Our investigation of moderation by labeling status yielded interesting findings. The significantly larger mediated relationship for those classified as labeled compared to unlabeled supports the general proposition of modified labeling theory that the negative outcomes of stigma are dependent on the labeling process (Link et al., 1989). The results are consistent with studies on non-substance-related psychiatric illness stigma that conducted stratified analyses or formal tests of moderation and found an interaction between stigma and labeling status, with previously treated individuals having poorer outcomes (Kroska and Harkness, 2006; Link et al., 1989, 1991, 1997). We caution that this finding should not be interpreted such that treatment should be avoided. Rather, decreasing stigma would be considered the appropriate target for intervention, which might include targeting the reduction of stigma among the general population and treatment providers (Livingston et al., 2012; McLaughlin and Long, 1996).
