Our objective was to systematically review clinical studies of incontinence treatments for women that used pad tests to assess outcome, to determine how closely the ICS guidelines had been followed. Our review (Medline 1988-2000, plus referenced studies) identified 75 relevant papers, carrying out pad tests in clinics ( n = 53) or patients' homes ( n = 28). Clinic pad tests lasted between 60 seconds and 2 hours, with inconsistent starting bladder volumes, activities carried out, other test details and presentation of results. Home pad tests lasted between 1.5 and 48 hours: the conduct and reporting of these tests were also variable. Only 25 studies used pad tests that were apparently consistent with ICS guidelines. Pad tests are important in identifying urine loss in clinical evaluations; however, we found wide variations in their conduct and reporting. We recommend that the ICS should review the guidelines, and that further research should develop clinically valid pad tests. Authors and journal editors should ensure that pad test details are fully reported.