Development and Validation of the Vaccine Barriers Assessment Tool for Identifying Drivers of Under-vaccination in Children under Five Years in Australia

"Understanding the changing social and behavioral drivers of vaccination is key to developing targeted, cost-effective vaccination strategies to improve and sustain uptake."

Data on routine childhood vaccination coverage can only tell us who is under-vaccinated; it cannot explain why vaccine coverage is low. Collecting data on the reasons behind under-vaccination is necessary to address key barriers and target interventions appropriately. In 2022, the World Health Organization (WHO) Behavioural and Social Drivers [BeSD] of Vaccination working group developed a flexible set of quantitative and qualitative instruments to help countries measure acceptance as well as access issues, mostly targeted to low- and middle-income country settings. However, these instruments have not been validated against behaviour or vaccine uptake in Australia. This study aims to develop and validate the Vaccine Barriers Assessment Tool (VBAT) for Australia. 

This paper presents the three phases of instrument development and validation, which took place between 2019 and 2023. Beginning with a broad item set derived from both behavioural theory and published research evidence, the goal was to refine and reduce the items to reach a final long (VBAT-LF) and short form (VBAT-SF) of the instrument. In all phases, insights generated through relevant qualitative or quantitative methods were considered and contextualised by the core investigator team, comprising eight experts; the project advisory group, which included public health policymakers and researchers, also provided guidance.

In brief, in Phase 1, the team developed a list of 80 items reflecting all potential parental barriers to childhood vaccination, derived from published literature and behavioural theory. Through cognitive interviews (n = 28), they refined this list to 45 items. In Phase 2, they conducted a two-wave online survey to test the reliability and validity of these items in an Australian sample of parents (n = 532) with structural equation modeling, further refining the list to 35 items. In Phase 3, they conducted a final parent survey (n = 156), administering these items along with the Parent Attitudes toward Childhood Vaccination (PACV) scale for comparison. The team then reviewed participants' immunisation register data to assess the predictive validity of the proposed models.

The final 6-item short form and 15-item long form VBAT assess access, communal benefit, personal risk, equity, commitment, social norms, and trust in healthcare workers. Per the researchers, the breadth of factors assessed in the VBAT-LF makes it a potentially suitable evaluation tool for multicomponent interventions that involve both vaccine communication and service delivery elements, which may have different mechanisms of effect. Where social norms are considered particularly relevant, the team advises users to apply the VBAT-LF. Both forms include an intention item that can be used as a proxy for vaccine uptake if register data are not available.

According to the researchers, in this study population, both forms of the VBAT exhibited an acceptable level of predictive validity, both in absolute and in relative terms. In particular, both forms of the VBAT predicted vaccine uptake to a greater degree than the gold standard PACV tool, which measures vaccine hesitancy. Among the range of available measurement instruments, only the WHO BeSD quantitative data collection instrument also captures access barriers or practical issues. The VBAT domains align generally with the overarching BeSD categories, with some differences between the specific constructs and items. For example, the VBAT domains of "communal benefit" and "personal risk" are similar to BeSD constructs within the category of "what people think and feel". The VBAT is not intended to replace the BeSD tool but is instead tailored specifically for Australia. 

The VBAT is being applied for national surveillance in Australia and will be adapted for additional populations and vaccines. For example, the team is working with Aboriginal and Torres Strait Islander researchers and communities to create a culturally tailored version appropriate for this population, which may include alternative modes of administration (e.g., through individual or group discussions). The researchers conclude that the VBAT "will be a valuable tool for serial surveillance at a national and regional level to inform vaccine policy and programs and as a measure for evaluating interventions."