Observational Studies

Observational Studies

– [Voiceover] Hey guys, this
is Dr. Joel with MedImmersion. Your are watching a mini-lecture in the Flash Flood Review Series, where I review the most important topics for your board exams, in as little time as possible. If you like this video, if this kind of thing
is right up your alley, click on the lower link here for the Flash Flood Review Series. You can check out more YouTube
videos on similar topics. Also, if you watch a lot of these videos, and you wanna save some precious time, while you’re studying and in the zone, go over to medimmersion.com
and get an account. In most cases, that will speed you up, by skipping this part of the video, and just jumping right into
the meat of the lecture. Okay, let’s get started. In this lecture, I’m going to be talking about three different studies. The case-control study, the cohort study, and the cross-sectional study. Alright, starting off
with case-control studies, and for this example, I’m going
to use two groups of people. The first group of people, the red group of people, are the cases, and those are the people
that have a disease. In this case, I’m gonna
say it’s skin cancer. The people in the bottom
are the second group, the green people, are those people that
do not have the disease. These are the controls. Because of this naming, you can see why it makes sense to call this a case-control study. So basically, this is a
type of retrospective study, meaning that we’re
looking backwards in time. We already know who has cancer, and who doesn’t have cancer, and we’ve already put them into groups. So we’re looking backwards in time, retrospectively, to see why
one group has a disease, and one group doesn’t. So in this case, we’re comparing the subjects
who have the disease, the cases, with those who do not have
the disease, the controls. And we’re asking ourselves “What happened? “What was different? “Why does the first
group have skin cancer, “and the second group
not have skin cancer?” We can make a hypothesis, like, we can say, well I think that
the group with skin cancer had excessive exposure to
the sun when they were young, and we look backwards in time, and sure enough, we see that, yes, they did have a higher
exposure to the sun. A case-control study is good
for giving us the odds ratio, which is the odds of getting
a disease after exposure, divided by odds of getting that same disease without exposure. So you need to understand
that the top group in red, not necessarily all of them did have excessive sun exposure as a youth, some of them just got
cancer, just happened. Likewise, the people in the green group, some of them may have had excessive sun exposure as a youth. But really what we’re
looking at is the odds. Are the odds increased to get cancer when you have excessive sun exposure, and the answer in this case
is yes, your odds are higher. Okay. Now into the cohort study. A cohort study is a type
of prospective study. Basically we’re looking
into the future now, in which we take two groups of subjects, with unknown outcomes. And of course that makes sense, because the future hasn’t
happened yet, right? We’re looking into the future, we don’t know what’s going to happen. But we’re comparing
these two groups based on what we suppose is going
to be a causal factor. So you can see, there’s two groups here. We know that one of these
groups has sun exposure, and the other group doesn’t. We’re asking ourselves “What
will happen in the future? “What is going to happen to these “two groups of people in 20 years?” A cohort study is good for
measuring the relative risk, which is the ratio of two probabilities. The probability of a disease
occurring in an exposed group, divided by the probability of the disease occurring in a non-exposed group. So how is the relative risk
different than the odds ratio? Well, with the odds ratio, we’re comparing two groups
of people with known disease. For the relative risk, we’re comparing two groups
with known exposure. So having an exposure doesn’t guarantee that you’re going to get the disease. Likewise, having the disease doesn’t guarantee that you have the exposure. So relative risk is kind of saying what’s the risk of getting the disease based on having a lot of
sun when you’re a child, and the odds ratio is what are the odds that all that sun
exposure is going to cause a disease later on or contribute
to a disease later on. Okay. Lastly, is the cross-sectional study. This is a study that analyzes data from a population of people or
whatever you’re studying, at one point in time, right? We’re not looking forward in time, we’re not looking backward in time. We’re looking right now. What’s happening right now? Thus it’s across the
current section of time, or it’s a cross-sectional
data, and it asks the question, not to be too repetitive,
“What is happening?” And cross-sectional studies are good at measuring the disease prevalence, which is the proportion of a population found to have a condition,
at a given point in time. Alright guys, thanks
for watching this video. I really hope you liked it. If you did, give me a like, subscribe, stay in touch, leave me a comment, let me know what you thought, and last of all, good luck in school. Hang in there.


  1. Really helpful, and with good explanations!!
    I´m from Brazil, and really enjoy your channel!
    please, keep posting your videos

  2. Thank you for the video. I like the way you explained it. Really helpful. Can you please do some examples( case scenarios) of each studies .

Leave a Reply

Your email address will not be published. Required fields are marked *