The broad aim of a meta-analysis is to provide a review of the literature on some scientific question and to summarize the information in a quantitative manner. It is hoped that by reviewing all the available studies a stronger consensus view can emerge. Meta-analysis taken in this sense is more than statistics and involves a whole methodology for research synthesis.

The statistical part of such a review covers several aspects, of which the combination of the statistical evidence from several experiments is the most important one. It is fair to say that unless the statistical methods being used are sound, the numerical summaries from a meta-analysis may be misleading and no benefits may accrue from combining the studies.

We are developing new statistical techniques for meta-analysis and research synthesis with application to research synthesis in medical and actuarial sciences. Of particular interest are multivariate and sequential methods of meta-analysis in random effects model setting.

EARS Seminars (East Anglian Research Synthesis)

NERC Advanced Training

References

  1. Kulinskaya, E., Morgenthaler, S., Staudte R.G.  Meta-analysis: A Guide to Calibrating and Combining Statistical Evidence. Wiley, Chichester,  2008
  2. Justine Naguib, Elena Kulinskaya , Claire Lomax , M Elena Garralda. (2009) Cognitive performance in children with type 1 diabetes – a meta analysis.-  Journal of Pediatric Psychology,   34: 271-282, http://jpepsy.oxfordjournals.org/cgi/reprint/jsn074
  3. Chisholm EJ, Kulinskaya E., Tolley NS (2009) Systematic review and meta-analysis of the adverse effects of thyroidectomy combined with central neck dissection as compared to thyroidectomy alone- The Laryngoscope, 119, 6, 1135-1139.
  4. Kulinskaya, E. and Koricheva, J. (2010), Use of quality control charts for detection of outliers and temporal trends in cumulative meta-analysis. Research Synthesis Methods, 1: 297–307. doi: 10.1002/jrsm.29
  5. Kulinskaya, E., Dollinger, MB and Bjørkestøl, K. (2011) Testing for Homogeneity in Meta-Analysis I. The One Parameter Case: Standardized Mean Difference, Biometrics, 67, 203–212, DOI:  10.1111/j.1541-0420.2010.01442
  6. Damodaram MS, Story L, Kulinskaya E, Rutherford M, Kumar, S. (2011) Early adverse perinatal complications in preterm growth restricted fetuses. The Australian and New Zealand Journal of Obstetrics and Gynaecology. 51: 204–209
  7. Kulinskaya, E., Dollinger, M. and Bjørkestøl, K.  (2012) On the moments of Cochran's Q statistic under the null hypothesis; with application to the meta-analysis of risk difference, Research Synthesis Methods, 2011, 2(4), 254-270, Article first published online: 4 MAR 2012 DOI: 10.1002/jrsm.54
  8. Kulinskaya, E. and Olkin,I. An overdispersion model in meta analysis. –Statistical Modelling: An International Journal. In print.
  9. Kulinskaya, E., Morgethaler, S. and Staudte R.G. Combining Statistical Evidence. International Statistical Review. doi:10.1111/insr.12037, in print
  10. Kulinskaya, E. and Wood, J. Trial sequential methods for meta-analysis. – Research Synthesis Methods, in print

Research Team

Prof Elena Kulinskaya, Samson Dogo, Ilyas Bakbergenuly

Collaborators

  • Prof Michael B Dollinger, Pacific Lutheran University, USA
  • Prof Richard Huggins, Melbourne, Australia
  • Prof Julia Koricheva, Royal Holloway, UK
  • Prof Stephan Morgenthaler, EPFL, Lausanne, Switzerland
  • Prof Ingram Olkin, Stanford, USA
  • Prof Robert G. Staudte, La Trobe University, Melbourne, Australia
  • Mr John Wood, UCL, UK