"Spatio-temporal Dynamics and Risk Cluster Analysis of Highly Pathogenic Avian Influenza HPAI (H5N1) in Poultry: Advancing Outbreak Management through Customized Regional Strategies in Egypt"

doi:10.21203/rs.3.rs-3361650/v1

Download PDF

Research Article

"Spatio-temporal Dynamics and Risk Cluster Analysis of Highly Pathogenic Avian Influenza HPAI (H5N1) in Poultry: Advancing Outbreak Management through Customized Regional Strategies in Egypt"

https://doi.org/10.21203/rs.3.rs-3361650/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Highly Pathogenic Avian Influenza HPAI (H5N1) has been endemic in Egypt for almost two decades, profoundly impacting both the poultry industry and public health. Egypt stands as a prominent epicenter for HPAI H5N1 outbreaks in Africa, marked by the highest number of positive human cases. Despite continuous governmental efforts, prior research underscores the inadequacy of strategies in controlling the virus spread. Our innovative epidemiological approach focuses on identifying spatiotemporal patterns and high-risk clusters of HPAI H5N1 outbreaks at the subdistrict level. This involves trial tracking of HPAI H5N1 endemicity dynamics, enabling tailored interventions at a regional level based on robust epidemiological investigations to address the persistent challenge of HPAI H5N1 in Egypt. The study aims to illuminate spatiotemporal outbreak dynamics, with specific attention on Menofia. Despite the region's early poultry impacts, initial outbreaks didn’t originate from Menofia in studied epidemic waves. Outbreak risk spatial distribution displayed an escalating pattern at the northern border, followed by risk reduction through the sixth epidemic wave. The predominant hot spot region was localized within rural districts, particularly villages, while urbanization coincided with lower outbreak density. Observed smoothed densities revealed epidemic propagation within urban centers, preceding its transition to new areas and establishing direct connections with neighboring cities. Primary cluster prognostication is plausible, occurring in regions previously hosting elevated relative risk clusters during preceding epidemic waves. Identification of enduring pinpoint clusters, persistent for extended durations, indicates close contact dynamics and localized viral circulation within populations. In conclusion, our study highlights the significance of Customized regional interventions based on the rigorous epidemiological framework. This approach is pivotal in the profound comprehension of endemicity dynamics, efficiently limits geographical infection spread, and contains outbreaks within delineated areas.

Highly Pathogenic Avian Influenza

Spatiotemporal Dynamics

Risk Clusters

Endemicity Dynamics

Outbreak Management

Tailored Intervention Strategies

Highly pathogenic avian influenza (HPAI) subtype H5N1 poses a significant threat to the global poultry industry due to its genetic variability and pandemic potential [1]. In Egypt, the emergence of HPAI H5N1 in poultry was noted in 2006, marking the second African nation to report such an infection [2, 3]. Egypt stands out with the highest number of HPAI H5N1 notifications of positive human cases [4, 5] and is considered a significant epicenter for HPAI H5N1 outbreaks in Africa [6]. Despite persistent governmental efforts to curtail its dissemination, previous publications have revealed the insufficiency of applied strategies in effectively controlling the HPAI-H5N1 virus [7]. As a result, HPAI H5N1 remains entrenched and exerts severe repercussions on both the poultry sector and public health [8, 9].

Egypt's poultry production predominantly occurs in Lower Egypt, with concentrated outbreaks in the Nile Delta region due to dense poultry populations and agricultural activities [10-14]. The reported outbreak rates, however, may underestimate the virus's true prevalence and it doesn’t fully reflect the true widespread endemic situation of the virus in the region due to the prevailing policy of under-reporting of the disease in the commercial sector in an attempt to protect their business interests and to prevent the mass culling of poultry stocks [15, 16]. In Egypt, HPAI H5N1clade 2.2 was discovered for the first time in poultry in February 2006, since then, it has spread quickly across both backyard and commercial flocks [2, 3]. Subsequent epidemic waves have been documented since 2006, with Clade 2.2.1 observed during the first wave [10, 17]. A novel variant, Clade 2.2.1.1, emerged during this time, possibly due to vaccination efforts [10]. Despite interventions, the virus continued to spread among vaccinated backyard and commercial poultry, causing significant social and economic losses [18]. Egypt was officially designated as an Avian Influenza endemic nation in 2008, underscoring the virus's evolving nature [19]. Clade 2.2.1.1 further diversified between 2008 and 2011 into two clusters [20]. Notably, these clusters ceased to emerge thereafter[20], but a new variant (subclade 2.2.1.2) emerged in early 2014, with no significant alterations were found in the circulating viruses until 2016 [21, 22].

A critical obstacle hindering the successful implementation of control strategies lies in the absence of detailed and meaningful epidemiological investigations [16]. Country-level aggregated data, at times, obscure smaller, high-risk clusters [14, 23]. Therefore, the detection of the spatial cluster of the outbreaks could have been made accurately when involving finer detailed data in sub-districts within villages, and more reliable surveillance data in addition to up-to-date reported outbreak data [24].

Our primary objective is to develop an innovative epidemiological approach to enhance control policies against HPAI H5N1 in Egypt. This approach centers on identifying spatiotemporal patterns and high-risk clusters at the subdistrict level, with tracking of HPAI H5N1 endemicity dynamics enabling tailored interventions guided by robust epidemiological investigations. This approach has the potential to contribute significantly to reducing virus circulation and mitigating the risk of disease transmission in both poultry and human populations. The success of these strategies hinges on the continuous assessment and evaluation of their implementation, ensuring their adaptability and effectiveness over time.

