PerspectiveIn the Journals Plus

Researchers identify risk factors for avian influenza at live bird markets

Show Citation

July 17, 2018

At live bird markets, the risk for avian influenza is higher among workers at large retail markets, women workers, and workers who clean, slaughter, defeather and eviscerate poultry, according to results of a recent meta-analysis.

Xiaoyan Zhou, MD, of the University of Queensland’s School of Veterinary Science, Australia, and colleagues said their findings may help improve infection prevention and control programs at live bird markets (LBMs), where avian influenza (AI) viruses can be maintained and transmitted over long periods of time, according to the researchers.

“After the emergence of [highly pathogenic avian influenza] H5N1 influenza in 2003, several studies have documented that LBMs could be sources of human AI infections,” the researchers wrote in The Journal of Infectious Diseases. “The importance of LBMs in the transmission of AI to humans was also highlighted by the emergence of influenza A (H7N9) viruses of low pathogenicity to poultry in early 2013, causing human infections without preceding or concomitant outbreaks in poultry. Exposure to H7N9-infected poultry at LBMs has been implicated as the main risk factor for human infection.”

Image of chickens
A recent meta-analysis examining risk factors associated with avian influenza at live bird markets could be used to inform decisions about infection control programs. 

Biosecurity practices have been implemented at LBMs to reduce the introduction and spread of AI infections among animal populations, Zhou and colleagues reported. These practices include limiting the number of poultry species that are sold at a single market, deploying sufficient cleaning and disinfection procedures and ensuring rest days, when markets are cleared and cleaned.

To determine the impact of biosecurity practices and better understand risk factors for AI, Zhou and colleagues conducted a meta-analysis of 79 studies published between 2003 and 2018. The studies — 25 in English and 54 Chinese — were conducted in China, Vietnam, Bangladesh, the United States, Egypt and Indonesia.

When examining market characteristics, the risk for AI infection was lower in LBMs:

  • small in size (OR = 0.55; 95% CI, 0.34-0.88);
  • selling only one poultry species (OR = 0.29; 95% CI, 0.11-0.76);
  • located in central city areas vs. noncentral city areas (OR = 0.74; 95% CI, 0.56-0.97);
  • with cleaning and disinfection protocols in place (OR 0 0.35; 95% CI, 0.17-0.73);
  • that separate different species (OR = 0.63; 95% CI, 0.43-0.90);
  • banning overnight storage (OR = 0.50; 95% CI, 0.29-0.86); and
  • that source poultry locally rather than from other areas (OR – 0.57; 95% CI, 0.35-0.94).

Zhou and colleagues also found that AI risk was lower;


  • after poultry had a day of rest vs. before a day of rest (OR = 0.20; 95% CI, 0.11-0.38);
  • in summer and autumn months (OR = 0.65; 95% CI, 0.44-0.96);
  • among male workers (OR = 0.68; 95% CI, 0.54-0.87);
  • in wholesale markets vs. retail markets (OR = 0.38; 95% CI, 0.22-0.65);
  • among workers who did not clean feed trays (OR = 0.34; 95% CI, 0.13-0.90);
  • among workers who were not in contact with ducks (OR = 0.28; 95% CI, 0.12-0.64);
  • among workers who did not slaughter poultry (OR = 0.12; 95% CI, 0.03-0.56);
  • among workers who did not defeather poultry (OR = 0.19; 95% CI, 0.07-0.51); and
  • among workers not involved in poultry evisceration (OR = 0.19; 95% CI, 0.07-0.52).

In light of these findings, the researchers noted that biosecurity practices should be targeted to LBMs that sell and slaughter poultry, as well as markets with multiple animal species. The results also highlight the need to focus surveillance efforts during winter and spring months.
– by Stephanie Viguers

Disclosures: The authors report no relevant financial disclosures.

itj+ Infographic

itj+ Perspective

Photo of Thomas Yuill

The study has clinical implications as they relate to two influenza strains that have been getting a lot of attention recently: influenza A(H5N1) and influenza A(H7N9). These strains are of concern clinically because they are associated with high case-fatality rates in people. Although A(H5N1) has produced fatalities in poultry, A(H7N9) has not. Therefore, epidemiologically, it makes it more more difficult to track A(H7N9) because it does not leave a trail of dead birds in its wake, giving a clue that something is going on.

The meta-analysis makes several interesting points, with statistical analyses to support them, about the risks associated with particular conditions related to live bird markets. Although not saying so explicitly, the researchers do not contend that the risk can be eliminated. Implicit in their studies is that live bird markets are going to continue. They are important economically and culturally in Asia and China, where there is a preference among many individuals to pick out birds in a live bird market rather than have them neatly prepared, slaughtered and packaged, as we are used to in the Western world. So, in practical terms, I do not think it is going to be possible to eliminate live bird markets. In the past with H7N9 virus, although not mentioned in this paper, health authorities have closed some markets for a fixed period because there were a number of human cases associated with transmission from the live bird markets. However, those reports also note that live bird markets cannot be permanently eliminated. The suggestions for reducing the risk for transmission in this study are practical. The question becomes, is there the political will given the economic and cultural forces to enable the implementation of the changes that would reduce the risk to people? That remains to be seen.

Thomas M. Yuill, PhD

ProMED viral diseases moderator
Professor emeritus, pathobiological science and forest and wildlife ecology University of Wisconsin-Madison

Disclosure: Yuill reports no relevant financial disclosures.