Validity and reliability of the Thai version of the barthel index for elderly patients with femoral neck fracture

Background: Hip fracture is a common osteoporotic fracture that requires a specific outcome measurement tool to evaluate functional status. The authors aimed to develop the Thai version of the Barthel Index (TVBI) and to evaluate the validity and reliability of TVBI for assessment of elderly patients with femoral neck fracture. Material and Method: The Barthel Index (BI) was translated into Thai using a forward-backward translation protocol. Fifty-three patients with low-energy femoral neck fracture were then prospectively enrolled and evaluated with TVBI, the de Morton Mobility Index (DEMMI), the EuroQol-5D (EQ-5D), the two-minute walk test (2MWT), and the timed get-upand- go test (TUG) within 2 weeks after surgery. Validity of TVBI was assessed by calculating the index of item-objective congruence and correlating TVBI scores with scores from other outcome measurements. TVBI reliability was evaluated by measuring test-retest reliability and internal consistency. Results: TVBI had high content validity and strong correlation with DEMMI (Spearman’s rho = 0.629; p-value <0.001), and moderate correlation with EQ-5D utility score, EQ-5D visual analog scale, and 2MWT (Spearman’s rho = 0.452, 0.313, and 0.413, respectively; p-value <0.05). Interobserver and intraobserver reliabilities of TVBI were high, with intraclass correlation coefficients of 0.714 and 0.968, respectively. The internal consistency of TVBI was acceptable (Cronbach’s alpha = 0.694). Conclusion: TVBI yields good validity and reliability, is without floor or ceiling effects, and can be used in all patients during early postoperative treatment after femoral neck fracture.