Meta-analysis: Formulating, evaluating, combining, and reporting

Meta-analysis involves combining summary information from related but indep endent studies. The objectives of a meta-analysis include increasing power to detect an overall treatment effect, estimation of the degree of benefit associated with a particular study treatment, assessment of the amount of v ariability between studies, or identification of study characteristics asso ciated with particularly effective treatments. This article presents a tuto rial on meta-analysis intended for anyone with a mathematical statistics ba ckground. Search strategies and review methods of the literature are discus sed. Emphasis is focused on analytic methods for estimation of the paramete rs of interest. Three modes of inference are discussed: maximum likelihood; restricted maximum likelihood, and Bayesian. Finally, software for perform ing inference using restricted maximum likelihood and fully Bayesian method s are demonstrated. Methods are illustrated using two examples: an evaluati on of mortality from prophylactic use of lidocaine after a heart attack, an d a comparison of length of hospital stay for stroke patients under two dif ferent management protocols. Copyright (C) 1999 John Wiley & Sons, Ltd.