Arts therapy is a popular intervention used to work through the effects of traumatic experience. We evaluate previous reviews and report a meta‐analysis of the effectiveness of arts therapy following trauma for reducing symptoms of PTSD, enhancing positive outcomes (e.g., quality of life) and decreasing negative outcomes (e.g., depression). Database searches identified 21 (N = 868) randomised controlled trials (RCTs). Outcomes were categorised as PTSD specific, positive non‐PTSD specific and negative non‐PTSD specific. Several moderators were tested: age, diagnosis type, trauma type, intervention instruction, control type, therapy mode and therapy duration. Overall, random‐effects analysis indicated that arts therapy was favoured relative to control for positive non‐PTSD‐specific outcomes (g = 1.53, p < 0.001), but not for negative non‐PTSD‐specific (p = 0.069) or PTSD‐specific outcomes (g = 0.89, p = 0.052). Regression analyses indicated that arts therapy was effective in reducing PTSD‐specific outcomes in children (Z = 2.81, df = 1, p = 0.005), positive non‐PTSD‐specific outcomes in group‐based arts therapy (Z = −2.40, df = 1, p = 0.016, I2 = 57.33) and for reducing negative non‐PTSD outcomes following acute traumas (e.g., combat‐related trauma or sexual abuse) (Q = 10.70, df = 3, p = 0.013, I2 = 77.09). We highlight the need for additional RCTs and standardised protocols to address heterogeneity. Our review provides an important benchmark for gauging the effectiveness of arts therapy in the treatment of trauma.