"The model used here , predicted the number of new HIV and HCV cases prevented based on the needle sharing rate. This included the impact of behavioral changes in injection activities outside of the SIF. The behavioral change, according to Table 2 and Table 3, was only considered twice (once for the first SIF and later for the second SIF)—this modeling decision is apparent in the marginal number of new HIV cases averted in Tables 3, 4 and 5. This calculation of behavioral impact is based on a conservative odds-ratio that falls within the limit specified by Kerr et al. (2005) .
"As expected, the results presented in Table 2 and Table 3 show that increasing the scope of SIFs through site expansion would result in a decrease of HIV infection cases. The model predicts: 14–53 fewer HIV cases and 84–327 fewer HCV cases annually, with the marginal range being much smaller: 5–14 fewer HIV cases and 33–84 fewer HCV cases annually.
"This range disparity, as outlined in Table 2 and Table 3, translates into substantial differences between the economic evaluation of SIFs with respect to the cumulative versus marginal estimates: the total effect of establishing SIFs and the effect of establishing each subsequent SIF, respectively.
"For example, according to Table 3, the cumulative annual estimates of new HIV cases averted, translates into a cost savings for society ranging from $0.764 million (benefit) for the first SIF to -$4.1 million (loss) for the seventh SIF. Benefit-cost ratios range from 1.35 to 0.73, and cost-effectiveness values range from $155,914 to $288,294 (cost per lifetime treatment). The cumulative annual estimates of new HCV cases averted translate into a cumulative cost savings that range from $0.769 million (benefit) for the first SIF to -$3.7 million (loss) for the seventh SIF. Benefit-cost ratios range from 1.35 to 0.73, and incremental cost-effectiveness values range from $25,986 to $46,727 (cost per lifetime treatment).
"In contrast, the marginal estimates of Montreal’s SIF expansion translate into a much smaller return. This is particularly true with respect to its benefit-cost and cost-effectiveness ratios. For instance, the marginal benefit-cost ratio varies from 1.35 to 0.77 for HIV and 1.35 to 0.76 for HCV. The marginal cost-effectiveness value for HIV ranges from $155,914 to $436,560 (cost per life- time treatment). The HCV marginal cost-effectiveness value ranges from $25,986 to $66,145 (cost per lifetime treatment)."
Jozaghi et al., "A cost-benefit/cost-effectiveness analysis of proposed supervised injection facilities in Montreal, Canada." Substance Abuse Treatment, Prevention, and Policy 2013 8:25. doi:10.1186/1747-597X-8-25.