The treatment options for the irritable bowel syndrome (IBS) are expanding as new therapies, including probiotics and serotonin receptor agents, become available. Before any new agents gain widespread use, they must be studied in appropriately designed clinical trials. Symptom improvement remains the key clinically but the best technique to measure symptom improvement is unclear. Many IBS therapy studies have used a binary endpoint such as "Have you had satisfactory relief of your IBS symptoms in the past week? Yes/No?" The study by Whitehead and colleagues in this issue suggests that "satisfactory relief" is affected by baseline symptom severity and may not always truly reflect the symptom burden. Future research needs to determine whether "satisfactory relief" is truly adequate, or whether alternatives such as the proportion of patients achieving a > or = 50% reduction in symptom severity would represent a superior approach to capture clinically important improvement.