Willingness to provide a hair sample for drug testing: results from an anonymous multi-city intercept survey.

Background: Hair provision for drug testing can provide secondary measurement to complement self-reported drug use data, thereby providing a more accurate representation of an individual's drug use. Understanding factors associated with hair provision offers valuable insights into recruitment methods. Objective: To identify demographic and drug-related correlates of providing hair samples in a multi-site venue-intercept study. Methods: We utilized venue-intercept sampling for our Rapid Street Reporting study across 12 US cities between January and November 2022. Participants reported past 12-month drug use and were asked if they would provide a hair sample. We conducted multivariable (generalized linear model with logit link) analyses on demographics and drug use characteristics correlated to hair provision for drug testing. Results: Among 3,045 participants, 55.8% were male, 13.6% provided hair samples. Compared to males, those identifying as "other gender" had higher odds of hair collection (adjusted odds ratio = 2.24, 95% confidence interval: 1.28-3.80). Participants identifying as Black (aOR = 0.32, CI: 0.23-0.45) or "other race" (aOR = 0.50, 95% CI: 0.29-0.80) had lower odds of providing hair than those identifying as White. All levels of reported drug use - one drug (aOR=1.50, 95% CI: 1.15-1.96), two-three drugs (aOR=1.51, 95% CI: 1.11-2.05), four or more (aOR = 2.13, 95% CI: 1.50-3.01) - had higher odds of providing hair samples than those reporting no drug use. Similar associations applied to reporting cannabis use with or without another drug (aOR = 1.52-1.81, 95% CI: 1.15-2.38). Conclusion: Differential hair provision based on participant sex, race/ethnicity, and drug use may introduce biases in drug testing, limiting generalizability to individuals from minority backgrounds.