Introduction

Poultry serves as a rapid and cost-effective source of high-quality animal protein in the form of meat and eggs, making it a popular dietary choice globally1,2. Unlike some meats restricted by religious practices, chicken is widely accepted across all major religions, making it a critical component of human nutrition3,4. In densely populated Bangladesh, the poultry industry plays a pivotal role in national food security, with many people relying heavily on chicken products as a primary source of protein. However, poultry production in Bangladesh often suffers from outbreaks of infectious diseases5, such as high pathogenicity avian influenza (HPAI), a zoonotic disease with potentially fatal consequences for both human and poultry6,7. In addition to repeated outbreaks of HPAI, low pathogenicity avian influenza (LPAI) such as the H9N2 virus strain is endemic in Bangladesh, with seroprevalence in commercial poultry often exceeding 40% and prevalence in live bird markets (LBMs) reported to be between 25% and 50% in recent years8,9,10,11. The LPAI H9N2 has been detected continuously since its first identification in Bangladesh in 2006, causing significant drops in egg production and predisposing flocks to secondary infections. Its ongoing circulation provides opportunities for viral reassortment that could increase pathogenicity or zoonotic potential.

Effective poultry production requires appropriate biosecurity measures to prevent the introduction and spread of pathogens of public health significance and ensure food safety12,13. In poultry farming, biosecurity refers to a set of technical, practical, and structural strategies designed to minimize the risk of disease outbreaks like HPAI14. These measures are equally important for preventing and controlling the spread of LPAI, which can circulate silently in poultry populations for extended periods. Without strong biosecurity, both LPAI and HPAI can persist on farms and in markets, undermining productivity and increasing the risk of human exposure8,11. Biosecurity in poultry farms includes isolation of poultry, control of human and animal traffic, sanitation of vehicles and equipment, use of physical barriers (e.g., fences and mesh), and appropriate design and management of poultry houses15,16,17,18. Together, these practices reduce the introduction, persistence, and dissemination of infectious agents, benefiting farmers, consumers, and the broader industry7,17, and generally encompass three main principles including isolation, traffic control, and sanitation.

In Bangladesh, government efforts to support the poultry industry include strategies for improving breeds, feed, and housing19,20. The Food and Agriculture Organization (FAO), in collaboration with the Ministry of Fisheries and Livestock (MOFAL) and the Department of Livestock Services (DLS), has developed a national biosecurity guideline for commercial poultry farms21. These guidelines include five major principles addressing different aspects of isolation, vehicle control, farm and personal hygiene, flock healthcare, and compliance with government regulations, which directly inform the practices assessed in this study.

Earlier studies reported that, many commercial poultry farms in Bangladesh lack adequate biosecurity measures22,23,24,25. For instance, workers are often use to enter industrial poultry farms without cleaning their shoes, clothes, or equipment23,24. Additionally, access of domestic and wild animals and improper waste disposal further compromise farm biosecurity23,24. Similar issues have been observed in other countries of the world including Nigeria26, Ethiopia1,27, and Cameroon7, where maintaining effective biosecurity in poultry farms remains challenging1,7,26,27.

The absence of proper biosecurity measures can lead to devastating disease outbreaks, as evidenced by the HPAI H5N1 virus, which nearly halved Bangladesh’s commercial chicken farms between 2007 and 2013 25. The continued high prevalence of H9N2 in Bangladeshi poultry further demonstrates that without effective control and prevention strategies, both HPAI and LPAI can establish sustained long-term circulation. This threatens animal health, human health, and poultry trade, and underscores the urgent need to strengthen compliance with biosecurity guidelines10. Many African nations like Egypt, Ethiopia, and Nigeria, have also experienced widespread mortality in the poultry industry due to the HPAI H5N1 virus7. Strengthening biosecurity remains the most effective way to prevent and control HPAI outbreaks28,29.

Despite its significance, there is limited research evaluating compliance with biosecurity guidelines across different poultry farming systems (broiler, layer, and Sonali) in Bangladesh. This study aims to assess adherence to national biosecurity standards among different farm types and to examine the association between specific biosecurity practices and the occurrence of HPAI outbreaks. The findings are intended to support improvements in current regulations and promote a One Health approach to disease prevention and control.

