In arm 1, sex workers and solvent users tended not to recruit other sex workers or solvent users (-0.202 and -0.083, respectively) while the opposite was true for arm 2 sex workers and solvent users (0.103 and 0.224, respectively).
Compared to arm 1 recruits, arm 2 seeds comprised more sex workers and solvent users, who tended to recruit people like themselves.
The population proportions estimated for solvent users and sex workers were higher in arm 2 than in arm 1 (0.43 for solvent users in arm 2 vs. 0.30 in arm 1 and 0.13 for sex work in arm 2 vs. 0.06 in arm 1).
IDU were those who had ever injected non-prescription drugs; solvent users were those who had ever sniffed any solvents (solvent use was a focus of our larger study and was included here to inform future analyses).
When we calculated risk estimates with a mutually exclusive categorical variable using nonusers of paint or solvents as the reference category, paint users had an OR of 1.84 (95% CI, 1.34 2.53), solvent users had an OR of 1.47 (95% CI, 0.93 2.32), and both exposures together produced an OR of 1.86 (95% CI, 1.37 2.53).
Examination of homophily values also revealed differences in recruitment from the two seed groups (e.g. in one arm of the study sex workers and solvent users tended not to recruit others like themselves, while the opposite was true in the second arm of the study).
The individuals in Arm 2, in particular the arm 2 seeds, may represent the most marginalized members of the overall population from which we were sampling (for example, based on their lower education and income levels and greater likelihood of being solvent users–see Table 2).
Arm 2 seeds only were more likely to have dropped out of school (OR of 1.83) and to have income from non-employment sources (OR of 2.47) while arm 2 recruits were more likely to be solvent users (OR of 1.60; arm 2 seeds approached significance for this variable with a p of 0.056 and OR of 1.62)) and less likely to be street-involved youth (OR of 0.34).
We further detailed exposures to paint and solvents by user (mother, father, others) and frequency of use (fewer than five times, or five times or more) in the corresponding time window.
