Abstract
Background: Psychometric properties of the 38-item version of the Dutch Objective Burden Inventory (DOBI) have been evaluated primarily in older female caregivers, with no published studies that vary by a different gender and age distribution.
Objective: The aim of this study was to test the construct validity and reliability of the DOBI in caregivers of individuals with heart failure from the United States and other countries.
Methods: This secondary analysis from a cross-sectional study used an online survey. Factorial validity was tested with confirmatory factor analysis, item performance was examined with ordinal item response analysis, and convergent validity was tested correlating DOBI subscale scores with Bakas Caregiving Outcomes Scale scores. Internal consistency reliability was assessed with Cronbach [alpha].
Results: Item response analysis removed 14 items. Confirmatory factor analysis retained the original 4-factor solution. Original and reduced instruments demonstrated good internal consistency. Validity was supported by meaningful associations with the Bakas Caregiving Outcomes Scale.
Conclusions: Both the original 38-item DOBI and the reduced 24-item DOBI had support for their construct validity and internal consistency. The 24-item DOBI may be a useful alternative to the 38-item version, because it maintains psychometric properties of the original instrument while decreasing data collection burden. However, more research is needed to assess whether the shorter version is useful in assessing objective caregiving burden.