Feasibility of Using SMS Text Message Survey Methods to Collect Weekly HIV-Risk Behavior Data Among 18-24 Year Old Urban and County Youth: A Success Story

Introduction: Youth aged 15-24 account for 45% of all new HIV infections, most infected via sexual intercourse.^1-3 This necessitates focusing on adolescent sexual risk-taking. Longitudinally assessing HIV-risk behaviors among inner-city high-risk youth is challenging due to transience, distrust, and other obstacles to multiple in-person visits.  Use of cell phones has emerged as a viable method to engage adolescents:  95% of 18-24 year olds own a cell phone and 97% of these owners text daily.⁴  Texting data collection (a) enhances retention (frequent contact; quick tracking of nonresponse); and (b) generates more accurate risk behavior data by reducing the reference period to just a few days or a week.^5-7  Short Message Service (SMS) text messaging applies to the most basic cell phone models; hence, reaching youth at all SES levels and contexts.

Methods:  Data are from a one-year study of 65 white and 65 black female and male inner-city and county youth ages 18-24 with two annual in-person visits and weekly interim texting.  Participants enroll in the texting component during the first in-person visit and receive 5-15 texts every Sunday afternoon.  Weekly items alternate among sex-only, drugs-only, sex and drugs, and violence.  Data collection began August 2012 and is on-going (current texting weeks range 1-16).  Respondents receive compensation each week for completing all texts.  Currently, 93 youth are enrolled in texting (53 females, 39 males). Enrollment will end at 130 youth in December 2012.

Results:  Of the 93 youth currently enrolled, 98% (N=91) have responded at least one week.  Two participants never responded.  Drop-out after the third week was rare.  Most participants completed all weekly items on the same day the initiating text was sent (mean = 89.7%) or the day after (mean = 6.2%).  The overall response rare across all weeks is 93%.  Nearly 80% (N=74) have complete data for all weeks with no gaps and 92.5% (N=86) have completed half or more of the weeks.  Reasons for non- or incomplete responding include incarceration, number changes, or inactive phones due to late payments (N=5; 5.4%).  Some participants are proactive, calling with their new cell numbers and to alert the team when their phones are not active and when phone activity will resume.   

Conclusions:  This study demonstrates that SMS texting is feasible, acceptable, and effective among adolescents ages 18-24 for data collection.  Strategies that reduce missing data and promote consistent responses include continuous compensation as well as timely monitoring and follow-up messaging.  The acceptability of this method has major implications for effectively intervening with and engaging youth in prevention and intervention programs.  Programs must reach youth “where” they are and when they need help.  Adolescents are a moving target, literally. Even youth without stable, permanent residences have cell phones with texting capabilities. Almost all youth ages 18 to 24 have a cell phone and most have unlimited texting plans.⁸  Sending and receiving text messaging is instantaneous. Youth read and respond to texts more quickly and consistently than phone calls or emails. This temporal, in-the-moment aspect of text messaging is invaluable for intervention efforts targeting youth.