Description of the Study Area

The Nile Delta region is about 40,000 km² which accommodates a substantial portion of Egypt's population and serves as a habitat for extensive poultry activity, including rearing, trade, and consumption [25]. This investigation's focal domain is the Menoufia governorate, a locale recognized for reporting elevated instances of HPAI [20]. Menoufia is one of the leading poultry-producing governorates in Egypt [26], where registered and unregistered commercial farms were found, in addition to the backyard which is randomly dispersed small-sized household flocks of mixed birds of different species and different ages with various vaccination protocols [27, 28].

Data collection

In Menofia, confirmed events of HPAI-H5N1 outbreaks in domestic poultry were officially reported to the Global Animal Disease Information System (EMPRES-I) provided by FAO [FAO, 29], and the Egyptian Ministry of Agriculture (General Organization For Veterinary Services) official reports for national surveillance from January 2006 to December 2017.

All data were integrated into one dataset and the villages were used as an administrative unit. The term outbreak was defined as 'the confirmed presence of disease with a clinical expression or not, in at least one individual in a defined location during a specified period’ [30]. The epidemic date and the outbreak centroids were utilized as spatiotemporal characteristics of each outbreak. The governorate map was constructed by (ArcGIS) to facilitate the presentation of data and the interpretation of results.

Method of analysis

Daily, weekly, and monthly epidemic curves, were constructed to display outbreak magnitude and temporal trends of HPAI-H5N1 outbreaks. An epidemic wave (EW) is a number of disease outbreaks that rapidly peaked and then gradually decreased till the end of the epidemic [31], and the number of outbreaks was calculated for each EW.

The interpolation tools were used to study the HPAIH5N1 outbreak risk. A raster risk map displays the interpolated surface, providing predictions for each location. The geostatistical analysis surface was derived using ordinary kriging by weighting the surrounding of the reported outbreak events to derive a prediction for the unreported locations. The general formula is formed as a weighted sum of the data values [32]:

where:

Z(si) = the reported outbreaks at the ith location

λi = an unknown weight for the number of outbreak events at the ith location

s₀ = the prediction location

n= the number of outbreak events

Spatially kernel density analysis (KDE) is a simple non-parametric technique that relies on a few assumptions about the structure of the observed data [33]. It is equivalent to a simple diffusion model that is a useful approximation to patterns of distribution frequently found in ecological data [34]. KDE was used to identify high-density areas [35]; run in Environmental Systems Research Institute ArcMap 10.5 software using reported cases to generate a density surface for each EW.

The quartic kernel function [36, 37] is given by:

where:

i = 1,…,n are the input points.
d_i is the distance between the point s and the observed event in location,
s_i and τ is the radius centered on s.

The formula to calculate the bandwidth is as follows [ESRI, 38]:

where:

SD is the standard distance
D_m is the median distance
n is the number of points if no population field is used, or if a population field is supplied, n is the sum of the population field values

Kernel density maps for the total number of cases were plotted for each EW to visualize the risk for the disease. Default cells and the output were selected in Square kilometers (Km²).

Retrospective space-time permutation scan statistics were performed to detect the Spatio-temporal clusters for each EW by investigating whether the outbreaks were interrelated in space and time using (SaTScan 8.2.1 software) [39-41]. The scanning window was a cylinder with the spatial and temporal dimensions as circular base and height, respectively. The temporal scanning window was set to <50% of the study period in each EW and a maximum of 50% of outbreaks were allowed in the spatial scanning window [42].

The likelihood ratio statistic was used to evaluate the possibility of a true spatiotemporal cluster in a window. The window with the maximum likelihood ratio statistic was considered the primary cluster while the remainder were considered secondary clusters. Their statistical significance was tested through Monte Carlo simulations of 999 replications [42]. The time units of day, week, and month were used, respectively.

ArcGIS 10.5 software was used to overlay results from different methods in a map for visual comparisons (ESRI, Redlands).

Temporal distribution of HPAI H5N1 outbreaks in Menofia governorate, Egypt

Throughout the study period encompassing 2006 to 2017, six distinct epidemic waves (EW1-6) of H5N1 outbreaks were discerned, as illustrated in Figure 1. The first epidemic wave (1^stEW) commenced in February 2006, persisted for approximately three months, and encompassed 66 reported outbreaks. This initial wave reached its peak in mid-March 2006, subsequently abating in May 2006. The second epidemic wave (2^ndEW) unfolded on November 21, 2006, and exhibited its peak in December of the same year, culminating in mid-April 2007. Notably, this wave spanned four months and comprised an estimated 57 outbreaks. Progressing to the third epidemic wave (3^rdEW), the Menofia governorate encountered 30 outbreaks, spanning a duration of three months from December 2, 2007, to mid-March 2008. The ensuing fourth epidemic wave (4^thEW) was characterized by its prolonged nature, reflecting the endemic status of the disease. This wave persisted from December 16, 2008, through February 11, 2012, resulting in a continuous surge of HPAI H5N1 outbreaks over a span of more than three years. The cumulative outbreak count during this extended period amounted to 378. The dynamics of the outbreak presented as four distinct epidemic cycles. The first cycle spanned from January 2009 to August 2009, followed by the second cycle from January 2010 to August 2010. The third cycle persisted from January 2011 to August 2011, with the final cycle peaking in January 2012. The fifth epidemic wave (5^th EW) started in September 2012 and persisted until late May 2013. This wave demonstrated a series of consecutive peaks, notably observed in October 2012, late November 2012, early February 2013, and mid-April 2013, constituting a cumulative total of 58 reported outbreaks. Subsequently, the sixth epidemic wave (6^th EW) emerged in October 2013, culminating on December 10, 2016, and registering 70 continuous outbreaks. A solitary definitive peak was observed during this wave in February 2015, as delineated in Figure 1.

Figures 2 and 3 offer a comprehensive depiction of outbreak distribution across localities for each of the six epidemic waves. Over the entire study duration, as depicted in Figure 2A, the localities reporting the highest numbers of outbreaks are Ashmun, Minuf, and Qwesna, in descending order. Following closely are Berkat_ElSabae, ElBagur, and Shibin_ElKom, constituting the second-highest category, succeeded by the third-highest category encompassing Tala, ElShohadaa, and Sadat_City. Table 1 and Figure 2B reveal that during the first epidemic wave (1^st EW), the predominant outbreak occurrences were observed in Qwesna, Berkat ElSabae, and Ashmun, representing 34.8%, 18.2%, and 16.7% of the total cases, respectively. Similarly, the second epidemic wave (2^ndEW) witnessed a concentration of outbreaks in Berkat ElSabae, Minuf, and Shibin_ElKom, accounting for 28.1%, 15.8%, and 12.3% of total cases, respectively, as indicated by Table 1 and Figure 2C. In the context of the third epidemic wave (3^rd EW), Ashmun, Shibin_ElKom, and Tala emerged as prominent outbreak regions, contributing to 33.3%, 26.7%, and 13.3% of total cases, respectively, as denoted in Table 1 and Figure 2D. Transitioning to the fourth epidemic wave (4^thEW), the focal points of outbreak notifications shifted to Minuf, Ashmun, and Qwesna, representing 24.6%, 17.7%, and 14% of total cases, respectively, as highlighted in Table 1 and Figure 2E. The fifth epidemic wave (5^th EW) underscored Ashmun as the locale with the highest outbreak instances within the Menofia governorate, comprising 43.1% of total outbreaks, closely followed by Qwesna at 34.5% of total outbreaks, as depicted in Table 1 and Figure 2F. Finally, during the sixth epidemic wave (6^th EW), the most substantial outbreak notifications in the Menofia governorate emanated from Ashmun, Minuf, and Shibin_ElKom, constituting 34.3%, 22.9%, and 12.9% of total cases, respectively, as illuminated by Table 1 and Figure 2G.

The spatial pattern of HPAI H5N1 outbreak in Menofia governorate, Egypt

The spatial distribution pattern of Highly Pathogenic Avian Influenza (HPAI) H5N1 outbreaks within the Menofia governorate is effectively depicted in Figure 3, which delineates the region into six distinct Epidemic Waves (EWs). The presented visualizations and predictive analyses through spatial mapping serve to enrich our understanding of the intricate dynamics characterizing HPAI H5N1 outbreaks within the Menofia governorate across successive Epidemic Waves. To facilitate comprehensive visual comparison, a village-based map is employed through ArcGIS 10.5 software (ESRI, Redlands, CA, USA).

Employing an ordinary kriging approach, the raster risk map illustrated in Figure 4 generates an interpolated surface that furnishes predictive insights into the likelihood of outbreak occurrences across various locations within the Menofia governorate during each Epidemic Wave. Notably, during the 1^st EW, Berkat ElSabae and Qwesna, situated along the northeast border of the governorate, exhibited the most pronounced predicted risk of HPAI H5N1 spatial occurrences. Subsequently, the subsequent Epidemic Wave witnessed a more widespread dissemination of outbreaks throughout the governorate, characterized by an overall heightened density. The highest predicted risk of outbreak occurrence was particularly concentrated in the northeastern cities (Berkat ElSabae and Qwesna cities), extending further to encompass the governorate's central cites (Minuf, ElShohadaa, and Shibin_ElKom cities), ultimately culminating in Ashmun city.

While in the context of the 3rd EW, a distinct shift in risk distribution was discerned, with the northeast segment of the governorate (encompassing Berkat ElSabae and Qwesna cities) displaying a notably diminished predicted risk of HPAI H5N1 occurrences. Conversely, the highest predicted risk was prominently identified along the governorate's northwestern borders (Tala city), extending towards its central region (Shibin_ElKom city), and extending southward to Ashmun city. Transitioning to the 4^th EW, outbreak dissemination followed a trend where the northern governorate borders exhibited the highest predicted risk. The 5^th EW witnessed a further expansion of outbreaks throughout the entirety of the governorate, with elevated predicted risk observed at both its northern and southern borders. In the 6^th EW, a discernible decrease in outbreak density was observed, with the highest predicted risk noted in Minuf and ElBagur.

The present investigation delves into a comprehensive depiction of outbreak density and cluster analysis of Highly Pathogenic Avian Influenza (HPAI) H5N1 outbreaks across six distinct Epidemic Waves (EWs) within the Menofia governorate. This analytical depiction is notably articulated through the illustrative framework presented in Table 2 and Table 3, as well as Figure 5. Outbreak density in the Menofia governorate was meticulously delineated via the utilization of adaptive kernel density estimation, wherein varying densities are visually represented by a gradient of monochromatic grey hues, with heightened density translating to a darker coloration. Concurrently, the geographic distribution of outbreaks is symbolized through the use of distinctive green dots. Turning our focus to the spatial dynamics characterizing each of the Epidemic Waves, it is observed that the initial outbreak wave (1^stEW) within the Menofia governorate registered its highest outbreak density concentrations within Qwesna and Shibin_ElKom.

As the progression of the second Epidemic Wave (2^nd EW) unfolds, the dispersion of outbreak densities spans across Qwesna, Berkat ElSabae, ElShohadaa, Minuf, and Ashmun. The third Epidemic Wave (3^rd EW) showcases a spatial extension of outbreak density that encompasses regions such as Tala, Berkat ElSabae, Shibin_ElKom, and Ashmun. The fourth Epidemic Wave (4^th EW) witnesses a notable proliferation of outbreaks enveloping the entirety of the governorate, with particularly heightened densities concentrated in Qwesna, Berkat ElSabae, ElBagur, and Ashmun. Shifting our attention to the fifth Epidemic Wave (5^th EW), the distribution of outbreak occurrences becomes more localized, primarily confined to Qwesna and Ashmun. Finally, within the sixth Epidemic Wave (6^thEW), areas such as Qwesna, Shibin_ElKom, ElShohadaa, Minuf, and Ashmun demonstrate a pronounced surge in outbreak density.

The integration of outbreak density and cluster analysis, as substantiated through the comprehensive presentation within Tables 2 and 3, along with the visual representation rendered by Figure 5, enriches our understanding of the intricate spatial dynamics underlying HPAI H5N1 outbreaks across successive Epidemic Waves within Menofia governorate.

Spatiotemporal clusters

Clustering, defined as an aggregation of diseases within specific spatial and temporal dimensions beyond what could be anticipated, is a fundamental concept in epidemiological analysis [43]. In this study, the identification of outbreak clusters across successive Epidemic Waves (EWs) in the Menofia governorate is achieved through spatiotemporal cluster analysis. However, it is noteworthy to acknowledge certain limitations inherent to this study, including the inability to fully account for various control measures, potential detection bias, and variations in demographic characteristics among at-risk groups. Tables 2 and 3, along with Figure 5, meticulously elucidate the precise locations and extent of diverse clusters within the Menofia governorate, offering a comprehensive overview

Over the course of six Epidemic Waves, the Menofia governorate exhibits discernible spatial patterns and the emergence of spatiotemporal clusters of HPAI subtype H5N1 outbreaks, manifesting at daily, weekly, and monthly intervals. Employing space-time permutation scan statistics, significant spatiotemporal clusters are discerned and characterized by both the most likely cluster (denoted by a red circle) and secondary clusters (indicated by blue dashed circles) within Tables 2 and 3, as well as Figure 5.

During the initial wave (1^st EW), the primary and most likely cluster emerged in Tala City towards the conclusion of the wave. Concurrently, secondary clusters encompassed Tala, ElShohadaa, Shibin_ElKom, Minuf, Qwesna, and Berkat ElSabae, spanning a radius of 26.37 kilometers based on monthly outbreak data. In the second Epidemic Wave (2^ndEW), the primary cluster (with a radius of 6.56 kilometers) manifested in Tala, ElShohadaa, and Shibin_ElKom towards the wave's end, considering both weekly and monthly outbreak data. Additionally, the primary cluster based on daily outbreak data extended to include Minuf city within a larger radius of 12.60 kilometers. Secondary clusters encompassed the entire governorate during this wave, with notable highest relative risk clusters situated in small villages such as Ashmun, ElBagur, and Shibin_ElKom.

Transitioning to the third Epidemic Wave (3^rd EW), the daily-based analysis revealed a primary cluster at the end of the wave, characterized by a radius of 6.87 kilometers and centered in ElBagur and Ashmun. The monthly-based data, however, showcased a smaller primary cluster (2.14 kilometers radius) exclusively in Ashmun. Furthermore, the weekly analysis detected a primary cluster encompassing ElBagur, Shibin_ElKom, Qwesna, and Berkat ElSabae, spanning a radius of 9.29 kilometers. Secondary clusters exhibited prevalence across the majority of cities, with prominent most likely clusters identified within the villages of Shibin_ElKom, Ashmun, Berkat ElSabae, and Tala.

The fourth Epidemic Wave (4^th EW) featured a primary cluster positioned in ElBagur, Qwesna, and Shibin_ElKom at the wave's midpoint, encompassing an area with a 10.21 kilometers radius as depicted in Figure 5, based on daily and weekly outbreak data. Contrastingly, the analysis of monthly-based outbreaks revealed a smaller primary cluster (5.17 kilometers radius) in Qwesna and Berkat ElSabae. Notably, secondary clusters sequentially covered the entire geographic expanse of the governorate during this wave, with the most elevated relative risk clusters predominantly concentrated in Minuf and Tala.

By the culmination of the fifth Epidemic Wave (5^th EW), a primary cluster characterized by a 14.01 kilometers radius emerged in Minuf, Ashmun, and ElBagur based on weekly outbreak data. Alternatively, the monthly-based analysis indicated smaller primary clusters (6.46 kilometers radius) in the same cities with the highest relative risk. Within the context of daily outbreak data, the primary cluster encompassed solely Shibin_ElKom and Qwesna, spanning a radius of 4.04 kilometers. Remarkably, the highest relative risk secondary clusters were identified in Tala, Berkat ElSabae, Ashmun, and Minuf.

Transitioning to the sixth Epidemic Wave (6^th EW), primary clusters materialized at various points throughout the wave, spanning daily, weekly, and monthly outbreak data. These primary clusters were consistently accompanied by secondary clusters that pervaded the entirety of the governorate. Within the daily-based analysis, primary clusters (7.97 kilometers radius) were detected in Minuf and Ashmun towards the close of 2014. Likewise, the monthly-based analysis unveiled a primary cluster (10.06 kilometers radius) encompassing Minuf, Shibin_ElKom, and ElBagur from January 2015, persisting for six months. Finally, the weekly-based examination identified a primary cluster (6.02 kilometers radius) located in Tala, Berkat ElSabae, and Shibin_ElKom at the end of 2015. Secondary clusters with elevated relative risk were notably situated in Ashmun, Qwesna, ElShohadaa, Shibin_ElKom, ElBagur, and Minuf, encompassing nearly all cities within the Menofia governorate.

Our study aimed to investigate the spatiotemporal pattern, identify, and trace the highest risk clusters of HPAI H5N1 outbreaks at the subdistrict/village level within the Menofia governorate, from 2006 to 2017 as a trial for tracking the HPAI H5N1 endemicity dynamics for better establishment of effective disease control strategies at that level.