Materials and methods

Study design, location, and duration

We conducted a cross-sectional study on commercial poultry farms in Bangladesh between September and December 2021. The study was carried out in six districts namely Dhaka, Gazipur, Manikganj, Munshiganj, Narayanganj, and Narsingdi of Bangladesh. These districts were selected because they have higher poultry production30. Unlike many other districts that usually have one dominant poultry species, these districts have broiler, layer, and Sonali farms repressing both meat and egg production systems. This allowed us to capture the full range of production systems within the same geographical areas.

Sample size and sampling

We selected a total of 331 commercial poultry farms which were identified through the local community veterinarian and referrals from other farm owners. The study comprised 113 broiler farms, 103 layer farms, and 115 Sonali farms. We purposively selected farms so that all three farm types were almost equal in number from each district. This approach allowed us to reflect on and evaluate biosecurity management practices across all farm types within each district and to ensure that the findings represented the diversity of management systems in high production areas. We produced a geospatial map using ArcGIS Pro to illustrate the locations of the study districts and commercial poultry farms (Fig. 1).

Fig. 1
figure 1

Geospatial distribution of study districts and sampled commercial poultry farms (broiler, layer, and Sonali) in Bangladesh.

Data collection

Data were collected through in-person interviews using a semi-structured questionnaire (pre-tested) designed to assess farm level biosecurity management practices in commercial chicken farms. The questionnaire was prepared based on the “Biosecurity guidelines of the Department of Livestock Services (DLS), Bangladesh, 2010 for the commercial poultry industry” 21. In addition to biosecurity measures, participants were asked whether their farms had experienced any HPAI outbreaks in the current or previous production cycle. All interviews were conducted only after obtaining informed consent from the farm owners. The research was carried out by trained field researchers who had completed formal ethical training in human subject research under the Collaborative Institutional Training Initiative (CITI Program). The training included modules on public health research, social and behavioral responsible conduct of research, biomedical responsible conduct of research, and social and behavioral research best practices for clinical research. For participants who were illiterate (had no formal education but were able to sign), the study objectives, procedures, and confidentiality assurances were verbally explained in the local language by trained field staff. After confirming understanding, informed consent was obtained from their guardian or legally authorized representative, with documentation provided through a signature or thumbprint in the presence of a witness. Field biosafety and hygiene protocols, such as shoe disinfection and hand sanitization, were strictly followed before entering any farm premises to minimize the risk of cross-farm contamination and ensure safe data collection procedures.

Data management and statistical analysis

The data collected were stored in Microsoft Excel for further analysis. The percentage distribution of farm and farmer’s characteristics with biosecurity was calculated in a univariate table. We used the Chi-square test to see if there was any variation in the maintenance of the individual rules of the biosecurity guideline among different poultry farms (broiler, layer, Sonali). To find the risk factors associated with previous outbreaks, all the factors were included in the univariate analysis where the outcome variable was “AI outbreak in past years on farms” and a chi-square test was performed to find the association between each independent and outcome variable. All the variables with p-values less than 0.05 were further included in the regression model. We checked all explanatory variables for collinearity before the regression model. Cramers’ V value between two factors greater than 0.4 was considered collinear and those variables were omitted from the final model (Supplementary Figure S1). We performed a generalized linear model (GLM) to find associated risk factors of biosecurity practices for occurring HPAI outbreaks on the farms. We checked the model for multicollinearity among the independent variables and conducted post-hoc pairwise comparisons to get the marginal means of significant variables using the “emmeans” package in R studio version 2024.09.0 31.

Results

Overview of farm and farmer demographics

The overall distribution of farmers’ characteristics and variation among different poultry farm types are presented in Table 1. The majority of farmers (94.6%) were male, 76.1% had secondary or higher secondary education, and 45.9% of the farmers had less than five years of farming experience. However, layer farmers had significantly more experience, with 42.7% having over five years (p < 0.001). 86.1% of farmers did not have formal training.

About half of the farms (51.1%) were between one and five years old, and 44.7%-layer farms were 5–10 year old (p < 0.001). Concrete floors were most common in layer farms (72.8%), while no layer chickens were raised on muddy floors. 94.7% of broiler farms were small-scale commercial farms (< 2000 birds), while approximately 33% of layer and Sonali farms were medium scale (> 2000 birds). Rearing systems also differed significantly (p < 0.001) since all broiler chickens were floor-reared, whereas 75.7% of layer chickens were housed in cages (Table 1).

