[Systematic reviews and meta-analysis in laboratory medicine: principles and methods]

Systematic reviews and meta-analyses can help health-care professionals to base their decisions on the highest levels of evidence, and are therefore the cornerstone in practicing evidence-based medicine. In the field of diagnosis in general, and in laboratory medicine in particular, systematic reviews are scarcer, and often of lower quality than in other areas of medicine.

To summarize the principles and methods that can be advised to perform systematic reviews of good quality in laboratory medicine.

A narrative review of the literature and discussions with members of the Committee on Evidence-Based Laboratory Medicine of the International Federation of Clinical Chemistry and Laboratory Medicine, in particular during our last three meetings, enabled us to adapt to laboratory medicine the methods which are currently being advocated to perform systematic reviews of good quality in other medical areas.

The process of systematic reviewing consists of six key steps: 1) preparation: formulation of question(s); 2) systematic search of the literature; 3) selection of primary studies; 4) critical appraisal of the quality of the selected literature; 5) extraction and synthesis of data; 6) interpretation. The main differences between systematic reviews of diagnostic interventions and those of therapeutic interventions bear on question formulation, the choice of study designs to be included or excluded, the assessment of study quality and the statistical methods used to combine their results. The quoted references will help interested readers to deepen by themselves the most technical or controversial points.

In order to improve quality of care as well as the balance between benefits, harms and costs of practice, each recommendation in guidelines should be based on a systematic review. In addition, such reviews can help to better identify gaps in biomedical knowledge, and improve critical skills, education, and training of professionals. One can therefore reasonably expect that the current multiplication of systematic reviews in laboratory medicine will favor the diffusion of diagnostic technologies with the highest levels of proven efficacy and effectiveness, at the detriment of the others.