Discrimination of changes in osteoporosis outcomes.

The purpose of this paper was to identify existing work related to discrimination, responsiveness, and minimal clinically important differences (MCID) for 4 key clinical outcomes in osteoporosis, to serve as a background to discussions about how to define MCID for both individuals and groups. The outcomes assessed were bone density, fractures, quality of life, and function. We conducted a systematic literature search in MEDLINE, EMBASE, and Current Contents for articles that discussed responsiveness, detectable difference, improvement criteria, and clinical importance. We used the Beaton cube to classify the studies depending on whether they compared differences between or changes within individuals or groups. Although a number of studies were identified that presented data on detectable differences beyond error and observed differences, few studies presented data on how to define clinically important differences. A key priority for future research is to define minimally clinical important differences for clinically important osteoporosis outcomes using a consensus based approach that will be accepted by the osteoporosis community at large. Furthermore, these MCID will likely be different for individual patients seen in clinical practice than for individuals in a clinical trial.