple-, or quadruple-digit integers indicating exact frequencies (e.g., number of times, number of distinct individuals, number of days).
  • Categorical and Dichotomous Variables: Binary Yes/No responses, polytomous route-of-administration indicators, and nominal codes.
  • Standardized Missing Data Codes: Uniformly coded across all fields: Don’t Know / Unsure = 7, 77, or 777; Refused = 8, 88, or 888; Logical Skip / Not Applicable = 9, 99, 999, or 666.
• Scoring and Data Management: As an objective behavioral inventory, there are no proprietary composite scales or cut-off scoring criteria. Raw data are recorded into computerized systems (such as the historical NOVA-DE software, REDCap, or relational databases) producing structured datasets with pre-labeled syntax files compatible with SAS, SPSS, and R. Researchers derive custom indices, such as aggregate unprotected sexual events, proportion of protected injections, or total substance diversity scores.

Permissions, Fee, and Test Year

The Risk Behavior Assessment (RBA) was originally formulated in 1991 and revised substantially through 1993–1994 by the Community Research Branch of the National Institute on Drug Abuse (NIDA), an operating agency of the National Institutes of Health (NIH), United States Department of Health and Human Services.

Copyright and Public Domain Status: As an instrument authored by United States federal government employees and academic investigators funded through federal cooperative research agreements, the RBA and its companion follow-up assessments (RBFA) reside entirely in the public domain. No royalties, licensing agreements, or authorization fees are required to use, reproduce, adapt, or digitize the instrument for academic, non-profit, clinical, or public health research purposes.

Access to Full Instruments: Annotated codebooks, administration manuals, and the complete instrument battery can be accessed through the Substance Abuse and Mental Health Data Archive (SAMHDA) hosted by the Inter-University Consortium for Political and Social Research (ICPSR) at the University of Michigan.

References

• Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall.
• Dowling-Guyer, S., Johnson, M. E., Fisher, D. G., & Needle, R. (1994). Reliability of drug users’ self-reported HIV risk behaviors and validity of self-reported recent drug use. Assessment, 1(4), 383–392. https://doi.org/10.1177/107319119400100407
• Edwards, J. W., Fisher, D. G., Johnson, M. E., Reynolds, G. L., & Redpath, D. P. (2007). Test-retest reliability of self-reported drug treatment variables. Journal of Substance Abuse Treatment, 33(1), 7–11. https://doi.org/10.1016/j.jsat.2006.11.002
• Fisher, D. G., Kuhrt-Hunstiger, T. I., Orr, S. M., & Davis, D. C. (1999). Hepatitis B validity of drug users’ self-report. Psychology of Addictive Behaviors, 13(1), 33–38. https://doi.org/10.1037/0893-164X.13.1.33
• Fisher, D. G., Reynolds, G. L., Creekmur, B., Johnson, M. E., & Deaugustine, N. (2007). Reliability and criterion-related validity of self-report of syphilis. Sexually Transmitted Diseases, 34(6), 389–391. https://doi.org/10.1097/01.olq.0000243623.51860.b7
• Fisher, D. G., Reynolds, G. L., Jaffe, A., & Johnson, M. E. (2007). Reliability, sensitivity and specificity of self-report of HIV test results. AIDS Care, 19(5), 692–696. https://doi.org/10.1080/09540120601087889
• Johnson, M. E., Fisher, D. G., Montoya, I., Booth, R., Rhodes, F., Andersen, M., & Williams, M. (2000). Reliability and validity of not-in-treatment drug users’ follow-up self-reports. AIDS and Behavior, 4(4), 373–380. https://doi.org/10.1023/A:1026415516035
• Johnson, M. E., Fisher, D. G., & Reynolds, G. L. (1999). Reliability of drug users’ self-report of economic variables. Addiction Research, 7(3), 227–238. https://doi.org/10.3109/16066359909004386
• Johnson, M. E., Pratt, D. K., Neal, D. B., & Fisher, D. G. (2008). Drug users’ test-retest reliability of self-reported alcohol use on the Risk Behavior Assessment. Substance Use & Misuse, 43(11), 1675–1686. https://doi.org/10.1080/10826080802241228
• Klahn, J. A., Fisher, D. G., Wood, M. M., Reynolds, G. L., & Johnson, M. E. (2004). Homelessness among out-of-treatment drug users in Long Beach, CA. Journal of Social Distress and the Homeless, 13(3–4), 345–368. https://doi.org/10.1023/B:JOSH.0000046140.54045.c8
• Napper, L. E., Fisher, D. G., Johnson, M. E., & Wood, M. M. (2008, October). Reliability and validity of drug users’ self-reports of amphetamine use. Paper presented at the 136th Annual Meeting and Exposition of the American Public Health Association, San Diego, CA.
• Needle, R., Fisher, D. G., Weatherby, N., Chitwood, D., Brown, B., Cesari, H., Booth, R., Williams, M. L., Watters, J., Andersen, M., & Braunstein, M. (1995). Reliability of self-reported HIV risk behaviors of drug users. Psychology of Addictive Behaviors, 9(4), 242–250. https://doi.org/10.1037/0893-164X.9.4.242
• Paschane, D. M., Fisher, D. G., Cagle, H. H., & Fenaughty, A. M. (1998). Gonorrhea among drug users: An Alaskan versus a national sample. American Journal of Drug and Alcohol Abuse, 24(2), 285–297. https://doi.org/10.3109/00952999809001714
• Schlicting, E. G., Johnson, M. E., Brems, C., Wells, R. S., Fisher, D. G., & Reynolds, G. L. (2003). Validity of injecting drug users’ self report of hepatitis A, B, and C. Clinical Laboratory Science, 16(2), 99–106.
• Tourangeau, R., Rips, L. J., & Rasinski, K. (2000). The psychology of survey response. Cambridge University Press. https://doi.org/10.1017/CBO9780511819322
• Weatherby, N. L., Needle, R., Cesari, H., & Booth, R. E. (1994). Validity of self-reported drug use among injection drug users and crack cocaine users recruited through street outreach. Evaluation and Program Planning, 17(4), 347–355. https://doi.org/10.1016/0149-7189(94)90035-3

Items of the Scale