Download stability of infection patterns with time, the case of infectious

STABILITY OF INFECTION PATTERNS WITH TIME, THE
CASE OF INFECTIOUS PANCREATIC NECROSIS VIRUS IN
SCOTTISH FARMED SALMON
Alexander G Murray and Rob S Raynard, FRS Marine Laboratory, Aberdeen AB11 9DB UK
Method
Is IPNV Persistent or Transient?
ƒ Infectious pancreatic necrosis virus is widespread and
increasing in Scottish salmon farms (90% marine sites)
ƒ Sort sample from each site into pairs (e.g. 1&2, 1&3
and 2&3)
ƒ Is a sample taken 1 or 2 years ago a good indicator of a
sites current infection status? Is infection persistent (or
repeated) or transient?
ƒ Divide pairs into two sets: I. earlier sample +ve or II.
earlier sample –ve
ƒ Sort lists by time T between earlier and later sample
If transient then:
ƒ the high prevalence indicates very high transmission
rates
ƒ samples rapidly lose their information value for infection
control
ƒ controls should be targeted above site level
ƒ Find probability of +ve versus T for list I
ƒ Multilevel logistic regression model used to account for
inter-regional (i) and inter-annual (j) variation
ƒ Repeat for list II
ƒ P(IT) = Logit(aij + bijIn(T+1))
A
X
B
A
0%
Time
1
Probability of positive result
Probability infected
100%
0.9
0.8
0.7
F+
0.6
>95% F-
0.5
0.4
0.3
0.2
0.1
<95% F+
F-
0
0
1
How to identify persistence
2
3
4
Years since earlier sample taken
Results from freshwater: incomplete convergence = some persistence
ƒ Ratio of conditional probabilities P(I+T|I+0)/P(I+T|I-0)
ƒ Theory A. Pathogen persists, or repeatedly infects the
same site, even after prolonged period sites that tested
+ve are more likely to still test +ve than are formerly -ve
sites
ƒ Theory B. Pathogen is transient an infects sites at
random, formerly -ve sites are just as likely as former
+ve sites to test +ve after T (converging on regional
average)
1
Probability of positive result
ƒ What is the probability that +ve site will test +ve, relative
to the probability that a site that tested -ve will test +ve
after a given time?
0.9
>95% M-
M+
0.8
0.7
0.6
0.5
<95% M+
0.4
M-
0.3
0.2
0.1
0
0
1
2
3
4
Years since earlier sample taken
Results from marine sites: infection convergence = dynamic
Conclusion
IPNV infection does not persist on marine sites for
periods of >2 years: Infection is transient. This may
reflect harvesting and fallowing cycle after 18
months. Some freshwater sites show response to
infection history several years later: infection is
persistent