Scribd
Upload
120 views2 pages

Sample Quest 13

The clinicians develop a hunch that a specific batch of marijuana may be responsible for students presenting with a severe cough and bloody phlegm based on interviews with early cases. This initial observation is classified as a case series design. To further investigate, the clinicians interview other students about marijuana use and symptoms, compiling the data into a 2x2 table. This resembles a case-control design as it compares cases (students with bloody phlegm) to controls (students without). The appropriate measure of association for this case-control design is the odds ratio. Calculating the odds ratio from the data in the 2x2 table results in a value closest to 36. The 95% confidence interval for this value

Uploaded by

Hundera Terefe
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
120 views2 pages

Sample Quest 13

The clinicians develop a hunch that a specific batch of marijuana may be responsible for students presenting with a severe cough and bloody phlegm based on interviews with early cases. This initial observation is classified as a case series design. To further investigate, the clinicians interview other students about marijuana use and symptoms, compiling the data into a 2x2 table. This resembles a case-control design as it compares cases (students with bloody phlegm) to controls (students without). The appropriate measure of association for this case-control design is the odds ratio. Calculating the odds ratio from the data in the 2x2 table results in a value closest to 36. The 95% confidence interval for this value

Uploaded by

Hundera Terefe
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Sample exam questions

Scenario 1: On Monday morning, students begin arriving at Health Services with severe
cough that includes bloody phlegm. The physicians and nurses interview some of the
early cases and find out that many of them mention attending a party at which a new (to
them) batch of marijuana was smoked. A prime source of epidemiological hypotheses is
‘clinical hunches’, of which this is an example – the clinicians develop the hunch that a
specific batch of marijuana may be responsible for this previously unrecognized
respiratory illness.

1. What is the name of the research design from which this hunch was derived?
a. Case control
b. Cohort (Retrospective)
c. Cohort (Prospective)
d. RCT
e. Case series

Answer: e. Case series. Why? The observation was based on a single series of cases.
There are no controls so it can’t be a case control study (or any of the other designs, all of
which are controlled designs). Course reference(s): Gordis discusses case series
(chapter 10, page 177), although he doesn't use the name.. Its role in generating
hypotheses is demonstrated by two n of 10 SARS papers that came out in the New
England Journal of Medicine early in the epidemic.

1a. This knowledge could also be tested using a short answer question. I would
probably have it worth two points, and ask:
What is the name of this design? __[case series]_________________
Give one reason to support your answer.___[it has no controls and the other answers are
all controlled designs]___

The clinicians immediately start asking other students attending clinic without the bloody
phlegm whether they were at the same party, whether they had recently smoked
marijuana, and if so, whether it was a new batch to them. They assemble the data into the
following table:

Smoked new batch Bloody phlegm + Other reason for visit


Yes 9 2
No 1 8

2. What epidemiological design does this most closely resemble?


a. Case control
b. Cohort (Retrospective)
c. Cohort (Prospective)
d. RCT
e. Case series
Answer: a. Case control. Why? The study was conducted using two samples: those with
the outcome (Cases) and those without (controls). Because it is now a controlled study, it
can’t be a case series. Because there is no randomization, it can’t be an RCT. Because
subjects were sampled on the basis of outcome (bloody phlegm +/-), it can’t be a cohort
study because cohorts are assembled on the basis of exposure.

3. Epidemiologists calculate the measure of association most appropriate for the


above design, which is:
a. Risk ratio
b. Odds ratio
c. Rate ratio
d. Pearson’s Product Moment Correlation Coefficient
e. Spearman’s rho

Answer: b. Odds ratio. Risk and rate ratios can usually not be calculated with case
control designs. Answers d. and e. are incorrect because they aren’t appropriate for two
dichotomous variables.

4. Calculate the value of the measure you chose in 3. Its value is closest to:
a. 0.03
b. 1.0
c. 7.4
d. 9.9
e. 36.0

Answer: e. (a x d) / (b x c) = (9 x 8) / (1 x 2) = 72 / 2 = 36.0 If you got c, you calculated


the Risk Ratio. If you got a., you have taken the reciprocal (2 / 72), possibly because you
substituted the cell diagonals (a x d) and (c x d) for each other.

5. The exact 95% confidence interval for the value from 4. (from OpenEpi) is (2.06
– 1766.95). Is the above value ‘statistically significant’?
a. Yes, because the lower bound excludes the null value.
b. Maybe, but we can’t be sure without seeing the p value.
c. No, but it might be clinically significant
d. No, nor is it clinically significant
e. Insufficient information

Answer: a. The null hypothesis is that OR = 1.0. The 95% CI excludes this value. Most
epidemiologists find confidence intervals of more use than simple p values because they
allow a test of the data against the null hypothesis, as well as indicating the precision of
the estimate. Therefore, b is wrong – we can be sure without seeing the p value. c. and
d. are both wrong because the value is statistically significant regardless of clinical
significance. e. is wrong because there is sufficient information for the calculation.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy