Semester 3 Practice Questions In Epidemiology

Explanation
Relative risk (RR) = Risk in exposed/Risk in unexposed
And, Risk = Number of cases (d)/ Total number of people at risk (N)

Therefore, RR for developing gastritis from herring consumption = Risk in ‘Ate’/ Risk in ‘Did not eat’

= (200/1000)/(100/1000)
= 2.0
NOTE: The risk can be calculated here because the total number of people who were at risk for falling ill is known. If the total number of people at risk of falling ill – say in a situation where a certain food chain is contaminated – then one cannot calculate risk but the odds of exposure is calculated instead (which should approximate the risk).

Explanation
Only odds can be calculated as a measure of incidence from case-control studies. In a case control study one is not sure about all the persons who had the particular exposure at the time in question and therefore the total number of persons at risk is unknown. Likewise, the duration of exposure to the risk factor in most instances and therefore rates cannot be calculated. The odds however gives a value that is numerically close to the risk.

Explanation
The crude mortality rate it the death rate that is calculated using all deaths as the numerator and the total population as the denominator. Therefore
Crude mortality rate = total deaths/total population
= 1000/100,000
= 10 per 1,000
NOTE: Age-adjusted mortality rates are reported when population mortality statistics are calculated. This means that the mortality rate has been standardized against a ‘standard’ population to account for any differences in age-group distribution in the population from year to year.

Explanation
The two groups where selected on the basis of the exposure (radium) and ‘followed’ over time (1945 to 1975) to see what the outcome experience (bone cancer) was. In fact because the investigators when back in time to establish the groups this study could also be called a retrospective cohort study (as compared to the usual prospective cohort study in which the groups are established in real time and followed up into the future.

Explanation
Randomization (or random allocation) is the method by which chance is used to allocate study participants to study groups. Random selection it the method by which chance (e.g. rolling dice or using a table of random numbers) is used to select participants for inclusion in a study. Blinding is where the participants are not aware whether they are receiving the intervention under investigation, current treatment or placebo. Cross-over is where one group in a cohort receives the intervention under study while the other receives current best treatment or placebo and then (following a wash out period) the groups switch around so that the group which was receiving placebo or current treatment now receives the intervention under evaluation. Factorial design is where a cohort study is set up to test more than one hypothesis at the same time so that the study group is randomized first to test one hypothesis and then those groups are randomized another time to create groups to test a second hypothesis.

Explanation
Sensitivity = (# correctly positive on screening test / # positive on gold standard)x 100
= (100/400) x 100
= 25%
NOTE: Validity (sensitivity, specificity etc.) questions are best answered by first setting up a 2x2 table before attempting to answer the question even if the answer seems to be obvious (so as to avoid simple errors). Make sure that your totals in the table are correct before you proceed with the calculation.

Explanation
You need to know which is the most likely pathogen and therefore you need to know what are the incubation periods for organisms which cause urethritis. The latent and infectious periods would be useful to know when trying to establish whether any other individuals may have been infected by this patient.

Explanation
The unadjusted odds ratio has a 95% confidence interval that does not include 1 and the adjusted odds ratio is associated with a P-value of