Patient recall and recall bias of health state and health status

Abstract

The reliability of recall patient reported outcomes, such as health-related quality of life, satisfaction and symptoms, varies substantially. The objectives of this special report are to identify key issues to consider in study design and provide suggestions for minimizing bias in studies including patient reported outcomes. A MEDLINE search identified several areas in which patient recall is subject to bias. Concordance between patient recall and baseline assessments (e.g., prior to an event or medical intervention) for these patient reported outcomes varies depending on the event being recalled, time since the event, and patient clinical and demographic characteristics. Symptom recall tends to be better than recall of health-related quality of life or pain intensity. Specific questionnaire techniques may help minimize the impact of recall bias. Further research is required to determine what factors or patient characteristics predict improved recall and what techniques minimize recall bias.