Epidemic waves of HPAI H5N1 outbreaks in Menofia governorate

The analysis of HPAI subtype H5N1 outbreaks in Menofia governorate revealed a notable pattern across six distinct Epidemic Waves (EWs), characterized by varying time scales of day, week, and month. These EWs exhibited consistent patterns, underscoring the robust epidemic nature of the disease. The duration of each EW ranged from 3, 5, 3, 38, 7, and 39 months, as depicted in Figure 1. Notably, the epidemic curve within Menofia governorate closely mirrored the temporal patterns observed in Egypt's broader epidemiological landscape, as elucidated by previous research conducted by [7, 14], and a global study by [44].

A recurring feature of the epidemic curve is the occurrence of disease peaks in December and January, indicative of an apparent noninterpreted epidemic cycle. This pattern aligns with the cyclic nature of the disease, highlighting its endemic status. The configuration of the epidemic curve, characterized by a brief vertical span and an extensive horizontal span, underscores the endemicity and suggests an expanding spatial distribution of outbreaks over time within the governorate.

The Menofia governorate played a pivotal role in the broader Egyptian epidemic waves. For instance, the 1^stEW in Menofia commenced merely one week following the disease's initial introduction in Egypt and concluded two months prior to the termination of the country's inaugural epidemic wave, according to the epidemic waves of HPAI H5N1 outbreaks in Egypt described by Y Elsobky, G El Afandi, A Salama, A Byomi, M Omar and M Eltholth [14]. Notably, the 2.2.1 clade emerged during the 1^st EW and persisted throughout the subsequent 2^nd and 3^rd EWs. Additionally, the 2.2.1.1 clade, introduced in the 2^nd EW, continued to be prevalent during the 3^rd EW, ultimately culminating in endemicity by 2008 [45]. Remarkably, Menofia governorate consistently reported higher outbreak numbers during distinct EWs, although it was not the origin of outbreaks. This phenomenon can be attributed to the high density of the poultry population in the governorate, thereby contributing to its prominence in outbreak notifications.

Further exploration of the 4^thEW in Menofia reveals a distinct epidemiological landscape. This wave was notable for being associated with the coexistence of two clades, namely 2.2.1.1a and 2.2.1.2, coinciding with the emergence of H9N2 [14, 45]. It is noteworthy that the 4^th EW exhibited a peculiar pattern wherein outbreaks initially occurred during warmer months, suggesting a possible link between clade stability and thermostability [45]. This underscores the virus's capacity for sustained endemicity through adaptive processes [46, 47].

During the 5^th EW, a protracted duration of continuous outbreak notifications was observed, characterized by irregular peaks and altered epidemic cycles. This phenomenon, corroborated by the presence of clades 2.2.1, 2.2.1.1a, and 2.2.1.2, aligns with the concept of viral adaptation and endemicity, a notion that concurs with findings by R El-Shesheny, A Kandeil, A Mostafa, MA Ali and RJ Webby [45]. Conversely, the 6^th EW culminated without distinct peaks or discernible patterns, potentially attributed to under-reporting and limitations within the surveillance system. This observation resonates with the findings of Y Elsobky, G El Afandi, A Salama, A Byomi, M Omar and M Eltholth [14] regarding epidemic waves in Egypt.

In essence, the analysis of HPAI H5N1 outbreaks in the Menofia governorate underscores the dynamic interplay between viral evolution, adaptation, and endemicity across distinct Epidemic Waves. The governorate's role in shaping the broader epidemiological landscape is pivotal, and the implications of these findings emphasize the importance of vigilance and refined surveillance strategies to address and manage the evolving dynamics of HPAI H5N1 outbreaks to contain outbreaks within delineated areas where the initiation phase starts before the outbreak acceleration phase.

The spatial pattern of outbreak density in Menofia governorate:

The geographic distribution of HPAI H5N1 outbreaks in Egypt has been extensively studied, consistently revealing a concentration of poultry outbreaks within the densely populated Nile Delta region, while demonstrating comparatively lower densities in upper Egypt [5, 14, 18, 48]. This pattern is attributed to the significant poultry stocks and human population present in the Nile Delta region [11]. Notably, more than half of Egypt's population is in direct proximity to poultry in the Nile Delta region, as reported by [49]. Consequently, the Menofia governorate, situated within Lower Egypt, assumes a notable role in the potential zoonotic transmission of H5N1, as outlined in a study by [5] study.

As evident from Figure 4, the spatial risk of HPAI H5N1 outbreaks escalates along the governorate's borders with neighboring provinces, particularly its northern border. Additionally, the elevated risk is observed in locales characterized by numerous small unlicensed farms and backyard poultry. These areas, known for their random distribution across Menofia's cities, underscore the importance of vigilance and active surveillance. In the context of the 6^th Epidemic Wave (EW), a decline in spatial risk is noted, potentially attributed to the under-notification of outbreaks, highlighting the urgency of enhanced surveillance coordination among governorates.

Effective control strategies are imperative and should encompass measures such as safeguarding borders, quarantining affected villages, appropriate disposal of deceased birds, restrictions on live bird transportation and markets, and meticulous vaccination coverage coupled with robust biosecurity practices. Notably, the longevity of HPAI H5N1 virus survival under diverse temperatures underscores the significance of these control measures [50, 51].

