A preliminary systematic review and meta-analysis of randomized-controlled trials of cognitive remediation therapy for anorexia nervosa.

Cognitive remediation therapy (CRT) for anorexia nervosa (AN) was developed as an adjuvant treatment to target set-shifting and central coherence inefficiencies important in AN and to ultimately improve clinical outcomes of those with AN. The primary aim of this preliminary systematic review and meta-analysis was to determine the effect of CRT for AN relative to control treatments in randomized-controlled trials (RCTs) on neuropsychological inefficiencies at end-of-treatment. Secondary aims were to assess the effect of CRT for AN on dropout, eating-disorder-related, and other psychological outcomes at end-of-treatment. Systematic review and meta-analytic procedures were conducted in accordance with PRISMA Guidelines. RCTs evaluating CRT for AN compared to a control treatment were identified via ProQuest, PsycINFO, PubMed, and SCOPUS. Seven RCTs and one quasi-RCT of CRT for AN were included. RCT quality ratings ranged from fair (n = 3) to good (n = 4). Random-effects meta-analysis was conducted using Hedge's g. Study heterogeneity was assessed using I2 and publication bias was assessed with Begg's adjusted-rank correlation and the trim-and-fill method. CRT was not associated with improvement in central coherence compared to control treatments at end-of-treatment (g = 0.25, 95% CI = -0.35, 0.85, k = 3). Set-shifting outcomes were mixed due to heterogeneity of set-shifting measures across studies. CRT may prevent dropout; yet, more studies are needed to draw conclusions. CRT did not confer advantage over control treatments for eating-disorder-related and other psychological outcomes at end-of-treatment. Future RCTs of CRT for AN should use precise measures to assess constructs (particularly for set shifting), increase sample size, and implement longitudinal follow-up. (Word Count: 247 words).