What matters for the scalability of prejudice reduction programs and interventions? A Delphi study

BACKGROUND: In many countries, policy makers and practitioners turn to prejudice reduction programs and interventions to tackle prejudice in the community. However, successfully addressing prejudice requires an effective intervention that can scale to match the broad span of the problem. The scalability assessment frameworks from health sciences have varying emphasis on four categories—intervention, delivery, costs, and context. For example, the high-level factors in the two Milat et al. scalability assessments are weighted towards details of the intervention (Milae et al. in Health Promot Int 28(3):285–981, 2013; Health Res Policy Syst 2:1–17, 2020). Investigation into scalability, specific to prejudice reduction, is necessary to understand how scalability frameworks apply in a different discipline. METHODS: Using a Delphi approach—a structured method to obtain consensus from experts (Milae et al. Health Promot Int 28(3):285–981, 2013; Linstone and Turoff in The Delphi method—techniques and applications, Addison-Wesley, 1975; de Meyrick in Health Educ 103(1):7–16, 2003)—to bring together 16 prejudice reduction experts from multiple sectors including NGOs, private, government and academia, we developed a scalability assessment framework of criteria that are important for the successful scaling of prejudice interventions. We then applied that framework to exemplars of prejudice reduction interventions published in the academic literature. RESULTS: For prejudice reduction interventions, contextual factors are key considerations for successful scaling. Commonly used prejudice reduction intervention approaches like contact, whether face-to-face or online, can have limited scalability. CONCLUSIONS: To reduce prejudice there needs to be consideration of scalability. This paper presents a first-of-its-kind framework for assessing scalability for prejudice reduction interventions. Applying the empirically developed framework to actual interventions demonstrated that for interventions to be effective and scalable, a greater focus on approaches beyond face-to-face contact is required.