Structural barriers to knowledge co-production in the context of public health crisis

Knowledge translation (KT) is widely recognized as essential for evidence-informed policymaking, yet the research-policy gap persists. Co-production, namely collaborative knowledge production among researchers, policymakers, and practitioners, has emerged as a promising approach to bridge this gap. However, existing literature has focused largely on articulating co-production's principles and anticipated benefits, while empirical investigations into why it fails in practice remain limited. We conducted a qualitative study using semi-structured in-depth interviews with 15 participants, including university-based researchers (n = 10) and government-funded institute researchers (n = 5) with direct experience in Korea's COVID-19 research response. Data were analyzed using framework analysis following Spencer's five-stage approach, informed by the Knowledge-to-Action model across four KT stages: knowledge production, exchange, utilization, and conversion to knowledge production. Barriers were identified across all four KT stages, reflecting five interrelated structural mechanisms: epistemological incommensurability and incomplete boundary work; institutional fragmentation and misaligned incentives; the dominance of biomedical epistemology and the systematic exclusion of uncomfortable knowledge; bureaucratic expert dependency and path-dependent decision-making; and the absence of equity and power redistribution. These mechanisms formed a self-reinforcing cycle that systematically foreclosed the preconditions for co-production, not as isolated failures but as structurally produced outcomes. Barriers to co-production are structural rather than communicative in nature. Addressing them requires concrete institutional responses: restructuring research funding to mandate interdisciplinary participation, reforming advisory committee composition and deliberative processes to enable substantive non-biomedical input, and establishing permanent knowledge brokering infrastructure rather than project-based arrangements. Co-production will remain aspirational without systemic interventions that realign institutional incentives, redistribute epistemic authority, and create sustained platforms for researcher-policymaker-public collaboration.