Spatial and temporal analyses of highly pathogenic avian influenza H5N1 re-emergence in Nigeria, 2015
Icomiare Lucky Adebudo, Babasola Oluseyi Olugasa, Eugene Odigie, Simeon Ajisegiri
Abstract
Icomiare Lucky Adebudo, Babasola Oluseyi Olugasa, Eugene Odigie, Simeon Ajisegiri
Abstract
Introduction: highly pathogenic avian influenza (HPAI) virus H5N1 kills up to 90 - 100% of domestic fowl flock during outbreaks globally. This study, conducted from January 8, 2015 to August 6, 2015, aimed to assess the space-time distribution pattern of the re-emergence of HPAI H5N1 in Nigeria. Methods: secondary analysis was done on confirmed cases of HPAI data abstracted from records of the Federal Department of Veterinary and Pest Control Service, Abuja. We sorted cases into confirmed and suspected HPAI. Geographic coordinates of affected farms was obtained using Google Earth Pros. We conducted analysis of the data with Kulldorffs two-dimensional spatial scan statistics based on the use of circular window to identify whether the HPAI cases were randomly distributed over space or not. Spatial scan was used to detect clusters in space (assuming a Bernoulli distribution). Primary and secondary clusters were determined using likelihood-ratio test statistics. Time and space-time permutation was developed for prediction of possible outbreak. Chi-square was calculated at significant level set at 5%. Results: the virus affected 20 states (55.6%) in Nigeria. HPAI H5N1 epidemics occurred in all the six geo-political zones of Nigeria. Primary cluster and two secondary clusters of purely spatial scan statistics were all significant at p ≤ 0.05, and revealed that 95.1% of H5N1-infected farms were located within a 0.091km radius (Log likelihood ratio = 25.68). The centre coordinates for the clusters were found to be areas of high human density and considerable swine production. The Space-time permutation model identified a significant prospective outbreaks within 3.5 years interval in Plateau State (χ2 = 58.08, P < 0.001). Conclusion: this finding portends significant potential for co-infection and re-assortment of the virus with the human and swine strains. We recommend the need for enhanced surveillance in both poultry and swine populations in Nigeria.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Introduction: highly pathogenic avian influenza (HPAI) virus H5N1 kills up to 90 - 100% of domestic fowl flock during outbreaks globally. This study, conducted from January 8, 2015 to August 6, 2015, aimed to assess the space-time distribution pattern of the re-emergence of HPAI H5N1 in Nigeria. Methods: secondary analysis was done on confirmed cases of HPAI data abstracted from records of the Federal Department of Veterinary and Pest Control Service, Abuja. We sorted cases into confirmed and suspected HPAI. Geographic coordinates of affected farms was obtained using Google Earth Pros. We conducted analysis of the data with Kulldorffs two-dimensional spatial scan statistics based on the use of circular window to identify whether the HPAI cases were randomly distributed over space or not. Spatial scan was used to detect clusters in space (assuming a Bernoulli distribution). Primary and secondary clusters were determined using likelihood-ratio test statistics. Time and space-time permutation was developed for prediction of possible outbreak. Chi-square was calculated at significant level set at 5%. Results: the virus affected 20 states (55.6%) in Nigeria. HPAI H5N1 epidemics occurred in all the six geo-political zones of Nigeria. Primary cluster and two secondary clusters of purely spatial scan statistics were all significant at p ≤ 0.05, and revealed that 95.1% of H5N1-infected farms were located within a 0.091km radius (Log likelihood ratio = 25.68). The centre coordinates for the clusters were found to be areas of high human density and considerable swine production. The Space-time permutation model identified a significant prospective outbreaks within 3.5 years interval in Plateau State (χ2 = 58.08, P < 0.001). Conclusion: this finding portends significant potential for co-infection and re-assortment of the virus with the human and swine strains. We recommend the need for enhanced surveillance in both poultry and swine populations in Nigeria.
Key concepts: Influenza A virus subtype H5N1, Highly pathogenic, Flock, Outbreak, Veterinary medicine, Virology, Avian influenza virus, Biology