Objectives: To assess six short-form versions of Zarit Burden Interview (ZBI-12, ZBI-8, ZBI-7, ZBI-6, ZBI-4, and ZBI-1) among three caregiving populations.
Study design and setting: Secondary analysis of carers' surveys in advanced cancer (n=105), dementia (n=131), and acquired brain injury (n=215). All completed demographic information and the ZBI-22 were used. Validity was assessed by Spearman correlations and internal consistency using Cronbach's alpha. Overall discrimination ability was evaluated using the area under the receiver operating characteristic curve (AUC).
Results: All short-form versions, except the ZBI-1 in advanced cancer (rho=0.63), displayed good correlations (rho=0.74-0.97) with the ZBI-22. Cronbach's alphas suggested high internal consistency (range: 0.69-0.89) even for the ZBI-4. Discriminative ability was good for all short forms (AUC range: 0.90-0.99); the best AUC was for ZBI-12 (0.99; 95% confidence interval [CI]: 0.98-0.99) and the second best for ZBI-7 (0.98; 95% CI: 0.96-0.98) and ZBI-6 (0.98; 95% CI: 0.97-0.99).
Conclusions: All six short-form ZBI have very good validity, internal consistency, and discriminative ability. ZBI-12 is endorsed as the best short-form version; ZBI-7 and ZBI-6 show almost equal properties and are suitable when a fewer-question version is needed. ZBI-4 and ZBI-1 are suitable for screening, but ZBI-1 may be less valid in cancer.