Related Experiment Videos
Validation of self-reported transfusion histories in renal dialysis patients
J S Sandhu1, P M Campbell, J K Preiksaitis
1Department of Public Health Sciences, University of Alberta, Edmonton.
The purpose of this analysis was to assess the validity of self-reported transfusion histories in dialysis patients. Using data from a cross-sectional study of a dialysis population being investigated for hepatitis C virus (HCV) infection, the correspondence between self-reported transfusion history and transfusion records was explored. Demographic data and dialysis histories were examined in relation to the accuracy of self-reports. Overall, the questionnaire data and the blood bank records agreed for 89% of participants. The Kappa statistic was 0.72 (95% CI: 0.61, 0.83) indicating an acceptable level of agreement. The effect of non-differential exposure misclassification on the risk estimates for transfusion history as a determinant of HCV infection is demonstrated. Exploring the discrepancies between self-reports and documented transfusion histories underlines the need to communicate clearly medical interventions in chronically ill patients. Additionally, it suggests that studies into transfusion-acquired blood-borne pathogens should use all available information sources to establish exposure.
The purpose of this analysis was to assess the validity of self-reported transfusion histories in dialysis patients. Using data from a cross-sectional study of a dialysis population being investigated for hepatitis C virus (HCV) infection, the correspondence between self-reported transfusion history and transfusion records was explored. Demographic data and dialysis histories were examined in relation to the accuracy of self-reports. Overall, the questionnaire data and the blood bank records agreed for 89% of participants. The Kappa statistic was 0.72 (95% CI: 0.61, 0.83) indicating an acceptable level of agreement. The effect of non-differential exposure misclassification on the risk estimates for transfusion history as a determinant of HCV infection is demonstrated. Exploring the discrepancies between self-reports and documented transfusion histories underlines the need to communicate clearly medical interventions in chronically ill patients. Additionally, it suggests that studies into transfusion-acquired blood-borne pathogens should use all available information sources to establish exposure.