gms | German Medical Science

MAINZ//2011: 56. GMDS-Jahrestagung und 6. DGEpi-Jahrestagung

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V.
Deutsche Gesellschaft für Epidemiologie e. V.

26. - 29.09.2011 in Mainz

Methodological quality of pharmacogenetic studies: Issues of concern?

Meeting Abstract

Search Medline for

  • Andrea L. Jorgensen - Department of Biostatistics, The University of Liverpool, Liverpool
  • Paula R. Williamson - Department of Biostatistics, The University of Liverpool, Liverpool

Mainz//2011. 56. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie (gmds), 6. Jahrestagung der Deutschen Gesellschaft für Epidemiologie (DGEpi). Mainz, 26.-29.09.2011. Düsseldorf: German Medical Science GMS Publishing House; 2011. Doc11gmds593

doi: 10.3205/11gmds593, urn:nbn:de:0183-11gmds5931

Published: September 20, 2011

© 2011 Jorgensen et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.



Genetic variation has been linked to variability in drug response since the 1950s, but the uptake of pharmacogenetic tests into clinical practice has been poor, not least due to the lack of randomised controlled trials to test their efficacy over standard clinical practice. Before investing in clinical trials of a pharmacogenetic test, evidence supporting the genetic association needs to be replicated in more than one pharmacogenetic study. Failure to replicate is a notorious problem in this field, with variability in methodological quality and inadequate sample sizes being significant contributing factors. Many authors have highlighted issues of concern regarding methodological quality of genetic association studies in general and we consider their relevance in the context of pharmacogenetic studies. Two additional issues unique to the conduct of pharmacogenetic studies are also identified: assessing for compliance with treatment, and the choice and definition of outcomes to best capture treatment response. The importance of the former is demonstrated in a pharmacogenetic study of warfarin, with results compared between analyses both with and without adjusting for the extent of compliance. The relevance of the latter is demonstrated in a systematic review of warfarin pharmaocgenetics, which identified that a third of studies did not provide a definition of stable dose, with 25 different definitions given in the remaining 36 studies. Methods to deal with these issues and the risks from not doing are also discussed. We also present a checklist of methodological quality that has grown from our research, for use in the design, analysis and reporting of pharmacogenetic studies. With a view to improving power to detect genetic associations, researchers are increasingly turning to systematic reviews and meta-analyses of pharmacogenetic studies. On reviewing the literature for guidelines on conducting such reviews, no specific criteria for assessing the quality of included studies – a critical part of the systematic review process - were identified. Our checklist of methodological quality is also proposed for this purpose. By assessing quality of studies in this way, their methodological rigor can be evaluated, appropriateness of their inclusion in a meta-analysis can be considered, and the current evidence base on genetic influences on treatment response can be formally assessed with any weaknesses identified.