The assessment of efficacy in clinical trials in Alzheimer's disease has been a controversial area. The US Food and Drug Administration has not accepted mental status assessment as a sole determinant of efficacy because of concerns about oversensitivity and lack of direct linkage to clinical importance. Global change instruments offer a number of advantages as primary outcome measures, chief among which is a wide range of responsiveness, a property well suited for trials of 1 to 2 years. Initial investigations with global change instruments have shown encouraging validity. Reliability has been only moderate, suggesting that the methodology needs further attention. A standardized format for interview and the acceptance of a set of anchor points that better define the ratings of change are two modifications that should improve reliability. Global change instruments, if properly explained to the patients, caregivers and primary physicians, will capture the concept of clinical importance.