Answers are grounded in this paper and cite their sources.
Food security is a public health priority, but publication output does not directly measure research implementation or societal benefit. This study examined contemporaneous ecological associations between annual food security related publication output and aggregated agrifood-system indicators across World Bank income groups. We integrated a corpus of 61,158 Scopus indexed publications with 24 productive, economic, resource use, and workforce indicators within an income-group-year analytical framework covering 2000–2025. Analyses comprised exploratory income-stratified bivariate regressions, mixed-effects models including centered calendar year and income-group random effects, and exploratory publication-moderator interaction models. Multiplicity was addressed using Holm adjustment for the stratified analyses and Benjamini-Hochberg false discovery rate (FDR) adjustment for the mixed-effects and interaction analyses. Of 96 income-stratified associations, 66 had nominal p values below 0.05, but only 11 remained below the Holm-adjusted threshold, indicating that much of the initial statistical evidence was not retained after multiplicity correction. Seven of 24 mixed-effects models had FDR-adjusted p values below 0.05, involving beef, chicken and pig meat production, aquaculture, fish and seafood production, agricultural output, and agriculture’s share of gross domestic product. Associations otherwise differed in direction and statistical support across productive, resource use, economic, and workforce indicators and across income settings. Moderator screening identified 294 FDR-adjusted interactions among 989 outcome-moderator combinations, showing that several publication-outcome relationships were attenuated, amplified, or changed direction across contextual conditions. The relationship between food security-related scientific production and agrifood-system conditions was selective rather than uniform. Ecological correspondence persisted for a limited set of productive and structural indicators, while many other associations weakened after multiplicity adjustment or differed across income and contextual settings. These patterns suggest that scientific activity cannot be interpreted independently of the outcome being examined or the environment in which knowledge is produced and potentially used.
No takes yet. Share an insight, caveat, or question.
Vergara-Rios et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: