Transcript 17 tcp/rer/3402/srv/fejzic - Assistance to Western Balkan Countries

Nihad Fejzić
International consultant
Draft surveillance design
- farm accreditation -
 Fresh water fish raised in tanks
 Farms with 2 to 20 tanks
 Each tank 1000 to 5000 fish
 Disease
 Present in the country
 Low infectivity
 During winter months
 Adult fish
Draft surveillance design
- farm accreditation  One stage survey
 95% confidence that disease will be detected if present
 Free status + maintenance of biosecurity
 OIE standards
 α= 5% probability of rejecting true Ho
 Ho – disease is present on farm with prevalence 5%
Draft surveillance design
- farm accreditation  Target population - all fish on a farm
 Clustering of disease minimal
 Stratification by tank proportional to population of
tank
 Sampling of fish for antigen based ELISA (Se=98%
95%CI 96,7% -99,2%, Sp=99,4% 95%CI 99,2%-99,6%)
 Confirmatory test for (possibly false) positive result –
culture Se=90%, Sp=100%
Draft surveillance design
- farm accreditation -

Draft surveillance design
- farm accreditation  Sample size using ELISA – 410 fish per farm
 Sample size using two tests – 169 fish per farm
 Adjustment to population size:
Population size
1000
2000
Sample size
157
163
5000
10000
166
169
Sample size calculation done by FreeCalc www.epitools.ausvet.com.au
Draft surveillance design
- farm accreditation -
Draft surveillance design
- farm accreditation -
 Selection of samples – as much random as possible
 i.e. 4 tanks with 1850,
4250, 4270 and 4880 fish = 15250
12,13% 27,87% 28%
32%
=100%
21
47
47
54
=169
 Harvesting – systematic sample 1850/21 – every 88th fish (first
chosen randomly)
 Dip net – different location of the tank
Draft surveillance design
- national accreditation  Fresh water fish raised in cages
 Small number of farms/importers – 2302
registered
 Farms with 3 to 60 cages
 Disease





Absent from the country
Highly contagious
No clinical signs, high mortality
Common in late summer, but appears all year round
Growing faze
Draft surveillance design
- national accreditation  Two stage survey
 OIE standards
 α= 5% probability of rejecting true Ho
 Ho1 – disease is present in the cage with prevalence 50%
(of cages within the farm )

95% confidence that disease will be detected if present
 Ho2- disease is present on the farm with prevalence 1%
(of farms within the country)

80% confidence that disease will be detected if present
Draft surveillance design
- national accreditation  Target population - all cages with susceptible
species in the country
 Two stage sampling
 Establishments breeder farms /importers
 cages
 Three tests
 Owner observation of mortality (Se=87%, Sp=68% -
determined by special study)
 PCR for cages with mortality (up to 20 fish) Se=98,6,
Sp=99%
 Transmission test for PCR positive cages Se=95%,
Sp=100%
Draft surveillance design
- national accreditation -

Sample size calculation done by FreeCalc www.epitools.ausvet.com.au
Recomended resources
 OIE Aquatic animal health cone
http://www.oie.int/internationalstandard-setting/aquatic-code/accessonline/
 Principles for the Design and Conduct of
Surveys to show Presence or Absence of
Infectious Disease in Aquatic Animals
http://www.daff.gov.au/__data/assets/pdf
_file/0004/146713/surveillance.pdf
 Surveillance and zooning for aquatic