Comparison of three intervention models for promoting circumcision among migrant workers in western China to reduce local sexual transmission of HIV.

Objective: Three models for promoting male circumcision (MC) as a preventative intervention against HIV infection were compared among migrant worker populations in western China.

Methods: A cohort study was performed after an initial cross-sectional survey among migrant workers in three provincial level districts with high HIV prevalence in western China. A total of 1,670 HIV seronegative male migrants were cluster-randomized into three intervention models, in which the dissemination of promotional materials and expert- and volunteer-led discussions are conducted in one, two, and three stage interventions. Changes in knowledge of MC, acceptability of MC, MC surgery uptake, and the costs of implementation were analyzed at 6-month and 9-month follow-up visits.

Results: All three models significantly increased the participants' knowledge about MC. The three-stage model significantly increased the acceptability of MC among participants and led to greatest increase in MC uptake. At the end of follow-up, 9.2% (153/1,670) of participants underwent MC surgery; uptake among the one-, two-, and three-stage models were 4.9%, 9.3%, and 14.6%, respectively. Multivariable Cox regression analysis showed that three-stage model was the most effective method to scale up MC, with RR = 2.0 (95% CI, 1.3-3.1, P=0.002) compared to the on-site session model. The two-stage intervention model showed no significant difference with either the on-site session model (RR=1.5, 95% CI, 0.92-2.4, P=0.12) or three-stage model (P=0.10).

Conclusions: A three-stage intervention with gradual introduction of knowledge led to the significantly increase in MC uptake among migrant workers in western China, and was also the most cost-effective method among the three models.