HPAI Outbreak: Species Comparison

Phase 1: Do transmission patterns differ between chicken and duck?

Published

February 16, 2026

Overview

The Phase 1 outbreak involves 103 confirmed cases across two species. This analysis examines whether transmission patterns differ between chicken and duck farms.

Table 1: Outbreak summary by species
Species Cases Farms at risk Attack rate (%) First confirmed Last confirmed Detection methods
Chicken 81 4435 1.83 Dec 22 Jan 13 contact_tracing, passive, preshipment
Duck 22 4725 0.47 Jan 02 Jan 13 passive

Chicken cases (81) outnumber duck cases (22) nearly 4:1. The chicken epidemic began 11 days before the first duck case was confirmed. Duck cases are detected exclusively through passive surveillance, while chicken cases also benefit from preshipment testing and contact tracing.

Epidemic curves by species

Figure 1: Daily and cumulative incidence by species

The chicken epidemic shows sustained growth from late December, while duck cases emerge from early January with a slower accumulation rate. Both species show accelerating case counts in the final days of Phase 1.

Between-farm growth rate by species

Figure 2: Daily growth rate of cumulative cases by species (7-day rolling mean)
Table 2: Growth rate summary by species (during period with ≥3 cumulative cases)
Species Mean r(t) Median r(t) Doubling time (days) Period
Chicken 0.22 0.203 3.2 Dec 25 – Jan 13
Duck 0.24 0.203 2.9 Jan 04 – Jan 13

Within-farm mortality growth

The mortality ledger data provides daily death counts for three farms: two chicken broiler farms (2395, 3013) and one duck conventional farm (8120). This allows comparison of within-farm disease progression by species.

Figure 3: Daily mortality by farm, coloured by species
Figure 4: Daily mortality normalised by farm capacity (%)
Table 3: Within-farm mortality escalation summary
Farm ID Species Capacity Escalation start Peak date Peak deaths/day Peak (% capacity) Growth rate (r) Doubling time (days)
2395 Chicken 2139 Jan 03 Jan 08 300 14.0 0.87 0.8
3013 Chicken 2197 Jan 09 Jan 14 300 13.7 0.64 1.1
8120 Duck 13291 Dec 14 Jan 09 1000 7.5 0.20 3.5
Figure 5: Mortality escalation aligned by onset, normalised by capacity

Spatial distribution by species

Figure 6: Spatial distribution of confirmed cases by species
Table 4: Median nearest-neighbour distance between cases (km)
Species Same-species NN (km) Cross-species NN (km) Cases
Chicken 4.2 5.6 81
Duck 7.5 3.7 22

Detection and response by species

Table 5: Detection method by species
Detection Method Chicken Duck
Contact Tracing 1 0
Passive 74 22
Preshipment 6 0
Figure 7: Time from suspicion to confirmation by species
Table 6: Preventive culling by species
Species Production Preventive culls
Chicken Broiler 1 8
Chicken Broiler 2 10
Chicken Layer 1
Duck Conventional 25
Duck Organic 8

Duck farms account for 33 of 52 preventive culls (63%), despite contributing only 22 of 103 confirmed cases (21%). This suggests a more precautionary approach to duck farm culling.

Movement network

Animal movement records are available exclusively for the chicken supply chain: all 7,187 recorded movements are from broiler_1 (grower) to broiler_2 (finisher) farms. No duck movements are recorded.

Table 7: Movement network summary (chicken only)
Metric Value
Total movements 7,187
Unique source farms (broiler_1) 1,248
Unique destination farms (broiler_2) 2,746
Movements from case farms 88
Movements to case farms 165
Direct case-to-case movements 3

The chicken movement network connects broiler_1 farms (source) to broiler_2 farms (destination). This supply chain may explain the disproportionately high attack rate in broiler_2 farms (1.8% overall for chicken vs 0.5% for duck) and the earlier onset of the chicken epidemic.

The absence of recorded duck movements suggests duck-to-duck transmission occurs through different pathways — potentially environmental contamination, shared water sources, or wild bird contact rather than commercial animal movements.

Summary

Key differences between species:

  • Timing: The chicken epidemic began 11 days before duck cases appeared, suggesting either earlier introduction or faster initial spread in the chicken population.

  • Scale: Chicken farms have higher attack rates (1.8% vs 0.5%) despite a larger population at risk (4,435 vs 4,725 farms).

  • Transmission pathway: Chicken cases are linked by a commercial movement network (broiler_1→broiler_2), providing a clear mechanistic pathway for between-farm spread. Duck cases lack recorded movements, pointing to alternative transmission routes.

  • Detection: Chicken cases benefit from multiple detection methods (passive, preshipment, contact tracing), while all duck cases are detected passively. This may affect the apparent timing and completeness of case ascertainment in each species.

  • Response: Preventive culling is applied disproportionately to duck farms, possibly reflecting geographic proximity to outbreaks or different risk assessments for duck production systems.

Limitations

  • The mortality ledger data covers only 3 farms (2 chicken, 1 duck), limiting within-farm growth comparisons.
  • The absence of duck movement data does not necessarily mean ducks are not moved — movements may occur through unrecorded channels.
  • Different detection methods between species may create ascertainment bias: earlier/more complete detection in chickens could inflate apparent attack rates relative to ducks.
  • Growth rate estimates are sensitive to the small number of duck cases, particularly in the early epidemic period.