lecture 9: case-control studies

Question 1

what are case-control studies

Answer 1

observational studies that allow the researcher to be a passive observer
group assignments are based on DISEASE STATUS
useful when studying rare diseases of investigating an outbreak
disease state comes first and exposure is what is found
RETROSPECTIVE FASION

Question 2

how are case-control study groups formed

Answer 2

people go out and find “diseased” cases to watch, and find a group of controls that are as exchangable as possible
–from these groups, people are randomly selected to be part of the study (NOT RANDOMIZED INTO GROUPS)

Question 3

pros to case-controls

Answer 3

good for assessing multiple exposures of one outcome
useful with rare diseases
useful in determining associations (NOT causation)
less expensive
useful when disease has long induction/latent period

Question 4

cons to case-controls

Answer 4

cant demonstrate causation
can be impacted by unassessed cofounders
can be impacted by various biases
limited by available data

Question 5

how cases are selected

Answer 5

cases are defined by the investigator using accurate, medically-reliable, efficient data sources
definable criteria are the best (diagnostic criteria)

Question 6

how controls are selected

Answer 6

MOST IMPORTANT
want them to be as equal as possible; expectation is that the controls represent the baseline risk of exposure
way that controls are selected is a major determinant in whether any conclusion is valid
–internal validity
–selection bias
must be selected irrespective of exposure status

Question 7

sources of control groups

Answer 7

population (state/community/neighborhood)
institution/organization/provider
spouse/relatives/friends

Question 8

how controls are selected for case-control studies

Answer 8

outbreak-sources of controls–> participated in same event

Question 9

exposed AND unexposed in same study??

Answer 9

individuals CAN occur as both
associated with…..
–outbreak investigation w/ multiple exposures
–situation of a brief changes in risk of the outcome in interest (hazard period)
»»known as case-crossover design

Question 10

case crossover design

Answer 10

when a situation of brief changes in the risk of the outcome of interest occurs

• -subjects are their own controls during the other times they don’t have the acute changes in risk
• -only design able to adequately attempt to address TEMPORALITY.

Question 11

nested case-control study

Answer 11

case-control studies conducted AFTER, or out of, a prospective previous study-design

• diseased are used in a new different study
• used to evaluate other exposures maybe not incorporated in previous studies

Question 12

selecting for CONTROLS in nested case-control studies

Answer 12

survivor sampling
base sampling
risk-set sampling

Question 13

survivor sampling

Answer 13

sample of non-diseased individuals at the END of the study period

Question 14

base sampling

Answer 14

sample of non-diseased individuals at START of study period

Question 15

risk-set sampling

Answer 15

sample of non-diseased individuals during study period at SAME TIME when case was diagnosed
–case found at certain time//control found at that same time when case was diagnosed.

Question 16

common biases encountered during case-control studies

Answer 16

selection bias

recall bias

Question 17

matching schemes used for selecting cases/controls

Answer 17

individual matching

group matching

Question 18

individual matching

Answer 18

matches individuals based on specific patient-based characteristics–> useful for controlling for confounding

Question 19

group matching

Answer 19

proportion of cases and proportion of controls with identical characteristics are matched

ex: 41% male cases = 41% male controls
* **cases must be selected first