Implementation of Daily Text Message Surveys Assessing Drug Use and Sexual HIV Risk Behaviors: Feasibility and Acceptability Among Urban Emerging Adults

Introduction: Drug use and sexual HIV risk behaviors (e.g., inconsistent condom use) are concerning among emerging adults who are reaching a developmental stage when both behaviors tend to peak. Daily and event-level research can help uncover the causal relationships between substance use and condom use behaviors across relationship types (e.g., main/regular partners, casual partners). With increasing mobile phone access, and young people’s preference for text messaging, we employed text messaging to assess 18-25 year-olds’ daily reports of drug use, sexual risk behaviors, motivations, and affect over 28 days. This study describes the implementation of this method and participants’ attitudes regarding the daily surveys. Methods: Participants were 111 emerging adults presenting in an urban Emergency Department who reported past 28-day drug use, sex without a condom, and having a cell phone with text messaging capability (M age = 22; 53% female; 45% Black, 38% White; 95% used marijuana). Participants received $2 for every completed survey, and a bonus of $5 for every 6 out of 7 surveys completed. Results: A total of 106 (95%) completed at least 1 of 28 surveys with an average of 18.9 surveys completed (SD = 7.9); 51% completed at least 75% of surveys. Among 93 participants (84%) who completed a 1-month follow-up, 97% said they would consider participating in a similar study. Most (77%) reported the surveys took less than 15 minutes to complete; 90% found the surveys not at all or only a little annoying. Participants were generally comfortable with the daily questions about drugs/alcohol (98%) and sex (95%) and also found it easy to provide accurate information about drugs/alcohol (98%) and sex (96%). Regarding privacy measures, 66% of participants used a password/code on their phone during the study and 72% said they always deleted the survey messages after daily completion. Although 10% reported that someone else saw their survey messages all of these participants indicated this did not bother them at all. Conclusions: In summary, participants in this sample seemed to find this methodology acceptable for assessing this sensitive information; however, compliance with daily surveys was lower than expected. Anecdotally, several participants noted that compliance was hindered by: interruptions in phone service (e.g., change carriers, late payment), network or carrier disruptions, or their personal schedules. In addition, survey length may have been burdensome, with over 20% of participants reporting that the surveys took 20+ minutes. Future analyses will examine drug use and other behavioral predictors of survey compliance.