Meta‐analysis of continuous outcome data from individual patients

Meta-analyses using individual patient data are becoming increasingly common and have several advantages over meta-analyses of summary statistics. We explore the use of multilevel or hierarchical models for the meta-analysis of continuous individual patient outcome data from clinical trials. A general framework is developed which encompasses traditional meta-analysis, as well as meta-regression and the inclusion of patient-level covariates for investigation of heterogeneity. Unexplained variation in treatment differences between trials is considered as random. We focus on models with fixed trial effects, although an extension to a random effect for trial is described. The methods are illustrated on an example in Alzheimer's disease in a classical framework using SAS PROC MIXED and MLwiN, and in a Bayesian framework using BUGS. Relative merits of the three software packages for such meta-analyses are discussed, as are the assessment of model assumptions and extensions to incorporate more than two treatments.

[1]  H. Petit,et al.  A double‐blind, placebo‐controlled, enriched population study of tacrine in patients with Alzheimer's disease , 1995, European journal of neurology.

[2]  N. Laird,et al.  Meta-analysis in clinical trials. , 1986, Controlled clinical trials.

[3]  S L Normand,et al.  Meta-analysis: formulating, evaluating, combining, and reporting. , 1999, Statistics in medicine.

[4]  S G Thompson,et al.  A likelihood approach to meta-analysis with random effects. , 1996, Statistics in medicine.

[5]  R. Moore,et al.  Single-patient data meta-analysis of 3453 postoperative patients: oral tramadol versus placebo, codeine and combination analgesics , 1997, Pain.

[6]  D. Spiegelhalter,et al.  Bayesian Analysis of Realistically Complex Models , 1996 .

[7]  M. Kenward,et al.  Small sample inference for fixed effects from restricted maximum likelihood. , 1997, Biometrics.

[8]  H Goldstein,et al.  A multilevel model framework for meta-analysis of clinical trials with binary outcomes. , 2000, Statistics in medicine.

[9]  R. Kempton,et al.  The application of REML in clinical trials. , 1994, Statistics in medicine.

[10]  S. R. Searle.  Linear Models , 1971 .

[11]  D O Stram,et al.  Meta-analysis of published data using a linear mixed-effects model. , 1996, Biometrics.

[12]  I Olkin,et al.  Comparison of meta-analysis versus analysis of variance of individual patient data. , 1998, Biometrics.

[13]  S G Thompson,et al.  Systematic Review: Why sources of heterogeneity in meta-analysis should be investigated , 1994, BMJ.

[14]  C. Morrell,et al.  Likelihood ratio testing of variance components in the linear mixed-effects model using restricted maximum likelihood. , 1998, Biometrics.

[15]  A Whitehead,et al.  Meta‐analysis of ordinal outcomes using individual patient data , 2001, Statistics in medicine.

[16]  J. Carlin.  Meta-analysis for 2 x 2 tables: a Bayesian approach. , 1992, Statistics in medicine.

[17]  S. Sharp,et al.  Explaining heterogeneity in meta-analysis: a comparison of methods. , 1997, Statistics in medicine.

[18]  A Whitehead,et al.  Borrowing strength from external trials in a meta-analysis. , 1996, Statistics in medicine.

[19]  R F Woolson,et al.  A double-blind, placebo-controlled multicenter study of tacrine for Alzheimer's disease. The Tacrine Collaborative Study Group. , 1992, The New England journal of medicine.

[20]  F. Ferris,et al.  The Efficacy of Tolrestat in the Treatment of Diabetic Peripheral Neuropathy: A meta-analysis of individual patient data , 1996, Diabetes Care.

[21]  L. Stewart,et al.  Practical methodology of meta-analyses (overviews) using updated individual patient data. Cochrane Working Group. , 1995, Statistics in medicine.

[22]  Karl J. Friston,et al.  Variance Components , 2003 .

[23]  C S Berkey,et al.  A random-effects regression model for meta-analysis. , 1995, Statistics in medicine.

[24]  L. Stewart,et al.  Systematic Reviews: Obtaining data from randomised controlled trials: how much do we need for reliable and informative meta-analyses? , 1994, BMJ.

[25]  S Greenland,et al.  Quantitative methods in the review of epidemiologic literature. , 1987, Epidemiologic reviews.

[26]  A Whitehead,et al.  A general parametric approach to the meta-analysis of randomized clinical trials. , 1991, Statistics in medicine.

[27]  Gastone G. Celesia,et al.  A 30-Week Randomized Controlled Trial of High-Dose Tacrine in Patients With Alzheimer's Disease , 1994 .

[28]  N L Foster,et al.  An enriched-population, double-blind, placebo-controlled, crossover study of tacrine and lecithin in Alzheimer's disease. The Tacrine 970-6 Study Group. , 1996, Dementia.

[29]  M Farlow,et al.  A controlled trial of tacrine in Alzheimer's disease. The Tacrine Study Group. , 1992, JAMA.

[30]  A Whitehead,et al.  Cholinesterase inhibition for Alzheimer disease: a meta-analysis of the tacrine trials. Dementia Trialists' Collaboration. , 1998, JAMA.

[31]  Sue J. Welham,et al.  Likelihood Ratio Tests for Fixed Model Terms using Residual Maximum Likelihood , 1997 .

[32]  Christopher H Schmid,et al.  Summing up evidence: one answer is not always enough , 1998, The Lancet.

[33]  H. Morgenstern.  Uses of ecologic analysis in epidemiologic research. , 1982, American journal of public health.

[34]  P. Solomon,et al.  A 30-week randomized controlled trial of high-dose tacrine in patients with Alzheimer's disease. The Tacrine Study Group. , 1994, JAMA.

[35]  Malcolm Man-Son-Hing,et al.  Meta-analysis of efficacy of quinine for treatment of nocturnal leg cramps in elderly people , 1995, BMJ.

[36]  D J Spiegelhalter,et al.  Bayesian approaches to random-effects meta-analysis: a comparative study. , 1995, Statistics in medicine.

[37]  L. Sullivan,et al.  Tutorial in biostatistics. An introduction to hierarchical linear modelling. , 1999, Statistics in medicine.

[38]  Elena Losina,et al.  An introduction to hierarchical linear modelling , 1999 .