Objective: To assess the agreement between patients and relatives concerning the formers' quality of life (QOL) before intensive care unit (ICU) admission.
Design: Prospective study involving direct interviews of patients and relatives during ICU stay.
Setting: Two four-bed surgical-medical ICUs in a 960-bed teaching hospital.
Patients and methods: A hundred seventy-two adult, co-operative patients consecutively admitted to ICU for more than 24 h, and their relatives were interviewed. The instruments used were two questionnaires suitable for ICU patients: QOL-IT and QOL-SP. Interobserver reproducibility was investigated in 36 patients.
Results: Interobserver reproducibility was nearly perfect (weighted Kappa 0.99 for QOL-IT and QOL-SP). Considering global scores, weighted Kappa was 0.78 for QOL-IT and 0.82 for QOL-SP, with the mean difference between patients and relatives lower than 0.3 for both scores but with limits of agreement wider than 4. Among the items, concordance was excellent in the areas of physical activity and social life for both questionnaires. Gender, living together with the patient and the degree of relationship of relatives did not influence the agreement.
Conclusions: The relatives give global scores for both instruments which can be regarded as acceptable substitutes for those given by patients. However, the wide limits of agreement should make investigators cautious in analysing together scores generated by patients and by relatives. The emotional dimension seems to be assessed less accurately by relatives than the physical one.