Page 33 - The efficacy and effectiveness of psychological treatments for eating disorders - Elske van den Berg

Chapter 2 33
Removing three outlier arms lowered heterogeneity to I2 = 30, 95% CI [0, 59]. No indi- cation for publication bias was found. Sensitivity analysis including only the largest effect size of each study resulted in an overall effect size of g = 0.10, 95% CI [-0.09, 0.28]. When only the smallest effect size was included, the pooled effect size was g = 0.06, 95% CI [-0.10, 0.21]. Including only high-quality studies in a sensitivity anal- ysis did not result in a significant difference between treatment and control condi- tions. Subgroup analyses showed that studies including patients of 18 years of age had a significantly higher effect size, g = 0.23, 95% CI [0.06, 0.40], than studies including patients under 18, g = -0.18, 95% CI [-0.41, 0.6], p = .006. High-quality studies differed significantly, g = 0.27, 95% CI [0.02, 0.52], from lower quality studies, g = -0.08, 95% CI [-0.24, 0.09], p = .025, with regard to the effect on weight gain.
Meta-regression analysis showed that publication year was not significantly associated with effect on weight gain, b = 0.002, 95% CI [-0.01, 0.02], p = .813.
FIGURE 2
Standardized mean differences of psychological treatments for anorexia nervosa compared with control conditions on weight gain
ABW = average body weight; FPT = Focal psychodynamic therapy; FT= family therapy; IBW = Ideal body weight; IP= inpatient; IPT = Interpersonal psychotherapy; MANTRA = Maudsley Model of Anorexia Nervosa Treatment for Adults; OP = outpatient; sub = subgroup