There is increasing evidence that contextual forms of cognitive-behavioral therapy (CBT) are effective in the management of chronic pain, yet little is understood about the factors that moderate or predict outcomes in these treatments. This systematic review aimed to identify pretreatment participant characteristics associated with positive treatment responses in contextual CBT for chronic pain. Medline, EMBASE, PsychINFO, and CENTRAL were searched to identify eligible studies. Studies were included if the participants were adults with chronic pain, designs were longitudinal, treatments focused on psychological flexibility or mindfulness, and reported results allowed for examination of moderators or predictors of standard treatment outcomes. Of 991 records initially identified, 20 were eligible for inclusion in the review. Some evidence suggested that baseline emotional functioning predicts treatment response, but the direction of this association varied between studies. Substantive findings were inconsistent and inconclusive, however, methodological limitations were consistent. These included treatment heterogeneity, and a lack of theoretical, a priori guidance in examining potential predictors. Future research should adopt a theoretically based approach to examining moderators in relation to specific treatment methods and therapeutic processes. Considering moderation without first considering mediation is probably a limited strategy. Perspective: In this systematic review we examined evidence for potential predictors or moderators of outcomes in contextual CBT for chronic pain. Substantive findings were inconclusive but important methodological limitations and a lack of theoretical guidance were found. Future research should explicitly plan relevant methods and follow clear theoretical models.