Assessing risk of bias in randomized controlled trials of methylphenidate for children and adolescents with attention deficit hyperactivity disorder (ADHD)

Int J Methods Psychiatr Res. 2018 Mar;27(1):e1586. doi: 10.1002/mpr.1586. Epub 2017 Sep 4.

Abstract

To test how reliable the tool recommend by Cochrane Collaboration for assessing risk of bias systematic reviews of randomized clinical trials is in the context of methylphenidate for children and adolescents with attention deficit hyperactivity disorder. Confirmatory factor analysis was used to evaluate a unidimensional model for the 7 indicators, applied to 184 Randomized Clinical Trial (RCTs) within a 2015 Cochrane systematic review titled "Methylphenidate for children and adolescents with attention deficit hyperactivity disorder." A unidimensional model resulted in excellent adequacy indices, but only 2 indicators had very high factor loadings and low measurement errors. In terms of content, the 7 indicators showed poor reliability (ω = 0.642); however, the set of indicators was precise in evaluating studies with a high amount of bias risk. The Cochrane model of risk of bias as it is, exhibited good fit indices but the majority of the items were not reliable to adequately capture risk of bias in the context of clinical trials of methylphenidate for ADHD.

Keywords: ADHD; confirmatory factor analysis; meta-analysis; methylphenidate; risk of bias.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Attention Deficit Disorder with Hyperactivity / drug therapy*
  • Bias*
  • Central Nervous System Stimulants / pharmacology*
  • Child
  • Female
  • Humans
  • Male
  • Methylphenidate / pharmacology*
  • Models, Statistical*
  • Outcome and Process Assessment, Health Care / standards*
  • Randomized Controlled Trials as Topic / standards*
  • Research Design / standards*
  • Risk

Substances

  • Central Nervous System Stimulants
  • Methylphenidate