The term meta-analysis refers to the quantitative combination of data from independent trials. Where the result of such combination is a descriptive summary of the weight of the available evidence, the exercise is of undoubted value. Attempts to apply inferential methods, however, are subject to considerable methodolo gical and logical difficulties. The selection and quality of the trials included, population bias, and the specification of the population to which inference may properly be made are problems to which no satisfactory solutions have been proposed. Insightful quantitative description ought not to differ materially from inferential conclusions; where discrepancies exist the inferential techniques should be regarded with extreme caution.
Get full access to this article
View all access options for this article.
References
1.
Oakes M.Statistical inference. Chestnut Hill: E.R.I., 1990. (Foreword by Greenland S.)
2.
Hedges L., Olikin I.Statistical methods for meta-analysis. New York: Academic Press, 1985.
3.
Louis TA, Fineberg HV, Mosteller F.Findings for public health for meta-analyses. Annual Review of PublicHealth1985; 6: 1-20.
4.
Sacks HS, Berrier J., Reitman D., Ancona-Berk VA, Chalmers TCMeta-analysis of randomized control trials. New England Journal of Medicine1987; 316: 450-55.
5.
L'Abbe KA, Detsky AS, O'Rourke K.Meta-analysis in clinical research. Annals of Internal Medicine1987; 107: 224-33.
6.
Light RJAccumulating evidence from independent studies: What we can win and what we can lose. Statistics in Medicine1987; 6: 221-28.
7.
Skinner JBOn combining studies. Drug Information Journal1991; 25: 395-403.
8.
Chalmers TCProblems induced by meta-analyses. Statistics in Medicine1991; 10: 971-80.
9.
Yusuf S., Peto R., Lewis J., Collins R., Sleight P.Beta blockade during and after myocardial infarction: An overview of the randomized trials. Progress in Cardiovascular Diseases1985; 27: 335-71.
10.
Yusuf S., Collins R., Peto R. et al. Intravenous and intravenous fibrinolytic therapy in acute myocardial infarction: Overview of results on mortality, reinfarction, and side effects from 33 randomized controlled trials. European Heart Journal1985; 6: 556-85.
11.
Hine LK, Laird N., Hewitt P., Chalmers TCMeta-analytic evidence of empirical long-term antiarrhythmic therapy after myocardial infarction. Journal of Clinical Epidemiology1989; 42: 1037-40.
12.
Fleiss J.Discussion contribution to Light RJ. Statistics in Medicine1987; 6: 221-28.
13.
Peto R.Discussion contribution to Light RJ. Statistics in Medicine1987; 6: 221-28.
14.
Yusuf S.Obtaining medically meaningful answers from an overview of randomized clinical trials. Statistics in Medicine1987; 6: 281-86.
15.
Glass GVCommentary on Rosenthal R, Rubin BD. Interpersonal expectancy effects: the first 345 studies. The Behavioural and Brain Sciences1978; 3: 394-95.
16.
Cook TD, Leviton LCReviewing the literature: A comparison of traditional methods with meta-analysis . Journal of Personality1980; 48,4:449-72.
17.
Sterlin TDPublication decisions and their possible effects on inferences drawn from tests of significance - or vice-versa. Journal of the American Statistical Association1959; 54: 30-34.
18.
Greenwald AGConsequences of prejudice against the null hypothesis. PsychologicalBulletin1975; 85: 845-57.
19.
Smith MLPublication bias and meta-analysis. Evaluation in Education1980; 4: 22-24.
20.
Rosenthal R.The 'file-drawer' and tolerance for null results . Psychological Bulletin1979; 86: 638-41.
21.
Cochran WGSome methods for strengthening the common chi-squared tests. Biometrics1954; 10: 417-51.
22.
Mosteller F., Bush R.Selected quantitative techniques. In: Lindzey G (ed). Handbook of social psychology, Volume 1, Theory and method. Cambridge, Mass.: Addison-Wesley , 1954.
23.
Rosenthal R.Interpersonal expectations. In: Rosenthal R, Rosnow RL (eds). Artifact in behavioural research. New York: Academic Press, 1969.
24.
Halperin M.Discussion contribution to Light RJ. Statistics in Medicine1987; 6: 221-28.
25.
Levy P.Substantive significance of significant differences between two groups. PsychologicalBulletin1967; 67: 37-40.
26.
Schmid JE, Koch GG, LaVange L.An overview of statistical issues and methods of meta-analysis. Journal of Biopharmaceutical Statistics1992; 1(1): 103-20.
27.
Mosteller F. , Chalmers TCSome progress and problems in meta-analysis of clinical trials. Statistical Science1992 ; 7: 227-36.
28.
Greenland S. , Salvan A.Bias in the one-step method for pooling study results. Statistics in Medicine1990; 9: 247-52.
29.
Thompson SG , Pocock SJCan meta-analysis be trusted? The Lancet1991; 338: 1127-30.
