Respondent-driven sampling in the recruitment of illicit stimulant drug users in a rural setting: findings and technical issues.

Increasingly, respondent-driven sampling (RDS) is being applied to study not-in-treatment users of illicit drugs. Although RDS has been successfully applied in recruiting active users in densely-populated, metropolitan areas, its utility with hidden populations in rural areas has yet to be determined. This study critically evaluates the sample of not-in-treatment, illicit stimulant drug users (n=249) recruited from the application of RDS in three rural counties in west-central Ohio. The findings of this study largely support the results of earlier studies in urban areas and suggest that RDS is a useful method of sampling hidden drug using populations in less densely populated rural areas. Some limitations of RDS are also discussed.