Does an integrated school eye health delivery model perform better than a vertical model in a real-world setting? A non-randomised interventional comparative implementation study in Zanzibar.

Background Few studies on school eye health programmes have shown they were cost-effective. We compared the performance (Reach, Effectiveness, Adoption, Implementation and Maintenance (RE-AIM)) between an integrated model (IM) and a vertical model (VM) of school eye health delivery in Zanzibar. Methods The set of RE-AIM performance indicators of the IM (n=9) and VM (n=10) cohorts was compared. The VM implemented only the eye health interventions, while the IM had the eye health interventions conducted within the school feeding programme. Semistructured interviews were conducted with 36 stakeholders to understand the challenges and outcomes experienced when implementing both models. Results The IM achieved higher screening coverage, voluntary follow-up rate, screening validity and spectacle compliance than VM. This was due to effective coordination between implementers, motivated teachers to prevent vision problems and related negative impacts in children, and activities implemented timeously post-training. Both models recorded low wearing compliance. All schools in the IM cohort completed screening activities, but two schools in the VM cohort did not. Both models ceased activities after the funding stopped. Local stakeholders emphasised that evidence from this study can be used to advocate for more resources for children's eye health. Conclusions The IM cohort achieved better reach, effectiveness, adoption rate and implementation performance than the VM cohort. The poor maintenance performance indicators in both arms postfunding call for improvement to the implementation strategy to ensure the sustainability of school eye health. In the optics of scaling up, an integrated approach is recommended. [ABSTRACT FROM AUTHOR]