Meta-analyses conclude that cognitive behavioural therapy (CBT) for perfectionism is efficacious without addressing indices of efficacy such as reliable improvement, deterioration, dropout, and change at follow-up. We addressed this through a reanalysis of the 16 randomized controlled trials included in Galloway et al. (2022), Robinson andWade (2021), andSuh et al. (2019). At posttreatment, small-to-large effects favouring CBT were found for certain perfectionism measures: high standards (g = −0.40), doubts about actions (g = −0.49), personal standards (g = −0.50), concern over mistakes (g = −0.85), and clinical perfectionism (g = −0.97). Small-to-medium posttreatment effects were found for symptoms of eating disorders (g = −0.29), anxiety (g = −0.37), and depression (g = −0.62). In contrast, self-oriented perfectionism (g = −0.60; 95% CI [−1.96, 0.78]), other-oriented perfectionism (g = −0.36; 95% CI [−1.15, 0.43]), socially prescribed perfectionism (g = −0.53; 95% CI [−1.58, 0.51]), perfectionistic cognitions (g = −0.70; 95% CI [−1.57, 0.17]), discrepancy (g = −0.48; 95% CI [−1.83, 0.87]), life satisfaction (g = −0.59; 95% CI [−1.02, 0.01]), and self-esteem (g = −0.53; 95% CI [−1.25, 0.18]) did not differ between treatment and control conditions. Risk ratios for reliable improvement were significant for perfectionistic cognitions (RR = 1.46), concern over mistakes (RR = 2.36), and clinical perfectionism (RR = 3.07). Dropout was 27.0% and higher in treatment than control conditions (RR = 1.78). At follow-up, all between-group effects were nonsignificant. Findings support the efficacy of CBT for certain perfectionism features while underscoring crucial limitations and areas for improvement.
Public Significance StatementThe importance of evaluating treatments for perfectionism is crucial given the deleterious effects of this personality variable. Although there are indications that CBT for perfectionism is efficacious, the current work indicates that a substantial proportion of participants do not experience reliable posttreatment improvements. Moreover, for those who experience improvements, it is unclear how much improvement extends beyond posttreatment. Dropout also appears higher in treatment conditions than in control conditions, suggesting that CBT for perfectionism might not be well tolerated by many individuals with perfectionism. Developing and refining treatment for perfectionism is urgently needed.