Meta-analysis in medicine
Abstract
Background: Meta-analysis is a process of using statistical methods to review and combine the results of different, independent clinical studies. Glass first used the term meta-analysis in 1976, when he and his co-worker Mary Lee Smith statistically combined the results of 375 studies that evaluated the efficacy of psychotherapy. Meta-analysis is of particular importance in the assessment of therapeutic efficacy as individual studies do not provide an overview over a topic in its entirety. As their samples are too small, individual studies cannot provide a quantitative evaluation of the effect of treatment, nor can they test the null hypothesis. Prior to meta-analysis, the traditional method was a narrative discourse on previous findings, which, however, could be misleading and subjective. In the past few years, meta-analysis has been increasingly used in all fields of science. This is particularly evident in the medical science, where two other terms are used as well – the systematic review and evidence based medicine. Other methods that have advanced markedly are the decision analysis and cost-effectiveness analysis. All these methods are connected, and the latter two are an upgrade of the first two. Systematic reviews are exact summaries of the best evidence related to exactly specified clinical dilemmas. Special centres, like the Cochrane Collaboration have been organised in different places around the world, where systematic reviews of scientific literature and their own findings are published in databases collecting data on most appropriate therapies of individual illnesses. These reviews support the synthesis of best evidence for treatment or establishment of best medical practices. In this case, meta-analysis has a broader impact and is not just a statistical method for collecting study results.
Conclusions: In addition to reviewing and combining results, meta-analysis also offers an upgrade option for discovering and exploring differences in the existing scientific literature in the research field of interest and may provide the most plausible explanations or even result in the discovery of new knowledge. If well performed, meta-analysis can give practical answers to contraversal clinical issues and save costs of additional clinical experiments.
Downloads
References
Huque MF. Experiences with meta-analysis in NDA submissions. In: Proceedings of the biopharmac. Section of the Am Stat Association. Rockville: American Statistical Association; 1988. p. 28–33.
Sacks HS, Berrier J, Reitman D, Ancona-Berk VA, Chalmers TC. Meta-analyses of randomized controlled trials. New Engl J Med 1987; 316: 450–5.
Hunt M. How science takes stock. The story of meta-analysis. New York: Russell Sage Foundation; 1997.
Lipsey MW, Wilson DB. Practical meta-analysis. Thousand Oaks: Sage Publications; 2000.
Cooper H, Hedges LV. The handbook of research synthesis 1994. New York: Russell Sage Foundation; 1994.
Hedges LV, Olkin I. Statistical methods for meta-analysis. New York: Academic press; 1985.
Petitti DB. Meta-analysis, decision-analysis and cost-effectiveness analysis. New York, Oxford: Oxford University Press; 2000.
Eysenck HJ. Meta-analysis: an abuse of research integration. J Special Educat 1984; 18: 41–59.
Mulrow C, Cook DC. Systematic reviews. synthesis of best evidence for health care decisions. Philadelphia: American College of Physicians; 1998.
Anon. Cochrane Collaboration. Contact details for Cochrane Centres. Dostopno na: http://www.cochrane.org/contact/centres.htm#16
Clarke M, Oxman AD ed. Cochrane reviewers’ handbook 4.1 [updated June 2000]. In: Review Manager (Rev Man) [Computer program]. Version 4.1. Oxford, England: The Cochrane Collaboration; 2000.
Geyman JP, Deyo RA, Ramsey SD. Evidence-based clinical practice: concepts and approaches. Boston: Butterworth Heinemann; 2000.
Keene BW. Towards evidence-based veterinary medicine (editorial). J Vet Intern Med 2000; 14: 118–9.
Wachter KW. Disturbed by meta-analysis? Science 1988; 241: 1407–8.
Spilker B. Guide to clinical trials. New York: Raven Press; 1991.
Plath I. Understanding meta-analyses: a consumer’s guide to aims, problems, evaluation and developments. Baden-Baden: Nomos Verlagsgesellschaft; 1992.
