Translation and validation of an Urdu version of the Vaccine Attitudes Examination (VAX) scale

Abstract

Abstract Introduction Despite being one of the most effective methods of preventing the spread of communicable diseases1, vaccination has been subjected to criticism. Several scales have been formulated that were used to document attitudes towards vaccination. However, they either focus on parental attitudes towards children’s vaccinations or are targeted towards an individual’s attitude towards a specific vaccine. A scale that could document an individual’s attitude about vaccinations in general was developed by Martin and Petrie2. Aim The study aimed to translate and validate an Urdu version of the Vaccine Attitude Examination (VAX) scale in a selected Pakistani sample. Methods A cross sectional study was conducted for a month (Sept. 2021) in an outpatient department of a healthcare facility in Karachi, Pakistan. All adult men and women, as well as their caregivers (≥ 18 years), who visited the clinic, spoke Urdu as their first language, and were eligible for COVID-19 vaccine were invited. Convenience sampling method was selected, and sample size calculation was based on an item response ratio of 1:20. An Urdu version of the VAX scale was developed. Data were analysed using IBM SPSS version 23. The reliability was assessed using Cronbach’s alpha (α) and intraclass correlation coefficient (ICC) was calculated. The model fitness was evaluated through IBM AMOS version 25. It done carried out using confirmatory factor analysis (CFA) and reporting of fit indices, namely goodness of fit index (GFI), adjusted goodness of fit index (AGFI), Tucker Lewis index (TLI), comparative fit index (CFI), and root mean square error of approximation (RMSEA). The scale was considered as reliable if Cronbach’s (α) was >0.7. Model fitness was considered acceptable if most fit indices were in acceptable ranges3, i.e., GFI, TLI, CFI > 0.9 and RMSEA < 0.6, as well as p value < 0.05. This study was granted ethical clearance by the Institutional Review Board of Tabba Heart Institute, Karachi, Pakistan (Reference# THI/IRB/FQ/22-09-21/017). Results A total of 211 responses were analysed. The reliability of the scale, i.e., Cronbach’s alpha value was 0.821, and remained the same, i.e., > 0.8 during item deletion. CFA using a 4-factor model revealed the following values for fit indices; GFI = 0.944, TLI = 0.953, CFI = 0.966, and RMSEA = 0.051 (p value = 0.008). Discussion/Conclusion Although a very small scale study, the findings suggest that the Urdu version of VAX scale is reliable and a valid instrument for use in an Urdu speaking population. The scale showed good internal consistency and acceptable model fitness. The use of convenience sampling and being conducted at a single site may affect the generalisability of the results. However, this was the first time the VAX Urdu version was made available for an Urdu speaking population. It is hoped that the attitudes toward vaccines that are measured by the Urdu VAX will provide valuable information as public health officials and others work to improve vaccine uptake. Availability of the Urdu version will enable clinicians to assess the Urdu speaking persons’ attitude towards general vaccination including COVID-19 vaccination.