Table 1 Distribution of farmer demographics and farm structure across different poultry farm types.

Assessment of factors related to biosecurity guidelines

Evaluation of compliance with biosecurity guideline principle 1 (Isolation)

According to the DLS biosecurity guidelines, the first principle emphasizes physical separation and limiting contact between farms and potential sources of contamination. Overall 93.7% of farms were located near residential areas, and 68.0% had a water body within 200 m. Only one third of the farms of all types had protective fences around their farms (Table 2). More than half of the farms (56.8%) allowed vehicle access to the premises and workers from 28.7% of farms reported visiting other farms. Only 34.7% of farms used separate cloths and footwear in their farms. Rodent access was reported in 73.4% of farms, with significant variation across farm types (p = 0.001). Wild birds had access to 76.7% of farms, and 45.3% of farms had backyard chicken access. Additionally, dogs and cats had access to 56.5% of farms, other domestic animals in 38.1%, and ducks in 31.7%. Workers of 25.7% of farms used to live inside the shed at night and workers from 13.3% of farms used to work on other farms.

Table 2 Maintenance of biosecurity guideline principle 1 (Isolation) across different poultry farm types.

Evaluation of biosecurity guideline principle 2 (Good farm hygiene)

This principle focuses on maintaining proper farm hygiene to lessen or get rid of the amounts of infectious agents in the environment. About 51.1% of farms cleaned and disinfected their surrounding areas, 59.8% disinfected the inside of poultry sheds, and 56.8% cleaned utensils during the rearing period (Table 3).

Regarding personal hygiene, among the farms surveyed, 63.1% practiced handwashing before each feeding, while 79.5% used to handwash after feeding. However, the use of disinfectant containing footbaths at farm entrances was low, with only 6.6% of farms following this measure.

Table 3 Maintenance of biosecurity guideline principle 2 (Good farm hygiene) across different poultry farm types:

Evaluation of biosecurity guideline principle 3 (Flock health care and monitoring)

This section is about flock health care and monitoring, which is necessary for early detection of disease. The veterinarian should be called immediately when birds appear sick or in an unusual condition and 74.6% of farms followed the rule (Table 4). There was significant difference in this practice among different poultry types. 64.0% followed the recommended practice of isolating sick birds. Only 11.8% of farms vaccinated their birds for agents that had previously caused problems. This practice varied significantly across farm types, with 2.7% of broiler farms, 33.0% of layer farms, and 1.7% of Sonali farms vaccinating their birds. Multiple antibiotics were used simultaneously at 41.7% of farms, while 32.9% of farms kept expired antibiotics for further use.

Table 4 Maintenance of biosecurity guideline principle 3 (Flock health care and monitoring) across different poultry farm types.

Evaluation of biosecurity guideline principle 4 (Good farm management strategies)

This principle addresses ventilation, waste management, and general husbandry practices. The rule that poultry buildings must be ventilated in accordance with breeding company instructions was followed by 77.0% of broiler farms, 59.2% of layer farms and 44.3% of Sonali farms (Table 5). Another guidance was birds that are dead, or sick should be sent in a safe and appropriate manner to the laboratory for analysis, according to the veterinarian’s instructions. 46.0% of broiler farms, 62.1% of layer farms, and 57.4% of sonali farms followed this practice. 78.9% of farms maintained the rule of downtown time being more than 14 days where 16.9% did not follow. 36.0% of people adhered to another rule regarding handling litter appropriately, which states that litter should be covered.

Table 5 Maintenance of biosecurity guideline principle 4 (Good farm management practices) across different poultry farm types.

Risk factors associated with AI outbreaks in poultry farms

Overall, 51 (15.4%, N = 331) farms reported a previous laboratory confirmed AI outbreak in their farm. The highest number of outbreaks were in layer farms (29.1%), followed by broiler (10.6%), and Sonali farms (7.8%). Additionally, there was no significant difference in outbreak occurrence between rural and peri-urban landscapes (Fig. 2).

Fig. 2
figure 2

AI outbreak prevalence (%) with 95% confidence intervals, stratified by (A) poultry farm type and (B) outbreaks by landscape (rural vs. peri-urban).

