Predicted effect size of lisdexamfetamine treatment of attention deficit/hyperactivity disorder (ADHD) in European adults: Estimates based on indirect analysis using a systematic review and meta-regression analysis - 26/02/15

Doi : 10.1016/j.eurpsy.2015.01.001

M. Fridman ^a,^{^{1
Tel.: +1 323 857 6618; fax: +1 323 857 6656.}}, P.S. Hodgkins ^b,^{^{2
Currently: Vertex Pharmaceuticals, 50 Northern Avenue, Boston, MA 02210 USA. Tel.: +1 617 961 7003; fax: +1 617 366 3967.}}, J.S. Kahle ^c,^⁎ , M.H. Erder ^d,^{^{3
Tel.: +1 484 595 8346; fax: +1 484 595 8178.}}
^a Statistics and Epidemiology, AMF Consulting, 846 S. Citrus Avenue, Los Angeles, CA 90036, USA
^b Global Health Economics & Outcomes Research, Shire, 725 Chesterbrook Boulevard, Wayne, PA 19087, USA
^c Research, BPS International, 3830 Valley Centre #705 PMB503, San Diego, CA 92130, USA
^d Health Economics and Epidemiology, Shire, 725 Chesterbrook Boulevard, Wayne, PA 19087, USA

^⁎Corresponding author at: BPS International, 3830 Valley Centre #705 PMB503, San Diego, CA 92130 USA. Tel.: +1 858 509 2711; fax: +1 858 509 2711.

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Thursday 26 February 2015
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder

Abstract

Background

There are few approved therapies for adults with attention-deficit/hyperactivity disorder (ADHD) in Europe. Lisdexamfetamine (LDX) is an effective treatment for ADHD; however, no clinical trials examining the efficacy of LDX specifically in European adults have been conducted. Therefore, to estimate the efficacy of LDX in European adults we performed a meta-regression of existing clinical data.

Methods

A systematic review identified US- and Europe-based randomized efficacy trials of LDX, atomoxetine (ATX), or osmotic-release oral system methylphenidate (OROS-MPH) in children/adolescents and adults. A meta-regression model was then fitted to the published/calculated effect sizes (Cohen's d) using medication, geographical location, and age group as predictors. The LDX effect size in European adults was extrapolated from the fitted model. Sensitivity analyses performed included using adult-only studies and adding studies with placebo designs other than a standard pill-placebo design.

Results

Twenty-two of 2832 identified articles met inclusion criteria. The model-estimated effect size of LDX for European adults was 1.070 (95% confidence interval: 0.738, 1.401), larger than the 0.8 threshold for large effect sizes. The overall model fit was adequate (80%) and stable in the sensitivity analyses.