Assessing the potential for scaling evidence-based interventions in African health systems: A deliberate dialogue

Background: The WHO Regional Office for Africa reviewed evidence-based interventions designed to enhance health systems outputs: access, quality, demand and resilience. Although there is eagerness to expand interventions, the extent to which they can be successfully scaled remains mostly unknown. This study evaluated their potential for scaling to enhance system outputs. Methods: Underpinned by two frameworks for knowledge transfer, the study utilized a deliberative dialogue approach and an integrated knowledge translation strategy. A steering committee of 8 men and women oversaw the process, and 25 experts from across Africa were invited to participate. Data collection employed the African Scalability Assessment Framework (AFROSAF), consisting of 15 attributes grouped into nine scalability components: health need (three attributes), development process (three), intervention content (one), political context (one), evidence for impact (two), resource availability (one), target unit (one), scaling setting (one) and sustainability at scale (one). The scoring was on a 4-point Likert scale. Intervention scores ranged from 0 to 100, where higher scores indicated a greater potential for successful scaling. Data analysis included frequency counts, arithmetic mean, standard deviation (SD), 95% confidence interval (CI) and hierarchical cluster analysis. Results: A total of 288 health services interventions were rated for scalability, with most focusing on disease prevention (n = 114; 39.6%) and newborn age group (n = 70; 24.3%). The scalability scores averaged 79.2 (SD 17.5). The highest component score was evidence for impact [mean (CI 95%) 87.9 (85.7, 90.2)] and lowest for health security [61.9 (58.0, 65.8)]. The scalability scores varied across public health functions and age cohorts. Interventions were clustered into three groups based on their scalability score: high (n = 185; 64.2%; mean 89.5; SD 6.1; range 78.7–100), medium (n = 77; 26.7%; mean 68.6; SD 5.3; range 58.3–76.9) and low (n = 26; 9.0%; mean 37.4; SD 14.2; range 3.7–55.6). Most high-scoring interventions focused on disease prevention, 78 (42.2%) and reproductive and newborn age groups 59 (31.9%). Conclusions: This study rated public health interventions for their scalability in African health systems. Disease prevention interventions for pregnancy women and newborns were most likely to be scaled. However, health security remained largely unexplored, and further investigation remains pivotal.