Browse the glossary:

• measurement bias (ascertainment bias, detection bias, observer bias)
  — Bias resulting from systematic differences in how outcomes are measured in treatment comparison groups in a study.; … (read more)
• meta-analysis
  — Statistical combination of estimates derived from two or more similar studies, to give an overall effect estimate; … (read more)
• minimal clinically important difference see smallest important difference
  — The smallest treatment effect that the people affected would identify as important; … (read more)
• minimization
  — A method used for allocating participants to treatments in a study, aiming to achieve balance between treatment comparison groups in respect of factors that are expected to affect outcomes; … (read more)
• moderate certainty of the evidence (moderate confidence in the effect estimate, moderate quality of the evidence, moderate strength of the evidence)
  — The research provides a good indication of the likely effect of a treatment. The likelihood that the actual effect of the treatment will not be substantially different is moderate; … (read more)
•   modified intention-to-treat analysis
  — An analysis of study results that excludes some participants; … (read more)
• monitoring
  — Checking and recording specified measures; … (read more)
• multicentre study
  — A study in which several sites (e.g. hospitals or primary care clinics) collaborate; … (read more)
• multiple statistical comparisons
  — The application of many tests of the statistical significance of associations when analysing the results of a study; … (read more)