The chi-square test indicated that AI outbreak history was associated with working person in the farms, age of the farms, rearing system, faming experience, maintaining distance between sheds, visiting nearby farms, wild bird access to farm, veterinarians visit in the farm, AI outbreak history in nearby farm, vaccination based on past problem, and litter covering (Supplementary Table S1). Other biosecurity practices like using PPE, isolating sick birds, cleaning and disinfecting around farms, footbaths in the entrance were not significant at 5% level but at higher significance level and these factors cannot be ignored in risk assessment. The multivariable model revealed the adjusted associations between AI outbreak history and key factors that were significant in the chi-square test and mutually uncorrelated. Farms located near other farms with a history of AI outbreak had 6.64 times higher odds of reporting their own AI outbreak history (Fig. 3). While comparing the marginal means of nearby farms without an outbreak history, the predicted probability decreased from 42% to 10% (Fig. 4). Veterinarians visit the farm was associated with an increase in the probability of AI outbreak history from 14% to 33%, likely reflecting reporting bias, as veterinarians typically visit farms where outbreaks have occurred. Farms where farmers or workers frequently visited other farms had 2.5 times higher odds of AI outbreak history. Farm management primarily by the owner or a family member was associated with 16% predicted outbreak probability. In contrast, when a worker or multiple people were involved in farm management, the predicted probability increased to 29%, with adjusted odds 2.12 times higher.

Fig. 3
figure 3

Forest plot of odds ratio of key factors influencing AI outbreaks in poultry farms. Asterisk signs indicate statistical significance of the factor. (* : p < 0.05, ***: p < 0.001)

Fig. 4
figure 4

Predicted probabilities (marginal means) of AI outbreak based on significant risk factors. Marginal means were adjusted for all other variables using GLM. Error bars represent 95% confidence intervals.

Discussions

We thoroughly evaluated biosecurity management practices along with farm structure of 331 commercial poultry farms of different regions of Bangladesh. To protect poultry from infectious disease outbreaks and ensure their health and welfare, biosecurity protocols are essential, as daily biosecurity policies established and implemented as part of best management practices help reduce the likelihood of disease spread. Since 2007, AI has shown both endemic and seasonal epidemic patterns in Bangladesh. This persistence is driven by interactions among commercial poultry, backyard poultry, wild birds and widely breaches of standard biosecurity measures. The AI spreads through direct bird to bird contact, contaminated surfaces, and shared farm resources, all of which have been identified as risk factors influencing AI outbreaks in commercial poultry farms across Bangladesh. This study findings highlight significant biosecurity gaps, particularly in farm management, hygiene, and disease monitoring, which contribute to the viral transmission and increased outbreak risk. This study is one of the few conducted in Bangladesh that quantitatively links specific biosecurity practices with AI outbreak risks using mixed-effects modeling and marginal mean predictions.

Previous research has underlined that demographic and operational farm characteristics can influence biosecurity practices and, in turn, affect the risk of AI outbreaks. In this study, we found that a lack of formal training and limited farming experience were associated with poor biosecurity awareness, contributing to AI outbreaks which is aligned with the earlier study32. Most farms, particularly broiler farms, were comparatively new with small flock sizes, potentially underestimation the necessity of training on farm biosecurity management, which could result in lapses in biosecurity measures and increasing AI risk33,34. Although concrete floors are an indicator of good biosecurity, a significant number of broiler and Sonali farms still had muddy flooring, which may increase the risk of AI transmission, given the higher susceptibility of these poultry types35. This study demonstrated that farms with muddy floors were more prone to AI outbreaks, reinforcing evidence that flooring is a critical structural biosecurity element. This is also consistent with earlier studies stating that farms having a muddy floors rather than concrete floors, heighten the odds of AI infection36.

Environmental factors also emerged as major contributors to outbreak risk. This study revealed that almost all farms had residential areas and waterbody nearby, posing significant biosecurity risks. Such proximity not only increases the risk of AI transmission but also contributes to environmental contamination through air, water, and land, creating health risks for both humans and animals23,37,38,39. Almost two-thirds of the farms did not have any protective fence, similar to findings in other studies24,25 which significantly threatens biosecurity, as fencing is the first line of defense against disease40. This study revealed inadequacy in critical biosecurity components including fencing, unrestricted movement of workers and vehicles, and improper disinfection, all of which elevated AI risk. This aligns with a recent study that demonstrated poor biosecurity measures can increase the AI outbreaks by 30% to 79%, depending on the specific parameters41.

