tetano
Editor, Senior Moderator
Front Epidemiol
. 2026 Jul 1:6:1813211.
doi: 10.3389/fepid.2026.1813211. eCollection 2026.
Pan-continental spillover risk: integrated spatiotemporal, transmissibility and surveillance analysis of avian influenza a(H5N1) in Africa
Maladho Diaby[SUP] 1 2 3 [/SUP], Alhassane Diallo[SUP] 3 [/SUP], Castro G Hounmenou[SUP] 1 [/SUP], Salifou Talassone Bangoura[SUP] 1 2 3 [/SUP], Emile Faya Bongono[SUP] 1 [/SUP], Kadio Jean-Jacques Olivier Kadio[SUP] 1 [/SUP], Aly Badara Touré[SUP] 3 4 [/SUP], Abdourahamane Barry[SUP] 1 [/SUP], Saidouba Chérif Camara[SUP] 1 [/SUP], Haby Diallo[SUP] 1 [/SUP], Thibaut Armel Chérif Gnimadi[SUP] 1 [/SUP], Sidikiba Sidibé[SUP] 3 [/SUP], Alexendre Delamou[SUP] 3 4 [/SUP], Alioune Camara[SUP] 3 [/SUP], Alpha-Kabinet Kéita[SUP] 1 5 [/SUP], Abdoulaye Touré[SUP] 1 2 3 [/SUP]
Affiliations
Background: The HPAI H5N1 panzootic represents a critical threat to human health in Africa, where traditional poultry systems and dense human-animal interfaces facilitate frequent zoonotic spillover. While sporadic human cases raise pandemic concerns, continent-wide integration of spatial dynamics, transmissibility indicators, and surveillance performance has been lacking. This study quantifies avian influenza transmission over two decades across Africa, identifies geographical hotspots, and evaluates the responsiveness of current surveillance systems.
Methods: We analysed 8,037 avian influenza outbreak events and 369 laboratory-confirmed human cases, predominantly caused by HPAI H5N1 (2004-2025), using harmonised data from FAO (EMPRES-i+), WHO, and WOAH. A Bayesian Besag-York-Mollié (BYM) spatiotemporal model estimated residual transmission risks and Incidence Rate Ratios (IRR) by subtype. The basic reproduction number (R₀) was derived via an exponential growth model applied to human outbreak phases across infectious durations of 7-30 days. Surveillance responsiveness was assessed by quantifying notification delays between clinical observation and official reporting.
Results: Risk of infection in animals: HPAI H5N1 was the dominant strain, representing 87.8% of animal cases, with Egypt acting as the primary epidemiological epicentre (66% of total records). The spatiotemporal model revealed that H5N1 is associated with a significantly higher risk of animal infection (IRR = 8.37; 95% CI: 6.65-10.53). Although 71% of outbreaks were reported within 5 days of detection, significant delays (≥15 days) occurred in 12% of cases, with notable regional disparities. Risk of infection in human: H5N1 was associated with a 67-fold increase in the incidence of human cases compared to other subtypes (IRR = 66.78; 95% CI: 25.29-176.37). Sensitivity analyses yielded R[SUB]0[/SUB] estimates ranging from 1.05 (95% CI: 0.91-1.31) to 1.23 (95% CI: 0.60-2.33), indicating localised epidemic potential.
Conclusion: Our findings highlight a persistent and geographically heterogeneous H5N1 reservoir in Africa with high zoonotic affinity. Although sustained human-to-human transmission remains limited, the identification of dual poultry-human hotspots and localised R0 peaks underscores the urgent need for geographically targeted One Health interventions. Strengthening real-time reporting systems and improving biosecurity in high-risk poultry value chains are critical to mitigating future pandemic threats on the continent.
Keywords: Africa; H5N1; avian influenza; bayesian modelling; one health; public health surveillance.
. 2026 Jul 1:6:1813211.
doi: 10.3389/fepid.2026.1813211. eCollection 2026.
Pan-continental spillover risk: integrated spatiotemporal, transmissibility and surveillance analysis of avian influenza a(H5N1) in Africa
Maladho Diaby[SUP] 1 2 3 [/SUP], Alhassane Diallo[SUP] 3 [/SUP], Castro G Hounmenou[SUP] 1 [/SUP], Salifou Talassone Bangoura[SUP] 1 2 3 [/SUP], Emile Faya Bongono[SUP] 1 [/SUP], Kadio Jean-Jacques Olivier Kadio[SUP] 1 [/SUP], Aly Badara Touré[SUP] 3 4 [/SUP], Abdourahamane Barry[SUP] 1 [/SUP], Saidouba Chérif Camara[SUP] 1 [/SUP], Haby Diallo[SUP] 1 [/SUP], Thibaut Armel Chérif Gnimadi[SUP] 1 [/SUP], Sidikiba Sidibé[SUP] 3 [/SUP], Alexendre Delamou[SUP] 3 4 [/SUP], Alioune Camara[SUP] 3 [/SUP], Alpha-Kabinet Kéita[SUP] 1 5 [/SUP], Abdoulaye Touré[SUP] 1 2 3 [/SUP]
Affiliations
- PMID: 42459179
- PMCID: PMC13368709
- DOI: 10.3389/fepid.2026.1813211
Background: The HPAI H5N1 panzootic represents a critical threat to human health in Africa, where traditional poultry systems and dense human-animal interfaces facilitate frequent zoonotic spillover. While sporadic human cases raise pandemic concerns, continent-wide integration of spatial dynamics, transmissibility indicators, and surveillance performance has been lacking. This study quantifies avian influenza transmission over two decades across Africa, identifies geographical hotspots, and evaluates the responsiveness of current surveillance systems.
Methods: We analysed 8,037 avian influenza outbreak events and 369 laboratory-confirmed human cases, predominantly caused by HPAI H5N1 (2004-2025), using harmonised data from FAO (EMPRES-i+), WHO, and WOAH. A Bayesian Besag-York-Mollié (BYM) spatiotemporal model estimated residual transmission risks and Incidence Rate Ratios (IRR) by subtype. The basic reproduction number (R₀) was derived via an exponential growth model applied to human outbreak phases across infectious durations of 7-30 days. Surveillance responsiveness was assessed by quantifying notification delays between clinical observation and official reporting.
Results: Risk of infection in animals: HPAI H5N1 was the dominant strain, representing 87.8% of animal cases, with Egypt acting as the primary epidemiological epicentre (66% of total records). The spatiotemporal model revealed that H5N1 is associated with a significantly higher risk of animal infection (IRR = 8.37; 95% CI: 6.65-10.53). Although 71% of outbreaks were reported within 5 days of detection, significant delays (≥15 days) occurred in 12% of cases, with notable regional disparities. Risk of infection in human: H5N1 was associated with a 67-fold increase in the incidence of human cases compared to other subtypes (IRR = 66.78; 95% CI: 25.29-176.37). Sensitivity analyses yielded R[SUB]0[/SUB] estimates ranging from 1.05 (95% CI: 0.91-1.31) to 1.23 (95% CI: 0.60-2.33), indicating localised epidemic potential.
Conclusion: Our findings highlight a persistent and geographically heterogeneous H5N1 reservoir in Africa with high zoonotic affinity. Although sustained human-to-human transmission remains limited, the identification of dual poultry-human hotspots and localised R0 peaks underscores the urgent need for geographically targeted One Health interventions. Strengthening real-time reporting systems and improving biosecurity in high-risk poultry value chains are critical to mitigating future pandemic threats on the continent.
Keywords: Africa; H5N1; avian influenza; bayesian modelling; one health; public health surveillance.