Landscaping the structures of GAVI country vaccine supply chains and testing the effects of radical redesign. Issue 36 (26th August 2015)

• Record Type: Journal Article
• Authors: Lee, Bruce Y.; Connor, Diana L.; Wateska, Angela R.; Norman, Bryan A.; Rajgopal, Jayant; Cakouros, Brigid E.; Chen, Sheng-I.; Claypool, Erin G.; Haidari, Leila A.; Karir, Veena; Leonard, Jim; Mueller, Leslie E.; Paul, Proma; Schmitz, Michelle M.; Welling, Joel S.; Weng, Yu-Ting; Brown, Shawn T.
• Is Part Of: Vaccine. Volume 33:Issue 36(2015)
• Page Start: 4451
• Page End: 4458
• Publication Date: 2015-08-26
• Subjects: Supply chains -- Immunization -- Modeling; Vaccines -- Periodicals; 615.372
• DOI: 10.1016/j.vaccine.2015.07.033
• Languages: English
• ISSNs: 0264-410X

Abstract

Highlights: Many GAVI-eligible countries' vaccine supply chains were designed in the 1970s. Supply chain design does not correlate closely with country characteristics. We used HERMES to model current and radically simpler designs in sample countries. Simpler designs reduce operating costs and warrant consideration in many countries. Abstract: Background: Many of the world's vaccine supply chains do not adequately provide vaccines, prompting several questions: how are vaccine supply chains currently structured, are these structures closely tailored to individual countries, and should these supply chains be radically redesigned? Methods: We segmented the 57 GAVI-eligible countries' vaccine supply chains based on their structure/morphology, analyzed whether these segments correlated with differences in country characteristics, and then utilized HERMES to develop a detailed simulation model of three sample countries' supply chains and explore the cost and impact of various alternative structures. Results: The majority of supply chains (34 of 57) consist of four levels, despite serving a wide diversity of geographical areas and population sizes. These four-level supply chains loosely fall into three clusters [(1) 18 countries relatively more bottom-heavy, i.e., many more storage locations lower in the supply chain, (2) seven with relatively more storage locations in both top and lower levels, and (3) nine comparatively more top-heavy] which do not correlate closely with any of the country characteristics considered. For all three cluster types, our HERMES modeling found that simplified systems (a central location shipping directly to immunization locations with a limited number of Hubs in between) resulted in lower operating costs. Conclusion: A standard four-tier design template may have been followed for most countries and raises the possibility that simpler and more tailored designs may be warranted.