Unstable housing as a factor for increased injection risk behavior at US syringe exchange programs.

OBJECTIVE: To assess variation in injection risk behavior among unstably housed/homeless injecting drug users (IDUs) across programs in a national sample of US syringe exchange programs.
METHODS: About 23 syringe exchange programs were selected through stratified random sampling of moderate to very large US syringe exchange programs operating in 2001-2005. Subjects at each program were randomly sampled. Risk behavior interviews were collected using audio-computer assisted self-interviewing (A-CASI). "Unstable housing/homelessness" was operationally defined as having lived "on the street or in a shanty" or "living in a shelter or single room occupancy hotel (SRO)" at any time in the 6 months prior to the interview. "Receptive sharing" was operationally defined as having injected with a needle or syringe that "had been used by someone else" in the 30 days prior to the interview. Six very large and nine moderate-to-large programs had at least 50 subjects who reported unstable housing, and these 15 programs were used in the analyses.
RESULTS: There was considerable variation among the 15 programs in the percentages of unstably housed participants (range from 35 to 74%, P < 0.0001), and in the percentages of unstably housed participants who reported receptive sharing (range from 8 to 52%, P < 0.0001). At each of the 15 programs, unstably housed exchange participants were approximately twice as likely to report receptive sharing than were stably housed participants. The weighted mean odds ratio was 2.02, 95% confidence interval, 1.68-2.41 (random effects model) and there was no statistically significant variation in these odds ratios. Across the 15 programs, receptive sharing among unstably housed participants was highly correlated with receptive sharing among stably housed participants (r = 0.95, P < 0.001, 90% of variance among unstably housed "explained" by variance among stably housed).
CONCLUSIONS: The programs clearly differ in the extent to which they are attracting unstably housed IDUs as participants. The consistency of more frequent injection risk behavior among unstably housed exchange participants and the lack of significant variation in the odds ratios for increased injection risk suggests that none of the programs were "better" or "worse" at reducing injection risk behavior among unstably housed participants. Reduction in injecting risk behavior among syringe exchange participants may require greater efforts to provide stable housing or the development of dramatically new interventions to reduce injecting risk behavior among IDUs with persistent unstable housing.