Veterinary Epidemiology Questions

Q1. This study estimates the prevalence of snake envenomation cases in dogs in NSW. There are three estimates identified in this paper (reported on p288); 0.96%, 0.54% and 0.31%.
a) Please explain why three estimates are given and why each of these estimates is useful for the reader
b) Why might the author have chosen to use prevalence rather than incidence as a measure of frequency of disease in this study?

Q2. A sample size calculation is reported in this paper.
a) Explain why the expected prevalence was included as 50% in this calculation
b) Why did the author decide to use 587 practices rather than the 112 – 232 that she estimated to be required through the sample size calculation?
c) If the author had based her sample size calculation on reports of snake envenomation from a different state, where an overall prevalence of snake envenomation of 10% occurred, what would the calculated sample size have been? (please show your working)

Q3. This paper reports the use of snake venom detection kits (SVDKs) for diagnosis of snake envenomation. SVDKs are highly specific, but not very sensitive for diagnosis of snake envenomation.
a) Please identify what this means in terms of false positives and false negatives for detection of snake envenomation using SVDKs in dogs.
b) In cases of snake envenomation why might you want to try to increase the sensitivity of a test for diagnosis as well? (hint: what is the cost of a false negative?)
c) How might you try to increase the sensitivity for diagnosing snake envenomation (describe in detail)?
d) What might be an issue associated with increasing the sensitivity of the test? (hint: what is the cost of a false positive?)

Q4. Describe a study that you might perform to follow up on this questionnaire study in order to validate some (could be one or more) of the findings reported in this questionnaire.