Instruments for assessing the burden of informal caregiving for stroke patients in clinical practice: a comparison of CSI, CRA, SCQ and self-rated burden

Objective: To compare the feasibility, convergent and clinical validity of three commonly used burden scales: Caregiver Strain Index (CSI), Caregiver Reaction Assessment (CRA) and Sense of Competence Questionnaire (SCQ), with a self-developed single question on self-rated burden (SRB). Subjects: Stroke patients receiving support from an informal caregiver ( n=148) and their caregivers were followed up to six months after stroke. Intervention: Feasibility was assessed with several measures of missing values. Convergent validity was assessed on the basis of the correlation patterns between the burden scales, and clinical validity through evaluation of expected associations between levels of burden and explanatory patients' and caregivers' characteristics. Results: Missing values were less often observed on CSI and SRB than SCQ and CRA. Significant correlation coefficients ( p<0.05) could be demonstrated between all burden scales, except for one subscale of CRA. Evidence for clinical validity was strongest for CSI and SRB, based on associations between higher burden scores and patients' disability, and patients' and caregivers' poor level of health-related quality of life (all p<0.05). Conclusions: A concise and simple measure would facilitate early detection of caregivers at risk in clinical practice and research. CSI and SRB are more feasible and at least as valid instruments for assessment of caregiver burden in stroke than the longer and more complex SCQ and CRA. SRB could be used for quick screening of caregivers at risk. CSI is indicated for further diagnosis of the burden of informal caregivers.