28. Systematic review of psychometric properties of health instruments measuring carer impact (177)

Michele Peters, Research Officer, Department of Public Health and Primary Care, University of Oxford, Oxford, United Kingdom
Co authors: Elizabeth Gibbons & Raymond Fitzpatrick
michele.peters@dphpc.ox.ac.uk

Abstract:

Carers play an important role in the care of chronically ill patients. The impact on health of caring for someone is increasingly recognized. There are indirect (generic) and direct measures of carer burden. The latter are specifically developed for carers. Forty-four articles were identified from the Patient-reported Health Instruments database containing over 12,000 articles. Twenty-six of these fulfilled the inclusion criteria. Another 49 articles were identified by supplementary searches. Psychometric properties according to Fitzpatrick et al (1998) were assessed. Information on psychometric properties was available for 5 indirect measures (15 articles) and 7 direct measures (57 articles). For indirect measures, only the SF-36 and the General Health Questionnaire had been evaluated in more than one article. For direct measures, 19 instruments had been evaluated in a single study only. The most widely evaluated direct measures were the Zarit Burden Interview (ZBI) (12 articles) and the Caregiver Strain Index (CSI) (9 articles), but many of these evaluations were carried out on different versions of the two instruments. Most articles focused on reliability and validity, with little, if any, information on responsiveness, precision, acceptability and feasibility. Due to the low number of evaluations on indirect measures, it is not possible to recommend a generic instrument for use in this population group. However, since generic instruments have been widely evaluated in other population groups, it is likely that they are also useful in carers. Most direct measures can also not be recommended for wide use, due to the limited psychometric information. The ZBI and CSI appear the most promising, with the CSI having the small but important advantage of having been shown to be feasible as a self-completion questionnaire. Overall, due to the limited psychometric information, further research is needed to confirm which indirect and direct measures performs best in the carer population.