Volume 21, Number 10—October 2015
Research
Effect of Live Poultry Market Closure on Avian Influenza A(H7N9) Virus Activity in Guangzhou, China, 2014
Table 2
Estimated effect of market closure and contaminated environmental sites on AIV and influenza A(H7N9) virus detection in 5 poultry markets under enhanced surveillance, Guangzhou, Guangdong Province, China, 2014*
Variable |
Retail LPMs, aOR (95% CI)† |
Wholesale LPM, aOR (95% CI) |
DPM, aOR (95% CI) |
|||||||
rRT-PCR |
H7N9 culture |
rRT-PCR |
H7N9 culture |
rRT-PCR |
||||||
AIV |
H7N9 |
AIV |
H7N9 |
AIV |
H7N9 |
|||||
Period | ||||||||||
Before market closure | Ref | Ref | Ref | Ref | –‡ | –‡ | Ref | Ref | ||
During market closure | 0.25 (0.16–0.39) | 0.21 (0.12–0.36) | 0.08 (0.02–0.42) | 1.60 (0.52–4.90) | 0.22 (0.10–0.50) | 0.11 (0.01–0.89) | 0.30 (0.09–0.98) | 0.68 (0.12–3.89) | ||
After market
closure |
1.78
(1.20–2.63) |
0.58
(0.35–0.95) |
0.73
(0.27–1.98) |
10.3
(3.52–30.3) |
Ref |
Ref |
5.27
(1.97–14.1) |
3.32
(0.68–16.1) |
||
Environmental samples tested | ||||||||||
Poultry cage | Ref | Ref | Ref | Ref | Ref | § | – | – | ||
Defeathering machine | 1.15 (0.61–2.14) | 1.66 (0.74–3.70) | 1.25 (0.20–7.87) | 2.49 (1.09–5.68) | 1.21 (0.40–3.65) | § | – | – | ||
Chopping board | 2.64 (1.60–4.37) | 2.12 (1.06–4.26) | 3.52 (0.88–14.0) | – | – | – | 0.56 (0.22–1.41) | 3.18 (0.98–10.3) | ||
Processing table | 1.16 (0.73–1.85) | 1.15 (0.59–2.25) | 1.09 (0.26–4.67) | – | – | – | Ref | Ref | ||
Bucket holding poultry meat | – | – | – | 0.97 (0.38–2.44) | 0.17 (0.02–1.40) | § | – | – | ||
Wastewater | 1.60 (0.95–2.67) | 1.23 (0.58–2.62) | 1.41 (0.28–7.14) | 1.38 (0.70–2.73) | 0.91 (0.37–2.22) | § | 1.15 (0.44–3.06) | 1.16 (0.31–4.36) | ||
Drinking water | – | – | – | 2.02 (0.44–9.38) | 2.32 (0.40–13.4) | § | – | – |
*AIV, avian influenza virus; aOR, adjusted odds ratio; DPM, dressed poultry market; LPM, live poultry market; ref, reference; rRT-PCR, real-time reverse transcription PCR; –, no samples tested.
†Also adjusted for potential market differences for the 3 retail markets.
‡No influenza A(H7N9) virus was detected before market closure in wholesale markets, and data from this period were excluded from the regression model.
§There were too few H7N9 virus–positive samples by culture in contaminated environmental sites in wholesale markets and DPM overall for us to estimate the effects. A simplified model was used for wholesale markets.
1These first authors contributed equally to this article.
2These senior authors contributed equally to this article.