Kernel density analyses depicted in Figure 5 elucidate the outbreak dynamics within Menofia, with rural districts, particularly those situated within the delta region, exhibiting prominent hotspots across all six Epidemic Waves. This phenomenon is linked to the prevalence of household poultry, both registered and unregistered small poultry farms, and informal markets. The pattern also demonstrates a decrease in outbreak density with increased urbanization, elucidating lower densities within more urbanized zones like main city areas and industrial regions such as Sadat City, consistent with findings by Gafi [52]. Observed smoothed densities revealed epidemic propagation within urban centers, preceding its transition to new areas and establishing direct connections with neighboring cities, the same as illustrated by [53]. This perspective indicates that the control measure is not effective in containing outbreaks within delineated areas, which refers to the importance of early actions to control the spread of disease successfully.

The identified clusters, both primary and secondary, reinforce the notion of successive outbreak occurrences within Menofia. Primary clusters, consistently identified towards the middle or end of each wave, often align with high relative risk clusters from preceding waves. These clusters are situated within the hot-spot region, albeit with varying locations and sizes. The proactive control of such clusters necessitates intensified interventions at the outset of a wave, thereby curbing spatial disease spread.

The sequential emergence of secondary clusters extends to encompass the entirety of the governorate's geographical expanse, demonstrating enduring persistence over extended periods, frequently spanning months. This phenomenon is concomitantly characterized by the proliferation of discrete pinpoint clusters, indicative of localized dissemination of infection through direct contact within poultry populations, frequently transpiring in the absence of efficacious intervention measures.

However, the success of control measures remains limited in effectively compressing spatial disease spread and curtailing epidemic progression. This emphasizes the urgency for tailored disease control strategies operating at subdistrict levels within each governorate, informed by a nuanced understanding of endemicity dynamics. Furthermore, continuous evaluation and refinement of control strategies based on real-world data are essential, given the challenges associated with obtaining accurate and comprehensive outbreak data.

Epidemic wave: EW

Highly Pathogenic Avian Influenza HPAI (H5N1): HPAI H5N1

Ethical approval and consent to participate: The authors confirm that the ethical policies of the journal, as noted on the journal’s author guidelines page, have been adhered to. No ethical approval was required as this is a research article with no original research data.

Consent for publication: Not applicable.

Availability of data and materials: All data generated or analyzed during this study are included in this published article.

Competing interests: The authors declare that they have no competing interests.

Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Authors' contributions: YE the first author is the one who did the research and wrote the manuscript’s first draft data collection and processing, and was a major contributor in cleaning, analysis, and interpretation. ME (Eltholth) and EA have given significant intellectual inputs in the final revision and participated in the conceiving and designing of the manuscript. MN, AS, and WM Made significant contributions to the cleaning and interpretation of the data. NE, GH, and AK participated in cleaning, and writing the manuscript. ME (Elkamshishi) participated in data collection, processing, and writing. All authors critically reviewed the manuscript and approved the final manuscript.

Acknowledgments: Not applicable.

