Tactile Contact as a Marketing Tool for Improving an HIV/STD Education Program's Compliance / Retention with Crack Cocaine Users.

Abstract

Background: This research brief reports results from an exploratory pilot study on the use of socially acceptable touch in a public setting that accompanies a request to improve program compliance with "street level" crack cocaine users.

Methods: Study participants consisted of 120 crack cocaine-using participants in a larger community-based HIV/STD prevention and research program targeting at-risk African-Americans. They were required to return for a series of four booster health education sessions over 2-5 days and 6 month and 1 year follow-up assessments. The most difficult aspect of this program was no-shows for the second booster session; study participants who attended at least two sessions were much more likely to attend all sessions and complete the entire lengthy program. The program director randomly approached some participants after the first visit in a public setting and briefly touched them as part of a handshake; then, the director asked them to return for their follow-up sessions. Whether they were approached or not was random. Analysis comprised descriptive and non-parametric statistics.

Results: Ninety-three percent of participants who were asked to return and were touched returned for the second session; only 75% returned who had been asked to do so but were not touched. A statistically significant difference favored being touched and complying, as measured by second-session returning participants (p < .01), though it appeared the touch / request had more of a preventive than a promotional effect. Extraneous demographic and background factors were ruled out with the exception of age (older participants), which contributed slightly.

Conclusions: Results suggest that a request "anchored" to a socially acceptable public touch is promising in terms of improving program participation and engagement. Limitations and implications for future research are discussed.