I thought I saw a MOOSE: Guidelines for Meta-analysis and Observation.

There is a general interest among the public, study volunteers, researchers and private and public funding organizations for well-done data gathering, analysis and reporting. In response to the modern emphasis on quality work, groups have developed guidelines to provide a framework for data reporting and analysis. This is especially useful when the analysis attempts to characterize disparate historical data sets. Because there is inherent heterogeneity among research subjects and datasets, a common problem is comparing apples with oranges. The MOOSE guidelines exist to help address this issue.

What are the MOOSE guidelines?

The Meta-analyses of Observation Studies in Epidemiology (MOOSE) guidelines were among the first collections of standardized suggestions advanced to address the issue of research data reporting. Briefly, the MOOSE guidelines focus on the routine parts of a science manuscript. These include:


Each section has several points, each crafted to that part and especially appropriate to analysis of prior published data. While centered on use of epidemiologic data, the guidelines would be germane to basic science studies.

Structure of the MOOSE guidelines

There are 35 separate items in the MOOSE guidelines arranged as a short, section by section, check list. The list prompts the researcher to confirm or not confirm that their study addresses each action item and includes a place to indicate on what page of the original manuscript the item is covered. Completing the checklist is easy and one can use it to identify areas for improving the manuscript. Additionally, MOOSE guidelines have implications for journal editors and reviewers of manuscripts. They are a metric for testing the mettle of the study and for making suggestions to improve the work prior to publication.

Implementation and promotion of MOOSE

However, the success of the MOOSE enterprise relies upon the diligence and sincerity of the researchers. Since following the guidelines is only suggested, the ability to enforce these parameters can be limited. A larger issue is revealed here, namely, how to standardize the data reporting in the source manuscripts. If one begins a meta-analysis with flawed primary papers, one’s efforts will not produce a reliable finding. The academic journals themselves bear heavily in this. They are the final gate keepers of science and carry a grave responsibility to insist upon quality of the highest magnitude in the manuscripts they publish. So, at every level the key to moving forward is unceasing effort to do critical, careful, and unbiased research.