Otte J, Hinrichs J, Rushton J, Roland-Holst D, Zilberman D: Impacts of avian influenza virus on animal production in developing countries. CAB Reviews: Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources 2008, 3(080):18.
Aly M, Arafa A, Hassan M: Epidemiological findings of outbreaks of disease caused by highly pathogenic H5N1 avian influenza virus in poultry in Egypt during 2006. Avian diseases 2008, 52(2):269-277.
Saad MD, Lu’ay SA, Gamal-Eldein MA, Fouda MK, Khalil FM, Yingst SL, Parker MA, Montevillel MR: Possible avian influenza (H5N1) from migratory bird, Egypt. Emerging infectious diseases 2007, 13(7):1120.
Harfoot R, Webby RJ: H5 influenza, a global update. Journal of microbiology 2017, 55(3):196-203.
Young SG, Kitchen A, Kayali G, Carrel M: Unlocking pandemic potential: prevalence and spatial patterns of key substitutions in avian influenza H5N1 in Egyptian isolates. BMC infectious diseases 2018, 18(1):1-13.
Ekong P, Fountain‐Jones N, Alkhamis M: Spatiotemporal evolutionary epidemiology of H5N1 highly pathogenic avian influenza in West Africa and Nigeria, 2006–2015. Transboundary and emerging diseases 2018, 65(1):e70-e82.
Elsobky Y, Nganwa D, El Afandi G, Byomi A, Reddy G, Abdalla E: A quantitative risk assessment to evaluate the efficacy of mitigation strategies to reduce highly pathogenic avian influenza virus, subtype H5N1 (HPAI H5N1) in the Menoufia governorate, Egypt. BMC Vet Res 2021, 17(1):210. DOI: 10.1186/s12917-021-02917-7.
Kalthoff D, Globig A, Beer M: (Highly pathogenic) avian influenza as a zoonotic agent. Veterinary microbiology 2010, 140(3-4):237-245.
Matrosovich M, Zhou N, Kawaoka Y, Webster R: The surface glycoproteins of H5 influenza viruses isolated from humans, chickens, and wild aquatic birds have distinguishable properties. Journal of virology 1999, 73(2):1146-1155.
Arafa A, El-Masry I, Khoulosy S, Hassan MK, Soliman M, Fasanmi OG, Fasina FO, Dauphin G, Lubroth J, Jobre YM: Predominance and geo-mapping of avian influenza H5N1 in poultry sectors in Egypt. Geospatial health 2016.
Arafa A, Suarez D, Kholosy S, Hassan M, Nasef S, Selim A, Dauphin G, Kim M, Yilma J, Swayne D: Evolution of highly pathogenic avian influenza H5N1 viruses in Egypt indicating progressive adaptation. Archives of virology 2012, 157(10):1931-1947.
El-Zoghby EF, Aly MM, Nasef SA, Hassan MK, Arafa A-S, Selim AA, Kholousy SG, Kilany WH, Safwat M, Abdelwhab E: Surveillance on A/H5N1 virus in domestic poultry and wild birds in Egypt. Virology journal 2013, 10(1):1-10.
Sheta BM, Fuller TL, Larison B, Njabo KY, Ahmed AS, Harrigan R, Chasar A, Aziz SA, Khidr A-AA, Elbokl MM: Putative human and avian risk factors for avian influenza virus infections in backyard poultry in Egypt. Veterinary microbiology 2014, 168(1):208-213.
Elsobky Y, El Afandi G, Salama A, Byomi A, Omar M, Eltholth M: Spatiotemporal analysis of highly pathogenic avian influenza (H5N1) outbreaks in poultry in Egypt (2006 to 2017). BMC Vet Res 2022, 18(1):174. DOI: 10.1186/s12917-022-03273-w.
Kayali G, Webby RJ, Ducatez MF, El Shesheny RA, Kandeil AM, Govorkova EA, Mostafa A, Ali MA: The epidemiological and molecular aspects of influenza H5N1 viruses at the human-animal interface in Egypt. PloS one 2011, 6(3):e17730.
Farnsworth ML, Hamilton-West C, Fitchett S, Newman SH, de La Rocque S, De Simone L, Lubroth J, Pinto J: Comparing national and global data collection systems for reporting, outbreaks of H5N1 HPAI. Preventive veterinary medicine 2010, 95(3-4):175-185.
Balish AL, Davis CT, Saad MD, El-Sayed N, Esmat H, Tjaden JA, Earhart KC, Ahmed LaE, El-Halem MA, Ali AHM: Antigenic and genetic diversity of highly pathogenic avian influenza A (H5N1) viruses isolated in Egypt. Avian diseases 2010, 54(s1):329-334.
Kayali G, Kandeil A, El-Shesheny R, Kayed AS, Maatouq AM, Cai Z, McKenzie PP, Webby RJ, El Refaey S, Kandeel A: Avian influenza A (H5N1) virus in Egypt. Emerging infectious diseases 2016, 22(3):379.
El-Shesheny R, Kandeil A, Bagato O, Maatouq AM, Moatasim Y, Rubrum A, Song M-S, Webby RJ, Ali MA, Kayali G: Molecular characterization of avian influenza H5N1 virus in Egypt and the emergence of a novel endemic subclade. The Journal of general virology 2014, 95(Pt 7):1444.
Arafa A, El-Masry I, Kholosy S, Hassan MK, Dauphin G, Lubroth J, Makonnen YJ: Phylodynamics of avian influenza clade 2.2. 1 H5N1 viruses in Egypt. Virology journal 2016, 13(1):49.
Arafa A, Naguib M, Luttermann C, Selim A, Kilany W, Hagag N, Samy A, Abdelhalim A, Hassan M, Abdelwhab E: Emergence of a novel cluster of influenza A (H5N1) virus clade 2.2. 1.2 with putative human health impact in Egypt, 2014/15. Eurosurveillance 2015, 20(13):21085.
Abdelwhab E, Hassan M, Abdel-Moneim A, Naguib M, Mostafa A, Hussein I, Arafa A, Erfan A, Kilany W, Agour M: Introduction and enzootic of A/H5N1 in Egypt: Virus evolution, pathogenicity and vaccine efficacy ten years on. Infection, Genetics and Evolution 2016, 40:80-90.
Jones SG, Kulldorff M: Influence of spatial resolution on space-time disease cluster detection. PLoS One 2012, 7(10):e48036. DOI: 10.1371/journal.pone.0048036.
Dhingra MS, Dissanayake R, Negi AB, Oberoi M, Castellan D, Thrusfield M, Linard C, Gilbert M: Spatio-temporal epidemiology of highly pathogenic avian influenza (subtype H5N1) in poultry in eastern India. Spatial and spatio-temporal epidemiology 2014, 11:45-57. DOI: 10.1016/j.sste.2014.06.003.
Abdelwhab EM, Hafez HM: An overview of the epidemic of highly pathogenic H5N1 avian influenza virus in Egypt: epidemiology and control challenges. Epidemiology and infection 2011, 139(5):647-657. DOI: 10.1017/s0950268810003122.
ElMasry I, Elshiekh H, Abdlenabi A, Saad A, Arafa A, Fasina FO, Lubroth J, Jobre Y: Avian influenza H5N1 surveillance and its dynamics in poultry in live bird markets, Egypt. Transboundary and emerging diseases 2017, 64(3):805-814.
Fasina FO, Ali A, Yilma J, Thieme O, Ankers P: Production parameters and profitability of the Egyptian household poultry sector: a survey. World's Poultry Science Journal 2016, 72(1):178-188.
Ali A, Ankers P, DeHaan N, Saad A, Hussein S, Lubroth J, Jobre Y: Mapping influenza A (H5N1) virus transmission pathways and critical control points in Egypt. FAO Animal Production and Health Working Paper 2013(11).
Food and Agriculture Organization: EMPRES‐i‐Global Animal Disease Information System. http://empres-i.fao.org/eipws3g/#h=0. 2017.
Toma B, Benet J-JC, Duford BC, Eloit MC, Marsh WC, Michel PC, Moutou FC, Sanaa MC, Vaillancourt J-P: Dictionary of veterinary epidemiology. In.: Iowa State University Press; 1999.
Zhang Z, Chen D, Chen Y, Liu W, Wang L, Zhao F, Yao B: Spatio-temporal data comparisons for global highly pathogenic avian influenza (HPAI) H5N1 outbreaks. PLoS One 2010, 5(12):e15314.
Oliver MA, Webster R: Kriging: a method of interpolation for geographical information systems. International Journal of Geographical Information Systems 1990, 4(3):313-332. DOI: 10.1080/02693799008941549.
Worton BJ: Kernel methods for estimating the utilization distribution in home‐range studies. Ecology 1989, 70(1):164-168. DOI: https://doi.org/10.2307/1938423.
O’Brien SH, Webb A, Brewer MJ, Reid JB: Use of kernel density estimation and maximum curvature to set Marine Protected Area boundaries: Identifying a Special Protection Area for wintering red-throated divers in the UK. Biological Conservation 2012, 156:15-21. DOI: https://doi.org/10.1016/j.biocon.2011.12.033.
Silverman BW: Density estimation for statistics and data analysis (1st ed.): Routledge; 1998. DOI: https://doi.org/10.1201/9781315140919.
Bailey TC, Gatrell AC: Interactive spatial data analysis, vol. 413: Longman Scientific & Technical Essex; 1995. DOI: http://www.personal.psu.edu/faculty/f/k/fkw/rsoc597/UgandaMaps.pdf.
Matsumoto PSS, Hiramoto RM, Pereira VBR, Camprigher VM, Taniguchi HH, de Raeffray Barbosa JE, Cortez LRPdB, Fonseca EdS, Guimarães RB, Tolezano JE: Impact of the dog population and household environment for the maintenance of natural foci of Leishmania infantum transmission to human and animal hosts in endemic areas for visceral leishmaniasis in Sao Paulo state, Brazil. PloS one 2021, 16(8):e0256534. DOI: https://doi.org/10.1371/journal.pone.0256534.
Environmental Systems Research Institute (ESRI): ArcGIS Desktop Help 10.2 Geostatistical Analyst. https://resources.arcgis.com/en/help/main/10.2/index.html#//009z000000110000002014.

