In addition to statistical significance, it is important to evaluate the magnitude of differences in health-related quality of life over time. Interest in establishing the minimal difference that is clinically important or the minimally important difference has burgeoned over the last few years. This review summarizes some of the leading approaches to estimating the minimally important difference, offers caveats on the minimally important difference estimation based on existing literature and provides recommendations for future work. The authors recommend using multiple anchors to estimate the minimally important difference, using only anchors that correspond to minimal change in health-related quality of life, reporting information about the variation around the estimates, and providing bounded estimates to reflect the uncertainty.