Ščuka L. The meta-analysis of clinical experiments of the use of salinomycin as a growth promotor in pigs. Zb Vet Fak Univ Lj 1997; 34(1): 101–9.
Hobbs FDR. What should cardiologists be telling general practitioners/family physicians? Eur Heart J Suppl 2002; 4 Suppl F: F54–F9.
Altman DG, Sakarovitch C, Glenny AM, Deeks JJ, D’Amico R, Bradburn M, et al. Indirect comparisons – their use in systematic reviews. In: 3rd Symposium on Systematic Reviews. Oxford: Centre for statistics in medicine; 2000.
Chapman A, Middleton P, Maddern G. Early updates of systematic reviews – a waste of resources? In: 4th Symposium on systematic reviews: pushing the boundaries. Oxford: Centre for statistics in medicine; 2002.
Clarke M, Hopewell S. Time lag bias in publishing results of clinical trials: a systematic review. In: 3rd Symposium on systematic reviews. Oxford: Centre for statistics in medicine; 2000.
Glasziou P, Sanders S, Pirozzo S, Doust J, Pietrzak E. Abstract screening – the value of two reviewers. In: 4th Symposium on systematic reviews: pushing the boundaries. Oxford: Centre for statistics in medicine; 2002.
Dickersin K, Berlin JA. Meta-analysis: state-of-the-science. Epidemiol Rev 1992; 14: 154–76.
Gelber RD, Goldhirsch GH. Meta-analysis: the fashion of summing-up evidence. Ann Oncol 1991; 2: 461–8.
Ščuka L. Enrofloksacin – metaanaliza učinkovitosti zdravljenja obolenj pri domačih živalih (doktorsko delo). Ljubljana: Veterinarska fakulteta; 2003.
Boissel JP, Blanchard J, Panak E, Peyirieux JC, Sacks H. Considerations for the meta-analysis of randomized clinical trials. Control Clin Trials 1989; 10: 254–81.
Gerbarg ZB, Horwitz RI. Resolving conflicting clinical trials: guidelines for meta-analysis. J Clin Epidemiol 1988; 41(5): 503–9.
Light RJ, Pillemer DB. Summing Up. The science of reviewing research. Cambridge: Harvard University Press; 1984.
Demets DL. Methods for combining randomized clinical trials: strengths and limitations. Stat Med 1987; 6: 341–8.
Eysenck HJ. Meta-analysis and its problems. Br Med J 1994; 309: 789–92.
Eysenck HJ. Meta-analysis, sense or non-sense. Pharm Med 1992; 6: 113–9.
Naylor CD, Phil D. Two cheers for meta-analysis: problems and opportunities in aggregating results of clinical trials. Can Med Assoc J 1988; 138: 891– 5.
Slavin RE. Best evidence synthesis: an alternative to meta-analitic and traditional reviews. Educ Res 1986; 15: 5–11.
Borenstein M, Rothstein H. Comprehensive meta-analysis: manual for meta-analysis. Eaglewood, New York: Biostat; 1999.
The Author transfers to the Publisher (Zdravniški vestnik/Slovenian Medical Journal) all economic copyrights following form Article 22 of the Slovene Copyright and Related Rights Act (ZASP), including the right of reproduction, the right of distribution, the rental right, the right of public performance, the right of public transmission, the right of public communication by means of phonograms and videograms, the right of public presentation, the right of broadcasting, the right of rebroadcasting, the right of secondary broadcasting, the right of communication to the public, the right of transformation, the right of audiovisual adaptation and all other rights of the author according to ZASP.
The aforementioned rights are transferred non-exclusively, for an unlimited number of editions, for the term of the statutory
The Author can make use of his work himself or transfer subjective rights to others only after 3 months from date of first publishing in the journal Zdravniški vestnik/Slovenian Medical Journal.
The Publisher (Zdravniški vestnik/Slovenian Medical Journal) has the right to transfer the rights, acquired parties without explicit consent of the Author.
The Author consents that the Article be published under the Creative Commons BY-NC 4.0 (attribution-non-commercial) or comparable licence.