Kulldorff M, Heffernan R, Hartman J, Assunçao R, Mostashari F: A space–time permutation scan statistic for disease outbreak detection. Plos med 2005, 2(3):e59.
Kulldorff M: Information Management Services, Inc.(2009) SaTScanTM v8. 0: Software for the spatial and space-time scan statistics. URL http://www satscan org 2010.
Onozuka D, Hagihara A: Spatial and temporal dynamics of influenza outbreaks. Epidemiology 2008, 19(6):824-828.
Zhang Z, Chen D, Chen Y, Davies TM, Vaillancourt J-P, Liu W: Risk signals of an influenza pandemic caused by highly pathogenic avian influenza subtype H5N1: spatio-temporal perspectives. The veterinary journal 2012, 192(3):417-421.
Porta M, ed: A dictionary of epidemiology. Oxford university press 2014.
Zhang Z, Chen D, Chen Y, Davies TM, Vaillancourt JP, Liu W: Risk signals of an influenza pandemic caused by highly pathogenic avian influenza subtype H5N1: spatio-temporal perspectives. Veterinary journal (London, England : 1997) 2012, 192(3):417-421. DOI: 10.1016/j.tvjl.2011.08.012.
El-Shesheny R, Kandeil A, Mostafa A, Ali MA, Webby RJ: H5 Influenza Viruses in Egypt. Cold Spring Harbor Perspectives in Medicine 2020:a038745.
Salaheldin AH, Kasbohm E, El-Naggar H, Ulrich R, Scheibner D, Gischke M, Hassan MK, Arafa A-SA, Hassan WM, El-Hamid A: Potential biological and climatic factors that influence the incidence and persistence of highly pathogenic H5N1 avian influenza virus in Egypt. Frontiers in microbiology 2018, 9:528.
Arafa A, El-Masry I, Kholosy S, Hassan MK, Dauphin G, Lubroth J, Makonnen YJ: Phylodynamics of avian influenza clade 2.2. 1 H5N1 viruses in Egypt. Virology journal 2016a, 13(1):1-11.
ELbayoumi KM, Mahgoub K, Mekky HM, Hassan ER, Amin Girh Z, Maatouq AM, El-Samadony HA, Rabie NS, MAA A KM: Molecular detection of H5N1, H9N2 and Newcastle disease viruses isolated from chicken in mixed infection in Egypt. World Applied Sciences Journal 2013, 27(1):44-50.
Abdelwhab E, Hafez H: An overview of the epidemic of highly pathogenic H5N1 avian influenza virus in Egypt: epidemiology and control challenges. Epidemiology & Infection 2011, 139(5):647-657.
Stallknecht DE, Shane SM, Kearney MT, Zwank PJ: Persistence of avian influenza viruses in water. Avian diseases 1990, 34(2):406-411.
Si Y, de Boer WF, Gong P: Different environmental drivers of highly pathogenic avian influenza H5N1 outbreaks in poultry and wild birds. PLoS One 2013, 8(1):e53362. DOI: 10.1371/journal.pone.0053362.
Gafi: "Industrial Zones of Governorate". Ministry of Investment Egypt. Archived from the original on 2018-11-23. Retrieved 23 November 2018. . 2018. DOI: https://www.gafi.gov.eg/English/StartaBusiness/InvestmentZones/Pages/Industrial-Zones.aspx
Norström M, Pfeiffer DU, Jarp J: A space-time cluster investigation of an outbreak of acute respiratory disease in Norwegian cattle herds. Preventive veterinary medicine 1999, 47(1-2):107-119. DOI: 10.1016/s0167-5877(00)00159-8.

Tables 1-3 is available in the Supplementary Files section.

No competing interests reported.

1.Tables.xlsx

Download PDF

Version 1

posted

You are reading this latest preprint version

"Spatio-temporal Dynamics and Risk Cluster Analysis of Highly Pathogenic Avian Influenza HPAI (H5N1) in Poultry: Advancing Outbreak Management through Customized Regional Strategies in Egypt"

Status:

Version 1

Abstract

Figures

Introduction

Data And Methods

Results

Discussion

Abbreviations

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1