Skip to main content
Search form
  • 00:00


  • 00:00

    GREG MARTIN: My name is Greg Martin.This is the second in a series of videosthat we're doing on epidemiology and study design and researchmethods.And in this video, we're going to take a quick lookat the difference between case-control studies and cohortstudies.And we're going to have a look at the relative strengthsof each study design.Let's start with case-control studies.Now, to do a case-control study, where you start

  • 00:23

    GREG MARTIN [continued]: is with a group of cases, so people thathave a particular condition.They've got a disease.And it's a great study if it's a rare conditionor a rare disease, right?So you find people that have a disease.Let's say, for example, we found a groupof people who, at the age of 21, suddenlyfind their hair goes green.So we collect those people in a group.We create an analogous group, people

  • 00:44

    GREG MARTIN [continued]: that don't have that condition, but are, in every other way,similar to the first group, right?Then we look back in time.We look retrospectively at their respective historiesand try and identify exposures that may have--and we can look at multiple exposures--that may have led to the condition, in this case,of green hair.

  • 01:05

    GREG MARTIN [continued]: And then we compare these two groups,the people with green hair and the people without green hair.And we try to determine if there isa difference in the extent to which theywere exposed to the group of exposuresthat we're interested in.OK, so, to summarize your case-control study,you've got a group of cases and controls.You're looking back in time.So it's retrospective.You can use rare outcomes, right?

  • 01:27

    GREG MARTIN [continued]: So you're looking at an outcome and then looking backwards.That's important because, when we talk about cohort studies,it's the opposite way around.So you're looking at rare outcomes.You're looking backwards in time.So it's retrospective.You're talking about looking at multiple exposures.Now, importantly, you can do it fastbecause you can do the study right now.You can get data today on what happened in yesteryear.

  • 01:49

    GREG MARTIN [continued]: And it's cheap, right?It doesn't cost much money to do.Interestingly or importantly, the evidencethat you get from a case-control studyis not considered to be strong evidence.OK, now let's talk about cohort studies, and we're going to--I want you to understand the difference between these twostudies.Now, remember, when we talked about a case-control study,we started with outcomes of interest,sometimes, a rare outcome, and we looked back in time

  • 02:09

    GREG MARTIN [continued]: at multiple exposures.A cohort study is exactly the opposite.We start off with an exposure of interest, right?So it might be a rare exposure.There's just a few people in the world thathave been exposed to something that's unusual,people that have been to the North Pole, right?So we find that group of people.We find an analogous group.We find a second group of people who are, in every other way,

  • 02:31

    GREG MARTIN [continued]: similar to the first group except they haven'thad the exposure of interest.Like these are people that haven't been to the NorthPole, but, in terms of their demographyand their age structures and their socioeconomic strata,et cetera, et cetera, you try and make surethat they're pretty much the samebecause you're going to compare these two groups.Now you look forwards in time, and you follow this group up,right?You follow the cohort up over time

  • 02:53

    GREG MARTIN [continued]: to see what outcomes emerge over time as a resultof that exposure, right?So, just to summarize, we're talkingabout a prospective study.You're looking at rare exposures.You can examine them for multiple outcomes.It takes a long time to do because you'vegot to follow this group up over time.It's expensive.

  • 03:13

    GREG MARTIN [continued]: So to run a study like that costs a lot of money,but what you do do--what you do get out of it is good, strong evidence.So it's stronger-- it's consideredto be stronger evidence than a case-control study.OK, I hope that was useful.I hope you found that helpful.In the next video, we're going to lookat randomized control trials and the idea of confounding.

  • 03:33

    GREG MARTIN [continued]: Thanks for watching.I'll be back.

Video Info

Series Name: Gregory Martin

Episode: 11

Publisher: Gregory Martin

Publication Year: 2013

Video Type:Tutorial

Methods: Epidemiology, Case-control studies, Cohort studies, Cohort analysis, Control groups, Data collection

Keywords: case-control studies; clinical trials; cohort analysis; cohort studies; epidemiology

Segment Info

Segment Num.: 1

Persons Discussed:

Events Discussed:



Greg Martin, Editor-in-Chief, Globalization and Health, provides an overview of cohort and case-control studies, including relative strengths of each.

Looks like you do not have access to this content.

An Overview of Cohort and Case-Control Studies

Greg Martin, Editor-in-Chief, Globalization and Health, provides an overview of cohort and case-control studies, including relative strengths of each.

Copy and paste the following HTML into your website