Introduction: A 'meta-analysis' is a multi-step statistical analysis of the results from independent studies, the goal to produce a single estimate of a treatment effect. Meta-analyses are becoming increasingly popular in the neurosurgical literature. AMSTAR is an 11 point scale used to assess methodologic quality. PRISMA is a 27-item checklist used to ensure the transparent and complete reporting of meta-analyses.
Methods: All meta-analyses ever published in the two main neurosurgical journals were retrieved. We first determined whether the term 'meta-anlaysis' was appropriate based on our simple definition, as described above. We then applied the Assessment of Multiple Systematic Reviews (AMSTAR) and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) to each metanalysis. Each metanalysis was then assigned a 'score' for each checklist, which was the number of components present within each paper for each checklist, expressed as a percentage.
Results: Sixty-three meta-analysis were reviewed, 39 from the Journal of Neurosurgery and 24 from Neurosurgery, from 1990 to 2011. Nineteen (30%) of the articles should NOT be considered a meta-analysis. The average AMSTAR score was 31.2% (range, 0-82%); the average PRISMA score was 52.6% (range, 3.7-92.6%). The AMSTAR and PRISMA scores showed improvement with time (5-year intervals from 1990 to 2010+), if one of the authors was a person who potentially had meta-analysis expertise (ex.biostatistician or epidemiologist), if it met our definition of a meta-analysis and if the authors solely focussed on conducting a meta-analsysis (as opposed to first reviewing their own institution's data and then secondarily performing a meta-analysis).
Conclusions: The overall methodology and reporting of meta-analyses in the neurosurgical literature is poor but is improving. One-third of papers inappropriately use the term "meta-analysis". Academic neurosurgery needs to hold these papers to high standards, which can be done by applying these checklists.
Patient Care: By advocating better quality meta-analyses which is a cornerstone of Evidence-Based Medicine.
Learning Objectives: By the conclusion of this session, participants should be able to: 1) Define a "meta-analysis" 2) Discuss, in small groups, the key methodologic and reporting components of a meta-analysis 3) Identify effective methods to improve methodology and reporting of meta-anslyses.