Though most of the farms were near waterbodies with inappropriate fencing, those farms are more prone to exposure to wild birds. This is complemented by an earlier study found that farms with unrestricted wild bird access were more prone to AI outbreak42. AI can also spread through aerosols and droplets, so farms located near infected premises are at higher risk, especially in windy or high-traffic areas43. The marginal mean plot estimation in this study revealed similar findings like proximity to an infected farm increased the risk of experiencing AI by 28%, consistent with previous studies44. Farms with access to domestic chickens and ducks were more prone to AI outbreaks23, aligning with this study findings. If nearby farms share water sources, feed suppliers, or other environmental conditions, the virus can spread through these shared pathways45. This emphasizes the role of environmental biosecurity and surveillance not only within farms but also in their immediate surroundings. Moreover, unrestricted vehicle movements, with vehicles frequently in contact with neighboring farms, increased contamination risk and AI transmission46,47. Therefore, restricting direct and indirect contact between farm birds is recommended to reduce outbreak risks39.

This study also revealed some other on-farm biosecurity lapses. Many farms with multiple sheds did not maintain the recommended minimum distance, increasing disease transmission risk. Cleaning and disinfection practices were inconsistently applied. While personal hygiene practices were relatively better, a major gap remained in entry-level hygiene: nearly 90% of farms lacked footbaths at the entrance, hindering pathogen control28.

Layer farm owners were more likely to consult veterinarians, yet overall vaccination coverage for known risk agents remained low48,49, with farm management largely responsible for decision- making. Farms managed by the owners were better positioned to implement timely disease control, while worker-managed farms often lacked adequate training facility and attorney for decision-making. This study calculated a 10% increase in predicted outbreak probability among farms not managed by owners. This underscores the importance of empowering farm owners through training and incentivizing biosecurity compliance. Additionally, over one-third of farmers did not follow quarantine protocols for sick birds or the all-in, all-out system. While this practice was better in broiler farms, it was lacking in layer farms, heightening transmission risk. Layer farms reported more outbreaks, likely due to their longer production cycles and prolonged flock presence, increasing cumulative exposure50. Longer flock durations without adequate monitoring and quarantine further compound this risk, suggesting tailored biosecurity plans for layer farms are needed.

Overall waste management practices were inadequate. Many farms failed to safely dispose of poultry waste, increasing AI risk, consistent with previous studies. As poor waste disposal harbors viral particles in the environment, targeted interventions in waste management must be prioritized. These findings point to the urgent need for targeted interventions to strengthen biosecurity at the farm level. Prioritizing farmer and worker training, regular monitoring, and promoting best practices particularly for high-risk farms like layer operations could be highly beneficial. The Department of Livestock Services and industry stakeholders should foster public–private partnerships to enhance biosecurity compliance and reduce future AI outbreak risks across Bangladesh’s poultry sector.

While this study has several limitations, the findings based on self-reported interviews which may be subject to recall bias and could not always be independently verified. The use of purposive sampling limits statistical generalizability, although it enabled us to capture a wide diversity of production systems (broiler, Sonali, and layer) and management conditions across multiple regions. Despite these limitations, the study provides actionable insights into critical gaps biosecurity practices and their association in AI outbreak risk. These findings highlight priority areas for intervention and offer a foundation for future longitudinal and intervention-based research to validate and strengthen risk-reduction strategies.

Conclusion

Commercial poultry farms in Bangladesh should strengthen compliance with the four core biosecurity principles recommended by the government to protect both poultry and public health. Although, some good practices were observed including locating farms away from LBMs, restricting worker visits to other farms, and limiting overnight stays are practiced, critical gaps remain. This study indicates weaknesses in structural and operational measures, particularly farms with poor hygiene, inadequate disinfection, unrestricted farm access, and limited disease monitoring, all of which strongly predict AI outbreaks. Strengthening biosecurity measures across all poultry farming systems will require stricter farm access control, regular use of protective gear and disinfection, improve waste management, and vaccination where appropriate. Equally important are farmer training, regular government monitoring, and awareness campaigns to help further reduce outbreak risks. These efforts should be closely align with Bangladesh’s One Health strategy to protect the poultry sector, reduce zoonotic risks, and safeguard public health.