Chapter 5

Question 1

Why can’t the piglet experiment be repeated in humans?

Answer 1

Firstly, gnotobiotics, which isn’t possible in humans. Then, would involve sampling patients multiple time per hour for no clear medical purpose. Even then, easiest thing to do would be to look at hospitalised patients anyway, not representative. In any case, might as well try to use data already available on hospital patients.

Question 2

Why is it important to understand within-host AMR diversity?

Answer 2

Since colonising strain most likely to infect, important to understand what’s there. Not clear how this works - eg most prevalent strain is most likely to infect? Hence importance of looking at relative prevalence changes over time. Can also help to understand instances where treatment might fail - eg if 10% of pop resistant, then give abx, enough for treatment to fail? Requires an initial understanding of what bacterial pop is made up of, and how this changes.

Question 3

Why did you use linear regression, Spearman’s, and Chi square?

Answer 3

• linear regression: tested assumption of a linear trend over time
• spearman’s: tests for monotonic relationship instead of linear like pearson’s, and no assumption of normally distributed values
• chi square: nonparametric, assumes mutually exclusive categories and expected sample size of at least 5 in each category

Question 4

What are the degrees of freedom in chi square test?

Answer 4

(number of rows-1)*(number of columns-1)

Question 5

Why the GOSH children hospital setting here?

Answer 5

Their database and access system. Lower MRSA in children than adults true, but maybe a bit higher S aureus colonisation.