Quantitative risk assessment for African horse sickness in live horses exported from South Africa
Loading...
Date
Authors
Sergeant, Evan S.
Grewar, John Duncan
Weyer, Camilla Theresa
Guthrie, Alan John
Journal Title
Journal ISSN
Volume Title
Publisher
Public Library of Science
Abstract
African horse sickness (AHS) is a severe, often fatal, arbovirus infection of horses, transmitted
by Culicoides spp. midges. AHS occurs in most of sub-Saharan Africa and is a significant
impediment to export of live horses from infected countries, such as South Africa. A
stochastic risk model was developed to estimate the probability of exporting an undetected
AHS-infected horse through a vector protected pre-export quarantine facility, in accordance
with OIE recommendations for trade from an infected country. The model also allows for
additional risk management measures, including multiple PCR tests prior to and during preexport
quarantine and optionally during post-arrival quarantine, as well as for comparison of
risk associated with exports from a demonstrated low-risk area for AHS and an area where
AHS is endemic. If 1 million horses were exported from the low-risk area with no post-arrival
quarantine we estimate the median number of infected horses to be 5.4 (95% prediction
interval 0.5 to 41). This equates to an annual probability of 0.0016 (95% PI: 0.00015 to
0.012) assuming 300 horses exported per year. An additional PCR test while in vector-protected
post-arrival quarantine reduced these probabilities by approximately 12-fold. Probabilities
for horses exported from an area where AHS is endemic were approximately 15 to
17 times higher than for horses exported from the low-risk area under comparable scenarios.
The probability of undetected AHS infection in horses exported from an infected country
can be minimised by appropriate risk management measures. The final choice of risk management
measures depends on the level of risk acceptable to the importing country.
Introduction
Description
S1 Dataset. Simulation results for base scenarios.
S2 Dataset. Simulation results for comparison of high and low PCR sensitivity.
S3 Dataset. Simulation results for sensitivity analysis on breakdown of vector protection during quarantine.
S4 Dataset. Simulation results for sensitivity analysis on breakdown of vector protection during loading.
S1 Fig. Map of AHS Controlled Area of South Africa.
S2 Fig. Results of sensitivity analysis on the probability of breakdown of vector protection during pre-export quarantine. A, Low-risk area, Breakdown = gamma(24, 2191); B, Low-risk area, Breakdown = gamma(1, 5000); C, Low-risk area, Breakdown = gamma(1, 500); D, Lowrisk area, Breakdown = gamma(1, 50); E, Low-risk area, Breakdown = gamma(1, 10); F, Low-risk area, Breakdown = gamma(1, 5); G, Endemic area, Breakdown = gamma(24, 2191); H, Endemic area, Breakdown = gamma(1, 5000); I, Endemic area, Breakdown = gamma(1, 500); J, Endemic area, Breakdown = gamma(1, 50); K, Endemic area, Breakdown = gamma(1, 10); L, Endemic area, Breakdown = gamma(1, 5).
S3 Fig. Results of sensitivity analysis on the probability of breakdown of vector protection during loading. A, Low-risk area, LoadingBreakdown = gamma(1, 500); B, Low-risk area, LoadingBreakdown = gamma(1, 50); C, Low-risk area, LoadingBreakdown = gamma(1, 10); D, Low-risk area, LoadingBreakdown = gamma(1, 5); E, Endemic area, LoadingBreakdown = gamma(1, 500); F, Endemic area, LoadingBreakdown = gamma(1, 50); G, Endemic area, LoadingBreakdown = gamma(1, 10); H, Endemic area, LoadingBreakdown = gamma(1, 5).
S2 Dataset. Simulation results for comparison of high and low PCR sensitivity.
S3 Dataset. Simulation results for sensitivity analysis on breakdown of vector protection during quarantine.
S4 Dataset. Simulation results for sensitivity analysis on breakdown of vector protection during loading.
S1 Fig. Map of AHS Controlled Area of South Africa.
S2 Fig. Results of sensitivity analysis on the probability of breakdown of vector protection during pre-export quarantine. A, Low-risk area, Breakdown = gamma(24, 2191); B, Low-risk area, Breakdown = gamma(1, 5000); C, Low-risk area, Breakdown = gamma(1, 500); D, Lowrisk area, Breakdown = gamma(1, 50); E, Low-risk area, Breakdown = gamma(1, 10); F, Low-risk area, Breakdown = gamma(1, 5); G, Endemic area, Breakdown = gamma(24, 2191); H, Endemic area, Breakdown = gamma(1, 5000); I, Endemic area, Breakdown = gamma(1, 500); J, Endemic area, Breakdown = gamma(1, 50); K, Endemic area, Breakdown = gamma(1, 10); L, Endemic area, Breakdown = gamma(1, 5).
S3 Fig. Results of sensitivity analysis on the probability of breakdown of vector protection during loading. A, Low-risk area, LoadingBreakdown = gamma(1, 500); B, Low-risk area, LoadingBreakdown = gamma(1, 50); C, Low-risk area, LoadingBreakdown = gamma(1, 10); D, Low-risk area, LoadingBreakdown = gamma(1, 5); E, Endemic area, LoadingBreakdown = gamma(1, 500); F, Endemic area, LoadingBreakdown = gamma(1, 50); G, Endemic area, LoadingBreakdown = gamma(1, 10); H, Endemic area, LoadingBreakdown = gamma(1, 5).
Keywords
Risk management measures, Infections, Sub-Saharan Africa, African horse sickness, AHS
Sustainable Development Goals
Citation
Sergeant ES, Grewar JD, Weyer CT,
Guthrie AJ (2016) Quantitative Risk Assessment for
African Horse Sickness in Live Horses Exported from
South Africa. PLoS ONE 11(3): e0151757.
DOI: 10.1371/journal.pone.0151757.