You must be logged in to vote!
poisonivy
totally agree, I dont understand why the right answer is Case control since that is not experimental
+2
howdywhat
am I subject to this kind of poor wording for the day of the exam?
+1
joeschmo
Case-control studies are largely retrospective by nature. You can retrospectively look back at samples and ask "what happened?"
+
You must be logged in to vote!
sherry
I agree. I was hesitating between the two choices. I still think cohort study is better regarding the "risk". I hope this kind of questions wont pop out on the real thing.
+3
soph
I think key here was they were measuring risk though
+1
yex
I also chose cohort, since it is comparing a given exposure.
+2
raspberryslushy
I was also thinking retrospective cohort study - just as time efficient, can look at risk, and the Q stem said the cancer was common, and I think of case-control for rare conditions. It's like they forgot a cohort study could be retrospective.
+2
boostcap23
The classic example they always give for why not to do retrospective cohort is because patients who have whatever disease your testing for are more likely to remember all their risk factor exposures than a normal person that doesn't have any disease. Of course in this case I'm sure the people running the study would be the ones who figure out how much arsenic was in the water but this also would be very time consuming to figure out for each individual person in the study. Thus a case-control study where you look at a group of people with >50 arsenic exposure and a group <5 arsenic exposure and simply see who has cancer and who doesn't would be easier and take less time.
+2
hunter_dr
According to the B&B and UWORLD, if question stem measures the exposure first, then it have to be a cohort study and you will do relative risk calculation, which is why i picked that as an answer. However, I do see 'Cases' and 'Control' both in the questions so maybe it is case-control study..
Bottom line is----WASTED MY $60
+
You must be logged in to vote!
You must be logged in to vote!
an_improved_me
I feel like you just described a prospective cohort? Find ppl with the disease (same populatin and outcome), and then see if they had a similar risk factor; then follow them to see if they had a risk factor.
Case-control would be: Have two groups of people, some with the Dx, others without. See if there is a difference in proportion that have/don't have a risk factor
Someone please correct me if i'm wrong?
+2
You must be logged in to vote!
kbizzitt
I believe what they are getting at with the "common cancer" is that it is easily definable as a case. This way you can easily determine who is a case and who would be a control. Then you go back and see if that case had exposure or not.
+1
You must be logged in to vote!
nbmehelp
If they were measuring risk shouldn't it be a cohort study though? By looking at first aid..
+6
270onstep1
They both can determine risk. Key here is the time efficiency of case-control studies when compared to cohort.
+1
suckitnbme
Case-control only determines odds ratio which is not calculating risk. In rare diseases the odds ratio can be used as an estimate of the risk ratio however.
+2
You must be logged in to vote!
drdoom
welcome, O great physician of the skull and oral cavity. we revere your intricate understandings of the face, jaw, maxilla and all their tiny and hidden foramina. teach us your ways.
+8
You must be logged in to vote!
submitted by โlsmarshall(465)
An experimental design or experimental study must have an intervention, by definition. Case-control studies are observational studies, not experimental. This question is technically incorrect. They wanted to amke a point that case-control studies are time and cost efficient since they don't require following patients over time or any resources besides reviewing/gathering information. Case series could not test this hypothesis.
Also, the wording "associated wit an increased risk" somewhat alludes to case-control studies only having the ability to find odds of an associations between exposure and outcome, but not establish causal relationship.