‘It’s far too complicated’: why fragmentation persists in global health

Abstract

BACKGROUND: Despite many efforts to achieve better coordination, fragmentation is an enduring feature of the global health landscape that undermines the effectiveness of health programmes and threatens the attainment of the health-related Sustainable Development Goals. In this paper we identify and describe the multiple causes of fragmentation in development assistant for health at the global level. The study is of particular relevance since the emergence of new global health problems such as COVID-19 heightens the need for global health actors to work in coordinated ways. Our study is part of the Lancet Commission on Synergies between Universal Health Coverage, Health Security and Health Promotion. METHODS: We used a mixed methods approach. This consisted of a non-systematic literature review of published papers in scientific journals, reports, books and websites. We also carried out twenty semi-structured expert interviews with individuals from bilateral and multilateral organisations, governments and academic and research institutions between April 2019 and December 2019. RESULTS: We identified five distinct yet interconnected sets of factors causing fragmentation: proliferation of global health actors; problems of global leadership; divergent interests; problems of accountability; problems of power relations. We explain why global health actors struggle to harmonise their approaches and priorities, fail to align their work with low- and middle-income countries' needs and why they continue to embrace funding instruments that create fragmentation. CONCLUSIONS: Many global actors are genuinely committed to addressing the problems of fragmentation, despite their complexity and interconnected nature. This paper aims to raise awareness and understanding of the causes of fragmentation and to help guide actors' efforts in addressing the problems and moving to more synergistic approaches.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 116 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.