Can a cohort study have no control groups (is a clinical series study a cohort)?

#1
Hi

AFAK a cohort study needs a matched control group. However, its definition doesn't necessitate a control group.

Moreover, I have read some publications in which the authors had investigated the complications of some treatments, for example rhinoplasty, during a period of time. They did not have any control groups, but the article said "this study was a cohort one".

I had a similar study, but I named it as a "case series" (= "clinical series"), which is defined as evaluating a cohort (without a control group) over time.

I just wanted to know if a clinical series is a cohort? Can I cite my study as cohort?

Thanks a lot in advance.
 
#4
It is still a cohort according to the original cohort study definition. But this is not the cohort you have in mind. You can't compute any RR or check for risk factors. The results have to be presented just like a cross sectional study.
 

Link

Ninja say what!?!
#5
You should remember that a cohort study is just one where a group of people are followed over time. Some examples are prospective, retrospective, and random cohorts. Having a cohort does not necessitate a control group.

Many researchers like to match control groups to cohorts in order to get a good comparison. Though this is reasonable, something to keep in mind is that the matched groups will almost always come from a different environment. So many of the things that your cohort would be exposed to, the matched population might not be.

Also, you still CAN get RR's and check for risk factors in cohort studies. Researchers do it all the time. The disclaimer is that you have to do it against a chosen reference group in your cohort. This is one of the main reasons why researchers always include what I call a "Table 1" in their papers (a table describing their study population). So that readers will see what their population is made up of and how they split off their reference group.

HTH
 
#6
Also, you still CAN get RR's and check for risk factors in cohort studies. Researchers do it all the time. The disclaimer is that you have to do it against a chosen reference group in your cohort. This is one of the main reasons why researchers always include what I call a "Table 1" in their papers (a table describing their study population). So that readers will see what their population is made up of and how they split off their reference group.
HTH

Thanks Link.
Could you please tell me how can I choose the reference group in my cohort? I have two such studies at hand and the statisticians failed to address this and I am interested in what you mentioned. They just assessed the changes in the cohort over time.