Lecture #11 - Case-Control Studies Flashcards
True or False:
Case‐Control studies are observational, descriptive
studies allowing researcher to be a passive observer
of natural events occurring in individuals with the
disease/condition of interest (Cases) who are
compared with people who do not have the condition
of interest (Controls)
False
Observational, Analytical NOT descriptive
Group assignments in Case-control studies is based on:
DISEASE STATUS
cohort=exposure
case control=disease status
In a 2x2 table for a case control study which table is known and what are the columns and rows?
Disease presence is known (its how you sort the groups)
Column is disease presence
Row is Exposure
What are the main reasons to use a Case Control design?
It’s RETROSPECTIVE
If you are unable to randomize (unethical, illegal, impossible)
Disease of interest is RARE
The strengths of a Case-control desig are:
Good for assessing multiple exposures of one outcome
Useful when diseases are rare
Useful when ethical issues limit Interventional studies
Useful when disease has a long induction/latent period (because it’s retrospective)
Why is the selection of controls difficult in a case-control study?
Goal: to assess for the presence of an association between exposure & known condition of interest by selecting non‐disease individuals from the sample population which produced the Cases
THIS IS HARD B/C its hard to find individuals who aren’t sick but are similar
The way the controls are selected is a major determinant in whether any conclusion is valid. This is commonly referred to as:
Think: are the groups you used applicable to the study
Internal Validity
If you aren’t able to find controls that are exactly like the studied group you should:
expand the range
Control groups can typically come from what 4 sources?
Institutional/Organizational/Provider
• Illness(es) of Controls should be unrelated to exposure(s) being studied
Spouse/Relatives/Friends/Neighbors
• Genetic, Environmental, Socio‐Economic, etc… similarities
Deceased (no‐longer supported/recommended)
• Inherently flawed
Participated in same event (e.g., picnic / convention)
True or False: An individual can actually function as both a Case
AND a Control in the same study
True
There can be multiple exposures (picnic -> who ate the salad, who ate the jello)
True or False: Subjects can be their own controls
True
Called Case-crossover design:
subjects are their own controls during the other times they don’t ahve the acute change in risk. Only used in situations of brief change in risk.
THIS IS THE ONLY CASE CONTROL STUDY DESIGN THE ADDRESSES TEMPORAILITY
Selection bias refers to….
the way Controls are chosen for the study
o Bias is any factor in the design/execution of a study
that leads to a spurious association between variables
NOT SIGNIFICANT during Case-crossover because controls are the cases.
Recall Bias refers to..
the amount/specificity that Cases or Controls recall past events DIFFERENTLY
How well do you remember (“I don’t recall”)
Individual matching of case/controls refers to..
Matches individuals (case/control) based on specific patient‐based characteristics
o Used when each Case has unique & important characteristics
Group matching of case/controls refers to..
Proportion of Cases & proportion of Controls with identical characteristics are matched
• Ex: 41% of Cases are male, so 41% of Controls are male
DON’T match on anything anything that might be a risk factor!!
