Volume 70, Issue 1
Original Article

Meta‐CART: A tool to identify interactions between moderators in meta‐analysis

Xinru Li

Corresponding Author

E-mail address: x.li@math.leidenuniv.nl

Mathematical Institute, Leiden University, The Netherlands

Correspondence should be addressed to Xinru Li, Mathematical Institute, Leiden University, PO Box 9512, 2300 RA Leiden, The Netherlands (email: x.li@math.leidenuniv.nl).Search for more papers by this author
Elise Dusseldorp

Institute of Psychology, Leiden University, The Netherlands

Search for more papers by this author
Jacqueline J. Meulman

Mathematical Institute, Leiden University, The Netherlands

Search for more papers by this author
First published: 28 January 2017
Citations: 11

Abstract

In the framework of meta‐analysis, moderator analysis is usually performed only univariately. When several study characteristics are available that may account for treatment effect, standard meta‐regression has difficulties in identifying interactions between them. To overcome this problem, meta‐CART has been proposed: an approach that applies classification and regression trees (CART) to identify interactions, and then subgroup meta‐analysis to test the significance of moderator effects. The previous version of meta‐CART has its shortcomings: when applying CART, the sample sizes of studies are not taken into account, and the effect sizes are dichotomized around the median value. Therefore, this article proposes new meta‐CART extensions, weighting study effect sizes by their accuracy, and using a regression tree to avoid dichotomization. In addition, new pruning rules are proposed. The performance of all versions of meta‐CART was evaluated via a Monte Carlo simulation study. The simulation results revealed that meta‐regression trees with random‐effects weights and a 0.5‐standard‐error pruning rule perform best. The required sample size for meta‐CART to achieve satisfactory performance depends on the number of study characteristics, the magnitude of the interactions, and the residual heterogeneity.

Number of times cited according to CrossRef: 11

  • Impact of dissociation on the effectiveness of psychotherapy for post-traumatic stress disorder: meta-analysis, BJPsych Open, 10.1192/bjo.2020.30, 6, 3, (2020).
  • Multiple moderator meta-analysis using the R-package Meta-CART, Behavior Research Methods, 10.3758/s13428-020-01360-0, (2020).
  • A flexible approach to identify interaction effects between moderators in meta‐analysis, Research Synthesis Methods, 10.1002/jrsm.1334, 10, 1, (134-152), (2019).
  • A history of meta‐regression: Technical, conceptual, and practical developments between 1974 and 2018, Research Synthesis Methods, 10.1002/jrsm.1338, 10, 2, (161-179), (2019).
  • Are Emotionally Intelligent People More Creative? A Meta-Analysis of the Emotional Intelligence–Creativity Link, Sustainability, 10.3390/su11216123, 11, 21, (6123), (2019).
  • Do automated digital health behaviour change interventions have a positive effect on self-efficacy? A systematic review and meta-analysis, Health Psychology Review, 10.1080/17437199.2019.1705873, (1-107), (2019).
  • Interventions to Promote Healthy Eating, Physical Activity and Smoking in Low-Income Groups: a Systematic Review with Meta-Analysis of Behavior Change Techniques and Delivery/Context, International Journal of Behavioral Medicine, 10.1007/s12529-018-9734-z, 25, 6, (605-616), (2018).
  • How effective are interventions in improving dietary behaviour in low- and middle-income countries? A systematic review and meta-analysis, Health Psychology Review, 10.1080/17437199.2018.1481763, 12, 3, (312-331), (2018).
  • Meta-Analysis of Complex Interventions, Annual Review of Public Health, 10.1146/annurev-publhealth-040617-014112, 39, 1, (135-151), (2018).
  • IMP2ART systematic review of education for healthcare professionals implementing supported self-management for asthma, npj Primary Care Respiratory Medicine, 10.1038/s41533-018-0108-4, 28, 1, (2018).
  • Key Elements of mHealth Interventions to Successfully Increase Physical Activity: Meta-Regression, JMIR mHealth and uHealth, 10.2196/10076, 6, 11, (e10076), (2018).