Self-help treatments have the potential to increase the availability and affordability of evidence-based treatments for anxiety disorders. Although promising, previous research results are heterogeneous, indicating a need to identify factors that moderate treatment outcome. The present article reviews the literature on self-help treatment for anxiety disorders among adults, with a total sample of 56 articles with 82 comparisons. When self-help treatment was compared to wait-list or placebo, a meta-analysis indicated a moderate to large effect size (g = 0.78). When self-help treatment was compared to face-to-face treatment, results indicated a small effect that favored the latter (g = -0.20). When self-help was compared to wait-list or placebo, subgroup analyses indicated that self-help treatment format, primary anxiety diagnosis and procedures for recruitment of subjects were related to treatment outcome in bivariate analyses, but only recruitment procedures remained significant in a multiple meta-regression analysis. When self-help was compared to face-to-face treatment, a multiple meta-regression indicated that the type of comparison group, treatment format and gender were significantly related to outcome. We conclude that self-help is effective in the treatment of anxiety disorders, and should be offered as part of stepped care treatment models in community services. Implications of the results and future directions are discussed.