ISO rates could be viewed as an indicator of the quality of healthcare facilities, enabling patients to assess their performance when they need to use them. It is therefore important to examine the validity of ISO rates derived from current monitoring programs. This article summarizes the main conclusions of a report published in June 2003 by the National Agency for Health Evaluation and Accreditation (ANAES) on the limitations of interpreting ISO rates derived from surveillance. ISO surveillance, when part of a comprehensive prevention strategy, is useful for improving the quality of care because its goal is to reduce the number of ISOs. However, the heterogeneity and subjectivity of ISO diagnostic criteria, the sensitivity of data collection—which is highly dependent on the ability to follow up on patients after their discharge from the hospital—as well as the number of risk factors (related to patients and surgical procedures in particular) that must be taken into account to allow for comparisons, limit the use of these data for comparisons between departments or healthcare facilities. (R.A.)