Lectures

Causal Relationship: Possibilities

by Raywat Deonandan, PhD
(1)

Questions about the lecture
My Notes
  • Required.
Save Cancel
    Learning Material 2
    • PDF
      Slides 12 ValidityReliability Epidemiology.pdf
    • PDF
      Download Lecture Overview
    Report mistake
    Transcript

    00:01 Now let's change tracks a bit, let's talk about a logical construct, called necessariness and sufficiency. This is a topic in classical logic, but it is relevant for medical science as well, because when we consider causal factors, we need to know whether or not the causal factor is both necessary and sufficient to cause the outcome we're interested in.

    00:23 So, there are four possibilities here when talking about a causal factor and an outcome. Either the causal factor is necessary and sufficient to cause the disease I'm interested in or it is necessary, but not sufficient or it's sufficient, but not necessary, or it's neither sufficient nor necessary. Let's go through each of these cases with an example. The first, necessary and sufficient. Think about the disease called SARS, which we know was caused by a variation of the coronavirus. So the causal factor here is a coronavirus, that causes SARS. It's necessary to have the coronavirus before you can have SARS, there is no such thing as SARS without infection of the coronavirus. Is it sufficient? Absolutely, it's all you need, you don't need anything else to cause SARS, just the coronavirus. So that virus is both necessary and sufficient to cause SARS. Next, necessary but not sufficient.

    01:18 Alright, consider an environmental trigger that produces a certain disease. An environmental trigger must trigger a gene that becomes activated, so without the environmental trigger, the gene is not activated. Without the gene, there is nothing to be activated and this causes the disease that we're interested in. So each one of those factors is necessary, but individually they are not sufficient, they must be present together to cause the disease. So neither the gene nor the trigger are sufficient, but they're both necessary. Next, sufficient but not necessary. Alright consider radiation and benzene poisoning, each one of those exposures can cause leukemia, alone. They are both sufficient, but neither one of them is necessary, because in absence of one, the other can do the job, so either radiation alone or benzene alone is sufficient to cause leukemia. Next, neither sufficient nor necessary. This one is complicated.

    02:21 Consider prostate cancer. Prostate cancer is caused by a whole bunch of complex interactions and variables. You've got hormones. You've got the things that you eat. You've got inflammation.

    02:30 You've got the kind of person demographically that you are. Individually one of those risk factors is not sufficient to cause prostate cancer, but combined with one or two of the other ones, it might be. Which one or two of the other ones? Well it's a soup, you can pick and choose which variables, any two or three of those could probably cause prostate cancer. It can be argued that this is the scenario that is most likely for most diseases that we perceive, but this is a case where the causal factors are neither sufficient nor necessary to cause the disease in question.

    03:02 Now let's talk about validity and reliability. These are important concepts in scientific proof, reliability is reproducibility. Can the results that you have collected in your study for example, can they be reproduced if you do the study again? That's actually the hallmark of science, reproducibility. That's why we publish our methodologies in papers, so that other people can read our methods, reproduce our studies and hopefully get the same results. Validity on the other hand is, do my results represent the real world? Are they valid when I extrapolate to the rest of the world? So validity and reliability as I mentioned are philosophical cornerstones to science. They are what we consider to be the bases of scientific proof. There are threats to reliability, for example, consider a poor


    About the Lecture

    The lecture Causal Relationship: Possibilities by Raywat Deonandan, PhD is from the course Validity and Reliability.


    Included Quiz Questions

    1. Inconsistency
    2. Biological plausibility
    3. Temporal relationship
    4. Strength of association
    5. Dose-response relationship
    1. Non-statistical association
    2. Temporal relationship
    3. Statistical association
    4. Genetic plausibility
    5. Biological plausibility

    Author of lecture Causal Relationship: Possibilities

     Raywat Deonandan, PhD

    Raywat Deonandan, PhD


    Customer reviews

    (1)
    5,0 of 5 stars
    5 Stars
    5
    4 Stars
    0
    3 Stars
    0
    2 Stars
    0
    1  Star
    0