Non-therapeutic risk factors for onset of tardive dyskinesia in schizophrenia: A meta-analysis


  • Diederik E. Tenback MD, PhD,

    Corresponding author
    1. Psychiatric Center Symfora Group, DB, Amersfoort, The Netherlands
    2. Department of Psychiatry, University Medical Center Utrecht, Utrecht, The Netherlands
    • Psychiatric Center Symfora Group, P.O. Box 3051, 3800 DB, Amersfoort, The Netherlands
    Search for more papers by this author
  • Peter N. van Harten MD, PhD,

    1. Psychiatric Center Symfora Group, DB, Amersfoort, The Netherlands
    2. University Medical Center Groningen, Groningen, The Netherlands
    Search for more papers by this author
  • Jim van Os MD, PhD

    1. Department of Psychiatry and Neuropsychology, South Limburg Mental Health Research and Teaching Network, EURON, Maastricht University, Maastricht, The Netherlands
    2. Division of Psychological Medicine, Institute of Psychiatry, London, United Kingdom
    Search for more papers by this author

  • This article is part of the journal’s online CME program. The CME activity including form, can be found online at

  • Potential conflict of interest: The authors have no conflict of interest to declare. All authors report no financial disclosures related to the research covered in the article.


A meta-analysis of prospective studies with schizophrenia patients was conducted to examine whether the evidence exists for risk factors for the emergence of Tardive Dyskinesia (TD) in schizophrenia. A computer assisted Medline/PubMed and Embase search was conducted in January 2008 for the years 1985–2007. Selected were truly prospective studies of incident cases of TD in a population with at least 80% patients with schizophrenia. Measures of relative risk were collected from the individual studies, either directly or by calculating the relative risk from the cox- or logistic regression coefficient provided in the article. Hazard Ratio's and Odds Ratio's were pooled using fixed and random effect models in case of multiple studies using the same measure of risk and outcome. Only eight studies satisfied the inclusion criteria reporting on 25 different single estimate risk factors. Of 25 risk factors, six concerned replicated estimates suitable for meta-analysis. Of these, non-white ethnic group and early extrapyramidal symptoms qualified as risk factors for the emergence of TD in schizophrenia. The association with older age was suggestive but inconclusive. Despite many reported risk factors for TD in schizophrenia, little conclusive evidence exists to corroborate this. However, the fact that early EPS predicts onset of TD has important clinical and research implications. © 2009 Movement Disorder Society