Development and Psychometric Properties of the Caregiver Burden Scale in Polish Caregivers of Stroke Patients

Purpose: Most stroke patients require long-term care of their family members. Excessive care burden entails several negative consequences; therefore, the severity of the burden should be periodically assessed. For this purpose, valid multidimensional measures are needed. Our study, which is a part of a larger research project, aims to translate and evaluate the psychometric properties of the Polish Caregiver Burden Scale (CBS) in relation to construct validity and internal consistency in caregivers of the patients after stroke. Patients and Methods: The sample of this cross-sectional observational study consisted of 366 informal caregivers to consecutive first-ever stroke survivors. The five-factor Polish CBS and the Hospital Anxiety and Depression Scale (HADS) were administered during the home visits at three to six months after patients' hospitalisation. Exploratory (EFA), confirmatory (CFA) factor analyses and a net analysis were performed to investigate the internal structure and a factorial construct validity of the CBS. Correlation analyses between the CBS and the HADS were carried out to examine convergent validity. Cronbach's alpha and item-total correlation were applied to assess internal consistency. Results: Three out of five factors identified by EFA were similar to the original indices of the CBS, while the remaining two deviated from the original structure of the CBS. The CFA five-factor model represented an acceptable fit (confirmatory fit index, CFI = 0.96, root mean square error, RMSEA = 0.04) but only after a modification. All subscale scores of the CBS were positively correlated with the HADS, supporting the convergent validity. Cronbach's alpha coefficients for the overall scale (0.92) and all subscales (0.72-0.87) except one (0.69) and item-total correlation results indicated good internal consistency. Conclusion: The Polish version of the CBS showed acceptable internal consistency and good convergent validity. Factorial validity and structural integrity were partially supported. The interrelationships between the CBS subdomains, their partial mutual contamination, and the scale's non-orthogonal structure should be considered when interpreting the results of further studies using this version of the scale.