3.3 Utility of animal health information systems
When asked “how frequently they use the animal health information
systems”, 0.93% of the respondents answered never, 8.88% rarely,
26.64% sometimes, 35.05% often and 28.50% always. There was no
association between the frequency of animal health information system
use and the type of information that they find useful in them (all types
of information were not significant in the Poisson model where frequency
of use was the outcome; p-value > 0.05).
The univariate Poisson regression model, using degree of usefulness as
the outcome variable and type of information (listed in questionnaireAppendix S1 ) as the explanatory variable (“treatment” being
the reference category) showed that there was a difference in the
usefulness of type of information available in the animal health
information systems (Figure 3). Overall the 5 most important
were “prevention measures” (extremely useful category with 42.52%),
“efficiency of currently available control measures” (extremely useful
category, 33.8%), “cases/incidence information” (extremely useful
category, 33.3%), “zoonotic potential” (extremely useful category
39.4%) and “evolution/spread of the disease during time (days, weeks,
months)” (extremely useful category, 33.3%). The 5 least useful ones
were “regulations which are currently in place regarding a specific
disease”, “methodologies for risk analysis which have been described
so far”, “produce risk estimation”, “pathogenesis of the disease”
and “treatment”, with 22.5%, 21.1%, 21.1%, 17.4% and 15%
respectively in the extremely useful category. The information
“methodologies for risk analysis which have been described so far”,
“produce risk estimation” and “pathogenesis of the disease” had the
same degree of usefulness as “treatment” given that they were not
statistically significant (p-value <0.05) in the Poisson
model.